SCIENCE JOURNAL
OF
TRANSPORTATION
Especial Issue No. 04
International cooperation Journals
MADI - SWJTU - UTC
Moscow - Chengdu - Hanoi
01 - 2012
Dear researchers, colleagues and readers,
Transportation is the means by which all people are connected, all
human activities occur. Nowadays, in strong globalizing process,
community activities have not been limited by countries’ borders; thus
transportation becomes non-confrontiers.
We, transportation makers, in this moment, have had a common forum
to together discuss, contribute, share and dedicate.
First Especial Issue of international co-operating transportation science
journals of State Technical University (MADI)-Russia, South West
Transportation University (SWJTU)-China and University of Transportation
and Communication (UTC)-Vietnam is published in spring – season of
blooming and developments.
We wish you and our transportation career were achieved, prosperous
and fruitful.
Science is non-limitation,
Transportation is non-border,
Friendship is non-confrontiers,
Aim toward the future, we will do our best to make transportation:
More intelligent and effective,
Faster and safer,
Cleaner and greener,
With that objective, by this forum, we together connect, endeavour,
research, create, contribute, share and devote.
Moscow-Chengdu-Hanoi
Board of Editors-in-Chief
ISSN 2410-9088
SCIENCE JOURNAL
OF
TRANSPORTATION
Especial Issue No. 04
International cooperation Journals
State Technical University-MADI
Southwest Jiaotong University (SWJTU)
University of Transport and Communication (UTC)
Moscow - Chengdu - Hanoi
06-2012
TABLE OF CONTENT
Pages 3
PROF. VIKTOR A. KORCHAGIN
LGTU, Lipetsk, Russia
PROF. VALENTIN V. SILIANOV
MADI, Moscow, Russia
ASSOC. PROF. YULIYA N. RIZAEVA
LGTU, Lipetsk, Russia
Dynamic
adaptive
by advancement of traffics of goods
Pages 11
Pages 52
TRAN TUAN HIEP
TRAN VU TUAN PHAN
Civil Engineering Faculty, UTC
QI WANG
HAILI LIAO
MINGSHUI LI
CUNMING MA
Research Center for Wind Engineering,
Southwest Jiaotong University,
Chengdu 610031, China
PROF. DR. NGUYEN VIET TRUNG
DR. TRAN VIET HUNG
University of Transport
and Communications, Vietnam
Y the oriented application of weathering steel
for bridge in Vietnam
control
Sustainable developing road transport system
for Hanoi
Pages 18
Pages 42
TIAN LI
JIYE ZHANG
WEIHUA ZHANG
Traction Power State Key Laboratory,
Southwest Jiaotong University,
Chengdu 610031, China
An improved algorithm for fluid-structure
interaction of high-speed trains under
crosswind
Pages 61
BOGDANOV N.K.
ZAMYTSKIKHY P.V.
KHADEEV A.S.
State Technical University - MADI,
Moscow, Russia
New approaches to the choice of architecture
for a supervisory system of gaz transportation
Influence of aerodynamic configuration of a
streamline box girder on bridge flutter and
vortex-induced vibration
Pages 65
Pages 29
KRASNYNSKIY M.N.
NIKOLAEV A.B.
OSTROUKH A.V.
State Technical University - MADI,
Moscow, Russia
Improvement quality of congestion controller in
ATM network by method using neural network
Application of virtual simulators for training
students in the field of chemical engineering
and professional improvement of petrochemical
enterprises personnel
environmental indicators for sustanable
development urban transport planning in
Vietnam
Pages 33
QIHENG LU
XIAOYUN FENG
Southwest Jiaotong University,
Chengdu 610031, China
Optimal control strategy for energy saving in
trains under the four-aspect fixed autoblock
system
Pages 72
Pages 78
TRAN XUAN TRUONG
Telecommunication Engineering Department,
UTC
MSC. VU KIM HUNG
InstituteofTransportPlanningandManagement
University of Transport and Communications
BARINOV K. A.
KRASNYANSKIY M. N.
MALAMUT A. J.
OSTROUKH A. V.
State Technical University – MADI,
Moscow, Russia
Algorithm of virtual training complex
designing for personnel retraining on
petrochemical enterprise
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
1
Pages 82
XIAOYU GUAN
YONG ZHAO
XIAOQIU JIA
Southwest Jiaotong University,
Chengdu 610031, China
Ni doping effects YBA2FE3O8 + W
Pages 89
PH.D. KURENKOV P.V
Department Transport business MIIT
GRADUATE POLYAEVA T.I
Department Economics
and Logistics in the transport SamGUPS
Economic competitive
speed lines
Pages 96
PhD. VU THE SON
University of Transport
and Communications
MA. NGUYEN VAN SUU
Transport Engineering Design
Joint Stock Incorporated South
MA. TRAN HUU BANG
Sai Gon Construction Quality Control
Joint Stock Company
XUEYI LIU
PINGRUI ZHAO
FENG DAI
Southwest Jiaotong University,
Chengdu 610031, China
Advances in design theories of high-speed
railway ballastless tracks
Pages 117
MAI VINH DU
DUOQIAN MIAO
RUIZHI WANG
Department of Computer Science
and Technology Tongji University,
Shanghai 201804, China
LE HUNG LAN
Electrical & Electronic Department,
The University of Transport
and Communications, Hanoi, Vietnam
An efficient method for Vietnam license plate
location
2
Pages 134
evaluation of high -
Calculation process of settlement - transition
pile net for bridge approach embankments
Pages 104
YINGXUE WANG
BO GAO
CHAO ZHANG
XUZHOU HE
School of Civil engineering,
Southwest Jiaotong University,
Chengdu 610031, China
Analysis on setting airshaft at mid-tunnel to
reduce transient pressure variation
Pages 127
TRAN THE TRUYEN
Department of Civil Engineering
University of Transport
and Communications
Service life estimation of reinforced concrete
structures with considering the damage of
concrete cover
BERNER, L.I.
Pages 142
KOVALEV A.A.
NIKOLAEV A.B.
State Technical University – MADI,
Moscow, Russia
The block of modeling and forecasting of gas
networks systems operation modes for
dispatcher decision support system
Pages 146
GALINA BUBNOVA
Moscow State University of
Railway Transport
Russia
Industrial integration and economical crisis
DR. LE VAN BACH
University of Transpor
and Communications
ENG. TRAN HUU BANG
Saigon Construction Quality Control
Jont Stock Company
Initial result of using cinder particles at some
steel factories in Ba Ria - Vung Tau province as
a mineral additive for cement concrete in
building motoray sufrace
Pages 156
Pages 163
DR. ALEXANDER CHUBUKOV
PROF. VALENTIN SILYANOV
State Technical University - MADI
Mechanisms for integration federal and
regional strategy for ensuring road traffic safety
in Russia
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
DYNAMIC ADAPTIVE CONTROL
BY ADVANCEMENT OF TRAFFICS OF GOODS
PROF. VIKTOR A. KORCHAGIN
LGTU, Lipetsk, Russia
PROF. VALENTIN V. SILIANOV
MADI, Moscow, Russia
ASSOC. PROF. YULIYA N. RIZAEVA
LGTU, Lipetsk, Russia
Summary: On all stages of functioning of transport systems approach of the systems is
used to the choice of rational management exploitation of motor transport. Intellectual control
system is offered transport streams in cities on the basis of traffic-light objects for reduction of
expenses of resources of by participants process of delivery of loads.
Key words: A transport system, transport stream, city.
CT 2
Transport possibilities of Russia, along with natural resources and geographical location, the competitive edge of our country. An increase of their attractiveness is a task state, as often
nascent transport corks and delays of people and loads on motorways with plenty of the lost and
trauma people, height of consumption of power resources and negative influence on an
environment in world practice it is accepted to characterize as strategic problems of national level.
Presently for many key objects of transport systems of our country uncoordinated and
coordinated not enough co-operation is very characteristic in-process motor transport. It,
foremost, contingently inharmonious development of associate infrastructural links of transport,
стыкующихся in knots, absence for them of single informative and normatively-legal base,
management center. As a result in transport knots and on going failures, congestions in
advancement of transport streams, scale outages of rolling stock, appear near them, the terms of
delivery of loads increase, both in internal and in international reports.
Deciding these tasks is possible on the basis of application of modern methods of
organization of motion of transport vehicles and pedestrians, by development and introduction
of intellectual control system (ICS) by transport streams of city. Without application of
computer facilities to decide this problem practically impossible.
Unsolved tasks in organizations of travelling motion, considerable increase of intensity of a
transport stream create serious problems in organization of transportations, worsen efficiency of
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
3
transport processes. One of variants of increase of efficiency of a transport process is a dynamic
management by travelling motion. A dynamic management has a row of advantages motion as
compared to a local management, because this system has the opportunity to predict nonstandard situations, forecast the changes of basic parameters of a transport stream, co-ordinate
work of technical equipments of adjusting, and also to coordinate managing parameters on a
network. It is important in transport knots to provide not only co-ordination in-process TCS but
also co-operating with them dispatch, insurance companies, broker-custom organs, freight
terminals, storages of temporal storage, operators of rolling stock and other partners.
From the out-of-control processes of motorization exceeding of admission possibility of
street-travelling network on 20-30 reduces the actual carrying capacity of crossing in 2-3 times, up
to the complete stop of a transport stream. For example, a middle rate of movement of transport on
the highways of Russia is 40-60 km/h and 80-100 km/h abroad. Consequently, in Russia loads are
moved for twenty-four hours to 250-300 kilometers against a 500-700 kilometer abroad. The
decline of rate of movement, in turn, conduces to the increase on a 20-30 prime price of
transportations, to the height of transport constituent in the last bid of products and services, that
reaches in Russia to the 15-20 (in USA and Europe this index does not exceed 7-10%).
Basic backlogs of perfection of a transport process are in rational organization on the basis
of engineering logistic of co-operation of all participants of transport-distributive chain of
motion of load, in the concordance of their interests and search of взаимоприемлемых and
mutually beneficial decisions. The logistic centers, created at the level of territories, municipal
educations, metropolises, integrated branch complexes, can promote the decision of task of
perfection of processes.
It appears that a basic task of perfection of work of motor-car transport at moving of loads is
reduction of general expenses in the process of satisfaction of transport necessities at maintenance of
the accepted (or concerted) parameters of quality of service and quality of the state of environment.
In a region, under act of dynamically changing terms of work of transport vehicles, both in
time and in space, there are processes of failure to observe of initial plan of a transport service,
and insolvency of administrative influences shows up. All of it results in situations, when the
existent models of vehicular process can not be effectively used for the operative management
of decision of tasks for acceptable one time.
The ramified of street road net gives an wide opportunities for varying transport operations
by optimization of routes of motion of transport vehicles between the points of service, local
correction of these routes, by the redistribution of transport vehicles on the routes of delivery,
minimization of expenses on providing of operative technical help or replacement of defective
transport vehicles.
The really folded situation stipulated a management necessity transport streams on crossing
of motorways with by the use of the traffic-light adjusting. The mathematical model of
intellectual control system (ИСУ) is offered by transport streams of city on the basis of traffic4
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
CT 2
light objects for reduction of expenses of resources of participants of process of delivery of
loads. From position of approach of the systems a management process is examined by freight
motor-car transportations under act of dynamically changing travelling, surrounding and
external environments and terms of service.
In a market competition economy attention increases to the every family losses. The
combined cost and time of delivery of load determine to a great extent, whether it will be maybe
that or another economic co-operation. Id est a main value is presented by an integral function is
delivery from a door to the door. But a function requires creation I correspond cabbage soup of
structure that would provide her implementation.
Already clear, that the necessity of creation came to a head for the cities of ICS, that must
provide the arranged technical co-operation of all participants of transporting and preparation of
loads. Thus the functions of the system it is been: complete and timely reception of requests on
transportation; calculation of the required park of rolling stock; serve of requests on providing
of loading of empty cars; controller's support of transportation; conduct of the permanent
monitoring of traffics of goods and determination of prospects of origin and redemption of
traffics of goods; operative co-operating with the contiguous types of transport and a load is
formed by enterprises.
Technology of work of regional ICS must be base on single through technological process
of work of street-travelling network of city, envisaging the concerted admission of traffics of
goods and rolling stock (carriages, cars) providing their on carriage.
CT 2
Operative management it is impossible a vehicular process to examine in tearing away
from a management by exploitation of roads and organization of travelling motion. The single
technological process of work of street-travelling network of city must be based on satisfaction
of economic interests of participants of vehicular process. The basic obligations of parties
participating in transportation must be certain in him, sufficient in an order to serve as the
regulator of their relations, and also to bear responsibility for their non-fulfillment.
Management in cities it maybe transport streams of street-travelling network to carry out
due to the change of the modes of operations of traffic-light objects on the managed crossing,
guided sign-boards. One of above all parameters of a transport stream on the basis of that made
decision about the redistribution of green time between the groups of motion there is a delay of
transport vehicles.
One of tasks of ICS is minimization of congestions on crossing of street-travelling network
and increase of speed of delivery of loads. Existent model of control system on the basis of trafficlight objects can not transport streams help management centers to decide a main task travelling
motion, and the considered variants of CAS of management by travelling motion by virtue of
locality of application and by the difficulties, conditioned by absence of reliable methods of
prognostication of distribution of transport streams at the different variants of project decisions a
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
5
System of logistic management freight motion on the street-travelling network of city is by
the integrated system. The necessary volume of informative arrays part of that is local with the
small period of storage is concentrated in her. Mathematical raising of task of optimization of
advancement of transport streams on the basis of ICS formulated:
Р1 (A, R , K, O, V) → opt
2
Р (A, R , K, O, V) → max
(1)
(2)
Where Р1- system of indexes of optimization of a transport activity(time, cost, normative
troop landing in the environment of and other);
Р2 - a profit of logistic center of management for optimization of transport activity;
A - parameters of motor transport;
R - parameters of shifting complex transport of knot;
K - limitations of objects of infrastructure;
V - technological possibilities and limitations;
O - influence surrounding and external environments on transport processes.
A planning task is a multicriterion, depending on the large number of factors, and for every
street-travelling network of region must decide separately coming from local terms and specific
of them works. Thus planning must be not discrete, and continuous with providing of permanent
depth of prognosis (for a 10-15 twenty-four hours forward and more), id est planning of traffics
of goods from a moment and places of their origin and formations.
As indexes must be worked out and accepted interdepartmental universal, balanced indexes,
characterizing complex efficiency of cooperation (synchronizations) of proprietors of load,
dispatch, and other participants of process of товародвижения in providing of transporting and
processing of loads in a knot and on the way their following from the places of origin.
As main indexes of choice most economically of effective variant of freight motion it is
recommended to use time, cost of delivery and normative troop landing in surrounding environment.
∑ t → min
∑ S → min
i
(3)
i
(4)
i
i
Where ti - time of transportation, clock;
Si - freightage, rubles;
Yi - normative extras in OS, conditional tons/year.
Three criteria create many criteria and needed methods of her overcoming. It is here
possible to recommend two approaches.
First. Linear displacing. Three criteria are taken to one by means of coefficients of coercion.
6
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
CT 2
Second. Hard priority. If one of criteria far excels by value the second, it is possible to
examine them in turn. For example, the term of delivery costs on the first place. Then at first
variants get out with minimum time of delivery, and among those, that is satisfied to the
requirement of time, gets out with the least cost.
At a management travelling motion the structure of management object is known for and
unchanging, and behavior of him depends on the row of unknown parameters. This task decides
in the class of the self-tuning systems in that the structure of regulator is set and it is required to
define the algorithm of tuning (algorithm of adaptation).
The task of synthesis of adaptive control system is considered. Let on the object of
management (ОM) measureable indignations influence (questioner influences) of х(t), not
measureable indignations of Q(t), surrounding and external environments of Fвi and Fпi and
managing influences of U(t), influence on the object of management, fig 1.
External
environment
FВi
Surrounding
environment
FПi
Filter
evaluations
Model
measuring
devic
Adjusting contour
CT 2
measuring
device
х(t)
Transport streams,
ОM
Ноб
Model
of object
PS
U(t)
Traffic-lights
Control system
d
Managing
system
система
Algorithm
adaptations
Contour adaptations
Regulator
Fig 1. Flow diagram of adaptive control system by transport streams
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
7
A communication network includes a tie line, on that entrance information and reverse
communication channel on that to the sensor-based system state information is passed
management object are passed. Information about the guided object, external and surrounding
environments perceived by the sensor-based system, processed in accordance with one or
another aim of management and as managing influences passed through traffic-lights on the
object of management (transport streams). A device for realization of purposeful influences is
the sensor-based system, a management object and communication network form the system
with a management (SM).
The supervisions a weekend is accessible to variables of object of management of Y(t). We
suppose thus, that behavior of object depends on the row of unknown parameters totality of that
is designated through D(t). In the process of management it is necessary to obtain freight motion
of a transport stream travelling motion in the set mode, that will give: possibility to shorten time
of passage on a network; to decrease the brought mass over of extrass of harmful substances in
an atmosphere; to shorten transport delays and charges on moving of loads (passengers) and
provide sufficient strength of travelling motion security. Because adaptive control system differ
from the traditional (not adaptive) systems the presence of contour of adaptation, then for
formalization of task of synthesis of algorithm of adaptation we use a concept "The Influenced
object" (Ноб), that plugs in itself all unchanging part of the control system (СS).
If primary sensors (PS) do not provide measuring of necessary for a management
parameters, such as speed, intensity, closeness of a transport stream, apply the filter of Кolmana,
dynamic prognostication or methods of analysis of temporal rows.
The algorithm of adaptive control has a three-level structure. The algorithm of the first
level (adjusting algorithm or algorithm of base-level) depends on the vector of parameters of D
(parameters of regulator), at each d ∈ D he must provide gaining end of management.
An algorithm the second level changes(influences) the vector of D such by character, to
provide gaining end of management at unknown d ∈ D .
On the basis of foregoing by us the three-level algorithm of adaptive control is offered by
transport streams on the street-travelling network of cities. The algorithm of adjusting of a
transport stream is modification of the coordinated management with introduction of new
managing parameter - balance of time of resolvent and forbidding phase of the traffic-light
adjusting ( 0 ÷ 1 ) and his rationed value. Management aim - the achievement of the set mode of
motion of a transport stream (smoothing of rate of movement) on the highways of city is arrived
at by the concordance of parameters of the traffic-light adjusting on a network.
The system described higher is a management task in the conditions of the vagueness
related to d ∈ D , fig 1. We will consider the parameters of adaptive control system a transport
stream of city. Entrance and output variables and managing influence are described as follows:
8
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
CT 2
U ( t ) = F3 ( Ц с , , V, B)
(5)
x ( t ) = F1 (П р , l, N cp , W)
(6)
u ( t ) = F2 (И, V, K c , Ц с , Y )
(7)
Where Пр- carrying capacity of street-travelling network, cars/hour;
l - length of driving, m;
Nср - presence of technical equipments of adjusting, 1 or 0;
W - influence external and surrounding environments;
И - intensity of a transport stream, cars/hour;
V - speed of a transport stream, kilometer/an o'clock;
Кс - composition of a transport stream;
Цс - a cycle of the traffic-light adjusting, (seconds);
Y - ecological economic damage;
B - normative troop landing in an environment;
F1, F2, F3 - unknown functions.
CT 2
In general case a management aim is set as having a special purpose inequality
Т в ≤ Т л at Т л > 0
(8)
Where Тв - time of passage on a street-travelling network, hour;
Тл - time of passage on a network at a local management.
As a regulator of adaptive control system a transport stream such technical equipments of
adjusting as traffic-light devices and guided sign-boards are used. The vector of unknown
parameters of d consists of traffic-light cycle indexes that is included in mathematical description
of management object and correspond to the optimal tuning of the traffic-light adjusting:
d = (d1 , d 2 , d 3 , d 4 , d 5 )
(9)
Where d1 - duration of basic phase of adjusting (seconds) cycle;
d2 - duration of prohibitive phase of adjusting (seconds) cycle;
d3 - duration of intermediate times of adjusting (seconds) cycle;
d4 - balance of time of resolvent and forbidding phase of traffic-light cycle;
d5 - a change of phases of the traffic-light adjusting of "relatively command" traffic-light.
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
9
The algorithm of adaptive control is offered by transport streams on the street-travelling
network of cities with the use of the traffic-light adjusting, taking into account an ecological
factor. The algorithm of adjusting of a transport stream is modification of the coordinated
management with introduction of new managing parameters - balance of time of resolvent and
forbidding phase of the traffic-light adjusting (0÷1) and his rationed value and account of extras
of harmful substances in an environment. Management aim - the achievement of the set mode of
motion a transport stream smoothing of rate of movement) on the highways of city and
normative troop landing in an environment is arrived at by the concordance of parameters of the
traffic-light adjusting on a network.
Because of territorial disconnect, and also distinction of parameters of functioning, for
effective work of freight transport there must be a single center of management, in that in good
time and operatively must act information about performance of every subject of a transport
stream indicators. Adjustment of strategy of development of a transport stream, and also, in a
short-term prospect, change of managing influences and redistribution of resources, must be
executed taking into account the results of analysis of acting operative information. At the same
time, as co-operation between subsystems comes true by means of informative and material
streams, her it is also possible to examine as intellectual control system, id est adaptive system
ticker-coil, where, where informative streams act part managers and correct of description
material streams.
Guidance of enterprises tests a requirement in reliable information, as on it management
quality depends by an enterprise and efficiency of planning of his activity in the conditions of
hard competitive activity. For informative support of every stage of management application of
ICS providing the choice of effective administrative decisions and operation ability of reaction
on the market state of affairs and changes external and surrounding environments is needed.
Approach of the systems to the choice of rational management must be used on all stages. On
every stage the characteristic are used for the decision of similar tasks methods and models.
The offered algorithms of management give an opportunity to shorten time of passage on a
network freight motion, to decrease the extras of harmful substances in an atmosphere, transport
delays and charges on moving of loads and passengers.
References
[1]. Korchagin V.A. Innovative ecological econom.-Lipetsk / LGTU, 2010. – 200p.
[2]. Korchagin V.A. Model of search of the effective functioning of motor transport natural economic
frame//of society the World transport and technological machines. - 2011. - № 2. - With. 25-31♦
10
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
CT 2
SUSTAINABLE DEVELOPING ROAD TRANSPORT SYSTEM
FOR HANOI
TRAN TUAN HIEP
TRAN VU TUAN PHAN
Civil Engineering Faculty, UTC
Summary: This presents characteristic Road transport System of Hanoi, and propose the
objective, principles and solutions for sustainable developing.
Key word: Hanoi, Road transport, sustainable developing.
I. OVERVIEW
In general, most of the urban in the world has been self-developed, derived from the factors
that support activities, manufacturing and transportation. Those are elements of geography, soil,
location and population, etc. For example:
- The confluences (supporting water way exchanges).
- The rivers, mountains and forests (supporting raw materials, fuel and trading).
- Delta area (supporting cultivation, transportation, fertilizing soil, construction, expansion, etc.).
- The location adjacent with the territories...
Such a place with such favorable natural conditions has been chosen to be where to form
city’s center. Over time, with the development of the economy-society, commercial exchange
develops more while pushing the development of transport system forming as centripetal
(Radial axle roads and urban ring roads).
Most of such urban has the central area with deep antique impression; urban density is very
high, crowded in the hub, stretching from inside to outside. Urban develops by the expanding
and invading type.
There is also young urban (with its road network) which develops in the form of fish bones,
chessboard, chain or mixture.
II. DEVELOPING ROAD TRANSPORT SYSTEM - OBJECTIVE & PRINCIPLES
Objectives
Road Transport System is a special system which consists of both entities structures and
regimes to operate it to serve the transport through, safely, smoothly, comfortably, economically
and aesthetically.
“Through” means: transport must be normally continuous during space: over service areas;
during time: over seasons without influence of climate or natural condition. To meet the people
demand of door-to-door and just in time
“Safely, smoothly, comfortably” show the service quality.
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
11
“Economically” expresses the effectiveness of all system which achieved by infrastructure,
transport means, legislation, and policies to manage and operate the system.
“Aesthetically” in accompany with safely, smoothly, comfortably response the
requirement, being friendly with environment and people.
Principles
•
Urban transport master plan has to be based on and originally generated from general
planning for economy-society city development, urban space development and land use.
•
later.
Planning has to be based on the long term strategic vision, at least for 50-100 years
•
Aiming all the effort to develop economy-society, for human, for the development of
human, for the harmony between people and people, people and nature, society; this means for a
healthy economy-socio-cultural and highly developed human environment.
•
Developing road transport based on smart growth, reasonable balance all modals of
transport, priority for ITS and environment friendly technology.
•
Transport system along with urban infrastructure has to go a step ahead in order to lay
a foundation and motivation for urban sustainable development.
•
The heritage, traditional value must be inherited, reserved and promoted, we should
synchronously acquire and study the advance achievements and experiment of world’s urban in
the globalizing cooperation, affiliation and development.
•
The urban of developed countries has to base on the characteristics of the region;
inherit, preserve, promote and study to find out their solutions in order to develop quickly and
sustainable with the criterion: “Tradition, Nation and Modern”.
III. TREND TO DEVELOP ROAD TRANSPORT SYSTEM
Born earliest and expanded largest, Road Transport has been being increasingly developed
all over the world.
In developed countries, now they tend to find out advanced breakthrough technology for
better Road Transport in future:
- To create “greener” road transport. The European Road Transport Advisory Council
(ETRAC) has proposed for a directive on promotion of clean and energy efficient vehicles
(2007)
- To encourage modal shift and decongesting transport corridors.
- To ensure sustainable urban mobility
- To improve safety and security
- To strengthen competitiveness
These trend lead to the concept of developing and integrated transport system, Road
transport and health, request to develop the ITS.
ULTra (urban light transport) in European seems a breakthrough technology of road
12
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
transport. This was applied at Heathrow airport as pilot project and shown many benefits:
- Immediate service: passengers rarely need to wait for a vehicle. Since the empty vehicle
management system ensures that one will already be at the station. Simulations demonstrate that
average waiting times, even in peak periods, average around 10 seconds.
- Nonstop travel: due to off-line stations, the journey is nonstop from start to destination,
anywhere on the network.
- No need to plan trips, considers schedules, or transfers between vehicles.
- Your own private cab you only share with your friends and family.
- Faster than other urban transport, typically by a factor of two or three. Although
maximum speeds are modest (25 mph), non-stop service ensures short trip times.
- Travel is reliable, predictable and congestion free affording passengers greater certainty
in their journeys.
- Travel is safe: ULTra’s target is safety levels at least as good as trains, approximately 10
times higher than automotive safety. Also segregation implies less conflict with non-users.
- Accessibility: The system is available to all, including the young, the old, and those with
disability.
In addition to user benefits, ULTra provides sustainable urban transport with major benefits
to non-users and to the community as a whole.
- ULTra is energy efficient: Light, small, efficient vehicles traveling non-stop and only on
demand result in significant energy savings. ULTra saves 2/3 rds of automotive energy
requirements, and is substantially more energy efficient than conventional public transport.
- ULTra meets Kyoto sustainability targets; providing the required 60% reduction in carbon
emissions over the car now, rather than by the 2050 target date of the Kyoto agreement – 35
years ahead.
- ULTra is exceptionally quiet: measurements on the prototype vehicle running at 6m/s
give 35 dBA at 10m, around 30 dB less than cars.
- Light weight vehicles permit ultra-light infrastructure: Automated control allows high
utilization. Small vehicles and guide ways imply less land take and less visual instruction.
- ULTra reduces congestion: Studies indicate significant modal shifts away from the car,
freeing up both road capacity and parking space.
- Installation flexibility: Small scale infrastructure may be readily integrated into buildings.
ULTra provides new ways to reclaim areas of the city now given over to the automobile.
- Rapid installation minimizes cost and disruption.
IV. SOLUTIONS FOR HANOI URBAN ROAD TRANSPORT SYSTEM TOWARD THE
SUSTAINABLE URBAN DEVELOPMENT
4.1. Characteristics of Hanoi urban road network
Hanoi has the center lays on the confluences of Duong River --- Hong River on the Delta
which has flat, large and rich land. With the spontaneous development of urban for more than
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
13
1000 years until now, this place forms a crowded urban with the population of no more than
10km2 (road transport system is primary, waterway is less developed, railway and airway do not
serve external transport relationship).
•
Road network
- Radial axle and ring road system merely developed (have just formed 3-4 lanes in the
inner city but hasn’t constructed and rehabilitated completely, hasn’t gotten suburb’s ring
roads). Therefore, external transports, national transport from different directions have to pass
through the center of city making it pressed, overloaded and congestion for the inner city.
- The streets mostly are two ways, the carriage way is narrow.
- The sidewalk is narrow, patchy and out of standardization.
- The inner city intersection system is at grade, much and closed to each other (200-400m).
- Residents’ houses system inserts heavily along the inner city’s street, is both living and
settling place, manufacturing, trading, etc. since long time ago; thus, it is very hard to clear
away, compensate and resettle in order to widen and improve the road (also build and improve
inner city’s public works).
- The urban system and the suburb’s road network are merely developed.
- The system of Red river bridges is not enough capacity. There are only 3 bridges: Thang
Long, Chuong Duong and Long Bien old bridge.
- The urban transport system is simple, lack of synchronous development of railway and
waterway.
•
The parking system
- The parking system is poor. The shops system crowded along the streets, small markets
and schools, etc. which create the problem that the traffic vehicles (cars, motorcycles…) occupy
pavement and sidewalk for parking so that the narrow streets become narrower.
•
The transport means
- The transport means are mainly cars, motorcycles and bicycles. Public transportation is
mainly bus, is given especial encouragement in 2000-2005, however just serve about 20% of the
travel demand but cannot develop because the road system is not satisfied yet.
•
The motorcycle boom and mix traffic flow
- There may be no such an urban in the world that has a many motorcycles as in Hanoi;
motorcycles serve more than 60% the travel demand of urban residents.
Until November, 2005, Hanoi has nearly 163,000 cars (with the increasing rate of 12-14%
per year), more than 1,5 million motorbikes (with the increasing rate of 14-16%) and more than
1 million bicycles (has the trend to a slight decrease).
Bicycles, motorcycles and cars put the mix traffic flow on the road is a outstanding
characteristic of Hanoi urban transport system, is currently a hard problem to the managers and
controller of urban transport.
14
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
•
The transport participants
- Almost Hanoi transport’s participants are not only the passengers but also the transport’s
controller (which means also the drivers). Coming from the habit of agricultural economy, small
manufacturing; the knowledge, the awareness for abiding by traffic regulations is not quite good.
There are even people who just care about when can go but need not to know when can be go.
•
The institution system, urban transport control and management:
Not really explicit and completed.
Fig. 8.4. Hanoi Road network
Fig. 8.5. Occupy carriageway, sidewalk for parking
Fig. 8.6. Mix traffic, and traffic jam
4.2. Solutions for Hanoi
•
Hanoi chooses its manner to develop from outside to inside
- Basic argument
As above mentioned, like many cities which has a long spontaneously developing process
has created Hanoi, crowded and overloaded central area. The difficulty of the clearance work
and relocation for improving and rehabilitating the interior area, the conflict between
reservation and development is the biggest impediment of the sustainable urban development.
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
15
Hanoi does not develop evenly; the suburb is currently a poor rural condition. It is a
favorable condition for constructing urban infrastructure synchronously and modernly.
The pressure of the internal area creates the inhabitant’s requirement for expanding to the
borders, which increases the trend of purchasing lands to build houses in the suburban area. This
creates sprawl construction in the suburb which is over controlled.
Many big cities in over the world have developed from inside to outside. Hanoi developed
slowly, and is still on its long way after other cities and urban, if goes on the same way, Hanoi
can never catch them. The problem here is to find its own way, which is to develop urban from
exterior into interior area.
- Concrete solutions:
First, to limit temporarily the development of the internal area.
Concentrating the sources on suburb’s development, building urban infrastructure to be
modern and synchronize. Specifically develop the radial road axle system and outer ring roads
with 15; 20; 25; 30 km far from the center as express way; next are to develop branch roads and
local roads into standard as well. Based on that, establishing the urban functional area can be
established to be modern as its planning.
Pulling and obligating some organs, universities, schools and administration offices from
internal area to exterior area.
Then reserving, improving and regenerating the interior area into a political, culture,
service, commercial, tourism, antique, modern and peaceful area.
- Hanoi axle road network proposal.
Fig. 8.7. Hanoi axle network proposal
16
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
• Hanoi certainly has to develop public transport system including buses and urban
railway system (metro). Within 15 year coming, Hanoi must to build at least 20-30 km of metro
•
Hanoi certainly has to eliminate motorcycles
Hanoi’s actual development has raised the problem of motorcycle explosion, this traffic
type develops so fast that makes the urban management over controlled, is also a reason for
traffic jams and traffic accidents.
Based on that fact, there have been many measures like increasing consuming tax,
prohibiting people from registering motorcycles in the inner districts, improving and investing
equipment; executing administrative and educational solutions, etc.; however, the numbers of
motorcycles running in the city have still been increased, the solutions are not effective yet.
- It is the time to answer definitively; (1) Can Hanoi eliminate motorcycles?, (2) When
will Hanoi eliminate motorcycles?
- Answer for question number (1): Show the decisive determination of the top leader to
attract and focus the sympathy power of the society and the people.
- Answer for question number (2): Determine the exact time for researchers, urban
managers studying to sketch out a reasonable process and comprehensive, synchronous and
feasible solutions for eliminate motorbike. Pay a special attention for the developing public
transport and dominate steps by steps private vehicles; so that urban residents have time to be
mentally prepared and choose the suitable solution for them and their own family.
We think that: in 2020, Hanoi has to say “No” to motorcycles. 15 years are enough to step
by step to build and develop public transport, step by step limit and then eliminate motorcycles,
simultaneously execute synchronous solutions for the economy and the society. If the progress
is too fast, it will be unfeasible; if it is too slow, motorcycles will develop until the level it
cannot be eliminated, causes bad affection to the urban development unable to be solved. Thus,
the urban planners and managers will be unsuccessful.
•
To invest and construct parking system to be enough and reasonable for the city.
•
Developing a special transport system, “smooth transport” by using traffic means
unpolluted and by establishing walking streets in the commercial and tourist area, old streets,
traditional and cultural handicraft villages.
•
To strengthen road transport regulations system and related policy in order to make
more effective urban transport management as well as providing, implementing administrative
and educational solutions to improve knowledge and awareness to obeying traffic regulations of
urban residents.
Besides road transport and urban railway, Hanoi also has a system of diversified rivers and
lakes system; thus, waterway transport serving tourism, as air transport for external relations
and international integrated transport are significant factors in developing our capital to be
modern and sustainable.
References
[1]. Martin V. Lowson, APRT Solusion for Europea Cities, Infrastructure Feb. 2007
[2]. ULTra Website, Advanced Transport System Ltd♦
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
17
INFLUENCE OF AERODYNAMIC CONFIGURATION OF A
STREAMLINE BOX GIRDER ON BRIDGE FLUTTER AND
VORTEX-INDUCED VIBRATION
QI WANG
HAILI LIAO
MINGSHUI LI
CUNMING MA
Research Center for Wind Engineering,
Southwest Jiaotong University,
Chengdu 610031, China
Summary: Streamline box girders are widely applied in the design and construction of
long-span bridges all over the world. In order to study the influence of modifications of
aerodynamic configuration and accessory components on flutter and vortex-induced vibration
(VIV), more than 60 cases were tested through a 1 ‫׃‬50 scale section model. The test results
indicates that the aerodynamic configuration and accessory components of streamline box
girders can significantly affect the wind-induced vibration of bridge, which is in good
agreement with the experience of past researchers. From the tests carried out, it is observed
that if the horizontal angle of the inclined web of the streamline box girder is below 16°, the
critical flutter wind speed of bridge will increase remarkably, and the VIV will diminish. The
test results also show that the 15° inclined web can restrain the formation of vortex near the
tail, and consequently improve the performance of aerodynamic stability of long-span bridges.
Finally, a new streamline box girder with 15° inclined web was presented and strongly
recommended in the aerodynamic configuration design of long-span bridges.
Key words: streamline box girder; aerodynamic configuration; wind tunnel test; flutter;
vortex-induced vibration.
I. INTRODUCTION
Since its successful application to the Severn Bridge of Great Britain, the streamline box
girder has been widely used in the design and construction of long-span bridges all over the
world. Its aerodynamic configuration and accessory components have great impacts on bridge
wind-induced vibration, especially the flutter and vortex-induced vibration (VIV). Larsen [1]
conducted a series of wind tunnel tests on the girder of the Great Belt Bridge, and discussed the
influence of railing, guide wing, and rostrum on the flutter. He found that the railings of low
porosity weakened the aerodynamic stability, the blunt rostra decreased the flutter wind speed,
and the guide wing strengthened the aerodynamic stability. Similar results were obtained by
Miyata [2] and Bruno [3]. Wilde et al. [4] designed an active deck flaps control system to
strengthen the aerodynamic stability of long-span bridges, but this system could increase the
construction and maintenance cost of bridges, and its feasibility and efficiency in real-world
applications still need to be validated. Yang and Ge [5] studied the effects of central stabilizer
on aerodynamic control of long-span bridges. The results indicated that the stabilizer did good
to aerodynamic stability, and its fixed position must be confirmed by wind tunnel tests.
18
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
However, the stabilizer did not fit for a traditional box girder. Through the wind tunnel tests and
experiments [1]-[5], it was found that the flutter of long-span bridges could be avoided by
optimizing the aerodynamic configuration and deck details. Compared with the studies on
flutter of the streamline box girders of long-span bridges, however, the studies on VIV are not
often seen. Larsen et al. [6] discussed the guide vane of box girder for suppression of the VIV of
the Great Belt Bridge. He provided a general design of the guide vane for the twin box girder of
Stonecutter Bridge in Hongkong [7], but this design would raise the construction cost. Diana et
al. [8] investigated the vortex-shedding phenomena of the multi-box deck shape of Messina
Strait Bridge and discovered the remarkable VIV phenomena in the bridge, but he did not
provide the effective measures to suppress the VIV of the triplet girder. Ge et al. [9] carried out
wind tunnel tests and field tests for the VIV of Xihoumen Bridge with a twin box girder, and
found that the windbreak could control VIV effectively but would increase the drag force of
girder and weaken the aerodynamic stability of the bridge. The aforementioned work indicates
that there are lack of effective and practical measures to suppress VIV and maintain the
aerodynamic stability of long-span bridges.
In order to clarify the influence of accessory components and aerodynamic configuration
on the aerodynamic stability and VIV of long-span bridges, this paper deals with wind tunnel
tests through a 1 ‫׃‬50 scale section model of streamline box girder. More than 60 cases of deck
configuration were tested, such as the bridge railings with different porosities in the sideway,
the position of the rail of bridge inspection car, the obliquity and width of guide wing, the edge
configuration of the section (rostrum), and the inclination of the inclined web. Based on the
tests, a new design for aerodynamic configuration of bridge girder is put forward, which could
ensure the superior aerodynamic stability and less VIV of long-span bridges.
II. WIND TUNNEL TEST MODEL
The design girder cross-section of the Nanjing 4th bridge, which crosses the Yangtze River
and is located in Jiangsu Province of China, with the main span of 1 418 m, was selected to
carry out wind tunnel tests. The bridge deck is originally designed a trapezoidal steel box girder
with overall width of 37.7 m and a height of 3.4 m, and the inclination of inclined web is 22°, as
shown in fig 1. For long-span bridges, the aerodynamic stability is a governing factor in the
design. According to the wind statistic data and the Wind-Resistant Design Specification for
Highway Bridges [10], the flutter checking wind speed of the bridge is calculated as up to
60.8 m/s. However, the critical flutter wind speed of the original girder is found by intensive
wind tunnel testing of the section model only 45 m/s. Apparently it falls short of the required
flutter checking wind speed. Hence, in the aerodynamic design of this bridge, it is necessary to
adopt some countermeasures to ensure the critical flutter wind speed meets the requirements of
wind resistance design. And on the other hand, we attempt to find the necessary measures to
suppress the VIV through the tests.
Fig 1. Outline of the streamline box girder in the tests
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
19
The wind tunnel tests were conducted in the second test section of XNJD-1 wind tunnel. The
model is 2.1 m long, and its width and height vary with the change of aerodynamic shape. The
mass of the model is 23.52 kg, and the mass moment of inertia is 1.407 kg·m2. The vertical and
torsional frequencies of the test system are 1.76 Hz and 4.03 Hz, respectively, and the wind speed
ratio between model and real bridge is 3.28. The damping ratios of the test system in vertical and
torsional directions are 0.45 and 0.48, respectively. The aeroelastic section model in the wind
tunnel tests is shown in fig 2.
Fig 2. Section model in the wind tunnel tests
III. INFLUENCE OF BRIDGE RAILING
The railings of two porosities (90% and 60%) at three different locations on deck were
tested under three attack angles (−3°, 0° and 3°). The three locations are as follows: one is on
the edge (the original design), the other two are 5 mm (prototype 250 mm) and 10 mm
(prototype 500 mm) away from the sideway edge. The test results are given in tab 1.
Porosity
90%
Tab 1. Critical flutter wind speeds for different railings (m/s)
Attack angle
Position
0°
+3°
−3°
Initial position
73.0
61.3
>74.6
60%
Initial position
>75.3
57.6
44.8
60%
250 mm inside
>74.8
58.6
42.4
60%
500 mm inside
>76.2
57.9
44.0
We find that the railings with high porosity strengthen the aerodynamic stability. Similar
findings were also reported in refs. [1]-[3]. The railings of low porosity mean the deck
configuration being in the form of an H-shape, which may lead to flow separation. Then a
rhythmic vortex shedding will lead to the generation of separation bubbles above the deck. The
vortex creation and drift process will not only dramatically weaken the stability of girder, but
also increase the amplitude of VIV [11]-[12]. The test results in tab 1 indicate that the railings of
low porosity weaken the aerodynamic performance of bridge girder. However, the railings of
high porosity are harmful to the safety and durability of long-span bridges. Therefore, in the
design of railings, both the safety of bridge and the aerodynamic performance of the bridge
girder should be ensured. Tab1 also shows that the critical wind speed of the girder is not
sensitive to the railing position change.
20
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
IV. INFLUENCE OF THE RAIL OF BRIDGE INSPECTION CAR
The aerodynamic stability of the girder may be sensitive to the distance between the rail
top and the girder bottom, called the rail gap, and also sensitive to the fixed position of rail on
the girder. Tests on three gaps were carried out to investigate the potential influence. The gap
was increased by 1 mm and 3 mm (prototype 5 cm and 15 cm) in the tests (the porosity of
railings is 90%). Flutter critical speeds of these sections were measured, and are summarized in
tab 2. It is found that increasing gap is beneficial to the critical wind speed at 0° and −3° attack
angle, but harmful to the aerodynamic stability of the girder at +3° attack angle. It can be seen
from tab 2 that the critical flutter wind speed decreases with an increase in the rail gap. Thus,
increasing the rail gap is not a good choice to strengthen the aerodynamic stability of the girder.
Case
Initial design
Increment of
1 mm
Increment of
3 mm
Tab 2. Critical flutter wind speed under different gap (m/s)
Attack angle
Position
0°
+3°
−3°
5 mm from the
73
61.3
>79.0
bottom
6 mm from the
59.6
>81
>80
bottom
8 mm from the
58.1
>84
>82.5
bottom
The sketch maps for three different fixed positions of rails are shown in fig 3, and the VIV test
results of the three cases under +3° attack angle are shown in fig 4. If the rail was close to the
corner of the girder bottom (see fig 3-(a)), which is the separation point of incoming flow, the
amplitude of VIV would become too large to exceed the code limit. If the rail was far from the
corner of the girder bottom (see fig 3-(b)), the amplitude of VIV would become small, or tend to
zero. If the guide vane was fixed at both sides of the rail in the original design, shown in fig 3-(c),
the VIV amplitude could decrease significantly. The position of rail has a large effect on the VIV.
800
(a) Original design
4 800
(b) The adjusted position of rail
800
(c) Original design with guide vane
Fig 3. Positions of the rail of bridge inspection car (unit: mm)
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
21
Amplitude of VIV (mm)
200
Original position
Adjusted position
Original position
with guide vane
150
100
50
0
2
4
6
8
10
Wind speed (m/s)
Fig 4. VIV amplitudes for different rail positions
V. INFLUENCE OF GUIDE WING
The guide wing on the edge of sideway, shown in fig 5, can smoothen the airflow passing
through the section, which may strengthen the aerodynamic stability [1]-[3]. A total of nine
different types of guide wings were employed in the intensive wind tunnel tests, with different
widths and obliquities. The railing porosity was 60% for the section model in the tests. Flutter
wind speeds were obtained in the tests for the section model under attack angles of +3° and 0°,
and the results are shown in tab 3.
Tab 3. Critical flutter wind speed with guide wings (m/s)
Guide wing
Attack angle
Width (cm)
Obliquity (°)
0°
3°
50
+15
52.2
44.9
50
0
51.47
41.98
50
−42
54.75
42.34
100
+15
54.96
52.05
100
0
52.78
50.23
100
−42
58.77
45.63
125
+15
60.23
63.88
125
0
56.94
55.48
125
−42
62.78
49.64
It is observed from tab 3 that, with an increase in the width of the guide wing with a
positive obliquity, the aerodynamic stability is strengthened even though the railings have a low
porosity. However, the guide wing will increase the complexity of the structure and its
construction, particularly in the location of rostra, and the maintenance cost will increase
accordingly. Therefore, a guide wing fixed at girder sides is not recommended in the design
unless there is no alternative.
22
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
Obliquity
Width
Girder
Guide
(b)
(a)
Fig 5. Outline of guide wing
VI. INFLUENCE OF ROSTRUM
The critical wind speed is sensitive to the shape of the rostrum [1]-[3], [11]. Therefore, the
rostrum with different widths and angles was taken into account in the tests. The angle of the
rostrum varied from 57° to 25°; correspondingly, the width varied from 1.9 m to 3.3 m. A total of
21 model cases were tested. The results are shown in tab 4.
Tab 4. Critical flutter wind speeds with different rostra (m/s)
Attack angle
Rostrum
Angle (°)
Width (m)
+3°
−3°
−3°
57
1.9
51.5
>69.6
>70.6
47
2.1
53.7
>70.7
>70.3
41
2.3
54.6
>71.2
>70.5
37
2.5
56.7
>70.8
>71.2
33
2.7
58.6
>71.5
>75.2
30
3.0
63.4
>72.5
>70.9
25
3.3
59.3
>71.3
>72.8
It is noted that with an increase in the rostrum width and a decrease in the rostrum angle,
the critical flutter wind speed increases. However, the critical flutter wind speed declines when
the width of rostrum exceeds 3 m, which may weaken the stability, and the reason will be
discussed in the later section of this paper.
Similar conclusions were obtained from the VIV tests. The amplitude of the VIV decreases
with a decrease in the angle of the rostrum (see fig 6). Thus, it can be inferred that the blunter
the rostrum, the larger the VIV amplitude. The reason is that being the first separation point of
the incoming flow, when the rostrum becomes blunter, the flow is easier to separate, and the
vortex will form and drift over the girder, which can intensify the VIV amplitude. According to
this result, a blunter rostrum, or the similar aerodynamic components which make the rostrum
blunter, such as the ordinary sidewalk board, is not recommended in the design.\
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
23
Amplitude of VIV (mm)
250
57° rostrum
41° rostrum
30° rostrum
200
150
100
50
0
2
4
6
8
Wind speed (m/s)
Fig 6. VIV test results with different rostra
VII. INFLUENCE OF INCLINED WEB
A wide and acuminate rostrum is difficult to fabricate and fix, implying more cost in design
and construction, although it can strengthen the aerodynamic stability of the girder. An alternate
way is to decrease the steepness of sidewall slope to make the deck cross-section more
streamlined. The critical flutter wind speed of the girder model with different slopes of the
inclined web is shown in tab 5 (the width of rostrum was 2.4 m), and the value increases with
the decrease of the slope. When the slope is decreased to 15°, the critical flutter wind speed rose
to 67.1 m/s, directly improving the flutter instability by 10%. From tab 5, we can obtain an ideal
section of girder for improving flutter instability: short rostrum, no guide wing, and railings of
low porosity.
Tab 5. Critical flutter wind speed with different slopes (m/s)
Slope of inclined web (°)
Attack angle
+3°
0°
−3°
22
44.8
57.6
>75.3
20
56.7
>70.8
>71.2
18
60.4
>72.5
>70.9
15
67.1
>71.3
>73.5
In addition, we have a new finding in the VIV tests. When the slope of inclined web is
decreased to 15°, without the railings and the rail of the bridge inspection car, there is no VIV
phenomenon. The results of VIV tests in different cases are shown in fig 7, where the slope of
inclined web is 22° for original design, 15° for the adjustments without railings and without rail.
Compared with the original design, the bare girder section without aerodynamic accessory
components does not experience VIV. There is a clear conclusion that the rail of the inspection
car on the bottom of girder will intensify VIV. The same conclusion was reached by Larsen [14]
in 2008, who first discovered and documented this phenomenon. Furthermore, he verified this
finding by considering two other bridges.
24
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
Amplitude of VIV (mm)
200
Original design
Without railing
Without rail
150
100
50
0
2
4
6
8
10
Wind speed (m/s)
Fig 7. Comparison between different VIV test results
VIII. DISCUSSION ON AERODYNAMIC PERFORMANCE IMPROVEMENT
Based on the experience from vortex shedding tests, Larsen [13] found that the streamline
box sections were similar to the airfoils at aerodynamic performance, and concluded that the
flow along the bottom plate would stay mainly attached if the slope of inclined web was less
than approximately 16°. He conducted a series of special tests and verified the above conclusion
through the flow visualization technique [14]. He also gave an example about the design for a
two span suspension bridge in Chile, whose inclined web angle was 14.8°, and no VIV was
observed in the wind tunnel testing. On the contrary, explicit VIV was observed for the box
girders of the Great Belt East Bridge and Osterøy Bridge, whose inclined web slope angles were
26.6° and 29.5°, respectively [14].
By analyzing Larsen’s test results and the test results of fig 7, we can draw the conclusion:
vortex shedding excitation originated from rhythmic vortex formation under the downwind
inclined web of the box section can be eliminated by choosing the slope of inclined web at
about 15° and shielding rail (see the test case without rail in fig 7).
Similar to a box section in the vortex shedding vibration status, in the flutter critical status,
two vortices with opposite directions were observed at the both sides of the nose-tail line of
Great Belt East Bridge section through the PIV (particle image velocimetry) technique [15]. The
two vortices could give the girder enough momentum to increase its VIV amplitude in a short
time and finally made the girder instable. When the wind speed was low and below the flutter
critical speed, the positive vortex below the nose-tail line was more powerful than the above
negative one, and the aerodynamic force was just a static lift force; when the wind speed
increased close to the flutter critical speed, the negative vortex above the nose-tail line was
strengthened to a level as powerful as the positive one below, and the aerodynamic force
fluctuated [15]. If the frequency and the phase of the fluctuated force induced by vortex shedding
are close to the bridge’s modal frequency and phase, the flutter of girder will occur soon.
In the tests conducted in this paper, when the slope of the inclined web is decreased to 15°,
there is a small dead wake region below the nose-tail line and the flow along the bottom plate
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
25
stays mainly attached to the web, which makes it more difficult for formation of a large vortex.
When the wind speed increases and approaches to the original critical flutter wind speed
(45 m/s), the opposite vortices can not become powerful, and can not give the girder powerful
and efficient excitation. Fig. 8 shows the two vortices near the tail of the girder at this status.
They are so asymmetrical that they can not produce the fluctuated aerodynamic force, and thus,
the bridge remains stable. When the wind speed continues to increase, the equivalent opposite
vortices will form at the similar position and the rhythmic excitation will return, which introduce
aerodynamic instability to the girder once more. Fig. 9 reflects the status of the vortex moment at
the wind speed approaching to the flutter critical value, and the equivalent vortices near the tail
of the girder can produce the fluctuated aerodynamic force that makes the bridge instable.
Fig 8. Vortex moment at the low wind speed (below the critical value)
Fig 9. Vortex moment when the wind speed approaches to the critical value
Therefore, vortex shedding excitation originated from the equivalent opposite vortices near
the tail of the box section can be restrained and delayed by choosing the inclined web angle at
15°. The test results in tab 5 can verify this conclusion. In aerodynamic stability design of longspan bridges, to increase the critical flutter wind speed, the same design of the bridge girder
with the inclined web at 15° is recommended.
This explanation can also be extended to interpret the results shown in tab 4. When the
slope of the inclined web is less than 16°, the increase of the width of the rostrum leads to the
formation of a large vortex, which is a disadvantage to the girder stability and make the critical
flutter wind speed decrease. Tab 6 shows the test results of the critical flutter wind speed with
different width of rostra for inclined web of 15°. We can conclude that the short rostra are of
benefit to the girder stability when the slope of inclined web is 15°.
26
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
Tab 6. Critical flutter wind speed (slope 15°) (m/s)
Rostra
Attack angle
Width (m)
Angle (°)
+3°
0°
−3°
2.4
40
67.1
>71.3
>73.5
2.6
36
62.0
>70.8
>72.7
2.8
33
61.1
>71.5
>72.4.
It should be noted that the explanation documented above is based on the Larsen’s research
[14] on the vortex shedding vibration of streamlined box girder, and Zhang’s research [15] on
the flutter critical status of Great Belt East Bridge by PIV technique. The explanation is
obtained by a logical deduction about aerodynamic stability improvement. It needs to be
verified by the further study of the wind tunnel tests using PIV technique and computational
fluid dynamics method in the future.
IX. CONCLUSIONS
In this paper, more than 60 cases were tested through a 1 ‫׃‬50 scale section model to study
the influence of aerodynamic configuration and accessory components on the flutter and vortexinduced vibration (VIV), such as the railing, the rail of inspection car, the rostrum, the guide
wing, and the slope of inclined web. From the tests, we can obtain several important conclusions
for the aerodynamic design of streamline box girder in long-span bridges. The girders with
railings of high porosity have higher critical flutter wind speed than the ones with railings of
low porosity. The wide and acutance rostrum can strengthen the aerodynamic stability of the
girder, but the width of rostrum cannot exceed 3 m. The rail of the bridge inspection car fixed at
the corner of bottom plate of box girder can intensify VIV. It is recommended that the rail
should to be shielded by guide vanes.
If the inclined web with a 15° angle is employed in the section design of long-span bridges,
the flutter performance can be enhanced or the critical flutter wind speed can be increased
dramatically. The VIV of long-span bridges under the practical structural damping can be also
suppressed by this same design.
Through the wind tunnel tests, we find that the final design section of the Nanjing 4th
bridge, including the railings of 60% porosity, shielded inspection car rail, inclined web with a
slope of 15° and 2.4 m wide rostrum, has a good aerodynamic stability, and the critical flutter
wind speed is up to 67.1 m/s. Of course, there is no vortex shedding vibration observed in the
1 ‫׃‬50 and 1 ‫׃‬20 section model tests. This section model offers a good demonstration for the
aerodynamic configuration designs of other long-span bridges. The points of this paper have been
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
27
validated by the aerodynamic configuration design of Huangyi Bridge crossing the Yangtze River,
which is located in Luzhou City, Sichuan Province.
References
[1]. A. Larsen, Aerodynamic aspects of the final design of the 1 624 m suspension bridge across the Great
Belt, Journal of Wind Engineering and Industrial Aerodynamics, 1993, 48(2-3): 261-285.
[2]. T. Miyata, Historical view of long-span bridge aerodynamics, Journal of Wind Engineering and
Industrial Aerodynamics, 2003, 91(12-15): 1393-1410.
[3]. L. Bruno, G. Mancini, Importance of deck details in bridge aerodynamics, Structural Engineering
International, 2002, 12(4): 289-294.
[4]. K. Wilde, P. Omenzetter, Y. Fujino, Suppression of bridge flutter by active deck-flaps control system,
Journal of Engineering Mechanics, 2001, 127(1): 80-89.
[5]. Y.X. Yang, Y.J. Ge, Some practices on aerodynamic flutter control for long-span cable supported
bridges, In: Proceedings of the 4th International Conference on Advances in Wind and Structures
(AWAS’08), Jeju, Korea, May 28-30, 2008.
[6]. A. Larsen, S. Esdahl, J.E. Andersen, et al., Storebaelt suspension bridge-vortex shedding excitation
and mitigation by guide vanes, Journal of Wind Engineering and Industrial Aerodynamics, 2000, 88(2-3):
283-296.
[7]. A. Larsen, M. Savage, A. Lafrenière, et al., Investigation of vortex response of a twin box bridge
section at high and low Reynolds numbers, Journal of Wind Engineering and Industrial Aerodynamics,
2008, 96(6-7): 934-944.
[8]. G. Diana, F. Resta, M. Belloli, et al., On the vortex shedding forcing on suspension bridge deck,
Journal of Wind Engineering and Industrial Aerodynamics, 2006, 94(5): 341-363.
[9]. Y.J. Ge, Y.X. Yang, F.C. Cao, VIV sectional model testing and field measurement of xihoumen
suspension bridge with twin box girder, http://www.dist.unina.it/proc/2011/ICWE13/start/papers/502
_8page_viv_sectional_model_testing_and_field_measurement_of_xihoumen_suspension_bridge_with_t
win_box_girder.pdf, 2011-10-30.
[10]. JTGT D60-01-2004, Wind-Resistant Design Specification for Highway Bridges.
[11]. J.Z. Song, Z.X. Lin, J.Y. Xu, Research and appliance of aerodynamic measure s about wind
resistance of bridges, Journal of Tongji University (Natural Science), 2002, 30(5): 618-621 (in Chinese).
[12]. C.J. Liu, Z.S. Guo, L.D. Zhu, Influence of railing curbstone structure on flutter stability of box main
girder, Bridge Construction, 2008(2): 20-22,44 (in Chinese).
[13]. A. Larsen, Aerodynamic stability and vortex shedding excitation of suspension bridges, In:
Proceedings of the Fourth International. Conference on Advances in Wind and Structures (AWAS’08),
Jeju, Korea, 2008.
[14]. A. Larsen, A. Wall, Shaping of bridge box girders to avoid vortex shedding response,
http://www.dist.unina.it/proc/2011/ICWE13/start/papers/633_8page_shaping_of_bridge_box_girders_to_avoi
d_vortex_shedding_response.pdf, 2011-10-28.
[15]. W. Zhang, Y.J. GE, Flow field mechanism of wind induced vibration response of large span bridge
influenced by guide vanes, China Journal of Highway and Transport, 2009, 22(3): 52-57 (in Chinese)♦
28
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
APPLICATION OF VIRTUAL SIMULATORS FOR TRAINING
STUDENTS IN THE FIELD OF CHEMICAL ENGINEERING AND
PROFESSIONAL IMPROVEMENT OF PETROCHEMICAL
ENTERPRISES PERSONNEL
KRASNYNSKIY M.N
NIKOLAEV A.B
OSTROUKH A.V
State Technical University – MADI,
Moscow, Russia
Summary: The structure and contents of training and control system formed on the basis
of LabVIEW programming environment are presented. The training and control system can be
used for teaching students in the field of chemical engineering and professional improvement
of petrochemical enterprises personnel. This work was supported by the Government of the
Russian Federation (Russian Ministry of Education) as part of the project under the Contract
№ 13.G25.31.0064 on October 22, 2010.
Key word: Chemical technology, informational system, virtual simulator, SCADA,
professional improvement, distant education.
Modern petrochemical industry actively uses automated control systems. These systems
not only help improving the product quality, but also provide convenient and simple tools to
monitor and manage technological processes and prevent possible extraordinary situations. In
order to use such systems, the enterprises’ personnel should be properly trained. Thus, it is
necessary to create simulators, which are intended to give students and chemical enterprises
employees an opportunity of practicing and improving their professional skills in the field of
using the above systems.
Today there is plenty of software intended for training the personnel of industrial enterprisessimulators, programs for testing and so on. Introduction of such software packages at the
enterprise raises the quality of personnel training and contributes to the formation of skills. It
should be noted that computer-aided training involves the use of visual methods; besides, it is very
convenient and easy to use. Simulators and training programs became very popular in chemical
and power industries since the employees at these kinds of enterprises quite often use remote
controls in their work; these operations can be easily performed by means of programs-simulators.
Besides, application of programs-simulators seems to be quite promising in training
students at higher educational institutions in the petrochemical industry. It is very important for
students to have both theoretical and practical knowledge which will enable them to efficiently
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
29
handle professional issues. This can be achieved through application of systems simulating the
work of particular technological lines. Thus, two goals are achieved: on the one hand, students
have deeper understanding of the studied material as representation of devices performance in
dynamics is frequently more informative than text descriptions and static illustrative materials;
on the other hand, students have an opportunity to get practical skills required for their further
work, without the necessity of using real equipment which is quite often unavailable. It is
obvious, that in such cases the use of simulators is, practically, a unique way to give students the
necessary knowledge and skills.
The department “Computer-aided design of the process equipment” of Tambov State Technical
University is engaged in the development of the automated system for control and training of
the students of the specialization “Flexible automated systems in technology of machines and
devices of chemical production” and the personnel of “Pigment” plc, Tambov. The given
system is developed on the basis of LabVIEW programming media made by the company
National Instruments.
The developed system includes the following components:
1. The virtual simulator of a workstation of the operator, controlling the work of pigment making device.
2. The module of trainee testing, providing the knowledge check of simulated technological
processes and chemical technologies as a whole.
3. The help system including the description of simulated production technology.
The training module represents a set of virtual tools created in LabVIEW system. It
consists of two basic components: a simulator intended for practicing actions in case of
emergency, and a simulator imitating the regular work of the technological circuit.
The simulator “PLAS-T” (fig 1) represents the virtual tool aimed at improving the
personnel actions in case of emergency, created on the basis of the existing Plan of Emergency
Localization (PEL) of the plant specialized in production of monometylaniline at “Pigment”
plc, Tambov. It represents a complex including the forward panel, the panel of preliminary
adjustments, subsystems of answers’ processing and those of testing results output.
The basic features of the system “PLAS-T” are:
- Conformity of the skills formed via the simulator, to the skills obtained through working
life (it is ensured by the fact that the simulator is based on the existing PEL which is the basic
document regulating the behavior of people in charge if the emergency occurs) and, thus, it
secures the identity between the actions which operator is obliged to take in case of real failure
and those which he does while being trained on the simulator;
- Inhibition of the skills giving a negative effect when put into real conditions as the trainee learns
about the mistake or the wrong answer thus helping them to avoid a similar mistake in further testing;
30
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
Fig 1. Main panel of simulator “PLAS-T”
- An opportunity to vary testing conditions; it is ensured by two time modes of testing and the
opportunity to choose a stage of technological process, its conditions and type of emergency;
- Registration of the results required for further analysis; these are presented inside the
program in the form of a chart showing correct and incorrect answers, and as an external file
containing complete information on testing;
- Methodical purposefulness of the simulator covering all possible options of emergency
occurrence and development as stipulated in PEL and a set of the exercises for students.
A number of training modules forming “The complex of virtual simulators for chemical
technology systems operators” have been developed by the students of the Tambov State
Technical University on the basis of the technique developed during the creation of the
described training modules, which you can find at http://www.170514.tstu.ru/tren/index.html
(fig 2) is developed.
The testing module is created on the basis of “Knowledge control system Knost 1.0.4.”
(www.scorp.ru), designed for the creation of electronic tests, testing and viewing of its results.
With the help of administrator module it is possible to create tests and carry out the remote
testing on a local network. The test created in the system Knost, represents a set of questions and
several answer options to each question. Each question may have an image in .bmp format, stored in
a test file. Besides, each question has the so-called maximum point received for the correct answer.
Thus, it is possible to identify the complexity of the question and its effect on the total score.
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
31
Fig 2. A web-page of the complex of virtual simulators for chemical technology systems operators
The set of questions together with the answer options covering both general knowledge of
chemical production technology and the particular production process presented in the module
are stored in the test file. When the test is over, the file is created; its name contains the ID of
the student who has completed the test, and the title of the test itself. This file can be viewed
only with the help of the program «Knowledge Control: Administrator». The score file contains
the following information: the title of the test and the student’s surname, the number of
questions, correct and wrong answers, factor of knowledge and a mark given according to a five
- point scale. Besides, the examiner can look through the answers given by the student.
The result of the represented work is the complex system which contains the testing
module for the high schools students and petrochemical enterprises employees intended to
identify the level of knowledge of technological processes. The application of the given system
makes it possible to arrange practical classes for the students, and teach them skills required for
petrochemical enterprises. The other important area of application of the given system is
training and retraining of the petrochemical enterprises personnel, revealing the degree of their
readiness to various situations, including cases of emergency.
References
[1]. Malygin E.N., Krasnyansky M.N., Karpushkin S.V., Mokrozub V.G., Borisenko A.B. New information
technologies in the open engineering education. Textbook. / / M.: "Machinostroeniye", 2003. P. 90-123♦
32
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
OPTIMAL CONTROL STRATEGY FOR ENERGY SAVING
IN TRAINS UNDER THE FOUR-ASPECT
FIXED AUTOBLOCK SYSTEM
QIHENG LU
XIAOYUN FENG
Southwest Jiaotong University,
Chengdu 610031, China
Summary: This paper deals with both the leading train and the following train in a train
tracking under a four-aspect fixed autoblock system in order to study the optimum operating
strategy for energy saving. After analyzing the working principle of the four-aspect fixed
autoblock system, an energy-saving control model is created based on the dynamics equation
of the trains. In addition to safety, energy consumption and time error are the main concerns
of the model. Based on this model, dynamic speed constraints of the following train are
proposed, defined by the leading train dynamically. At the same time, the static speed
constraints defined by the line conditions are also taken into account. The parallel genetic
algorithm is used to search the optimum operating strategy. In order to simplify the solving
process, the external punishment function is adopted to transform this problem with
constraints to the one without constraints. By using the real number coding and the strategy of
dividing ramps into three parts, the convergence of GA is accelerated and the length of
chromosomes is shortened. The simulation result from a four-aspect fixed autoblock system
simulation platform shows that the method can reduce the energy consumption effectively in
the premise of ensuring safety and punctuality.
Key words: Leading train; following train; four-aspect fixed autoblock system; optimal
control strategy of energy-saving; train tracking; dynamic speed constraints; genetic
algorithm.
I. INTRODUCTION
Railway transportation departments, depending on energy heavily, have a responsibility to
save energy. Normally, train diagrams allow drivers to select different operating strategies,
which correspond to different energy consumption.
A number of studies have been conducted to optimize train operating strategies for saving
energy. With the help of genetic algorithm (GA), and using the control of coasting position,
Chang and Sim [1] optimized the operating strategy of subway trains. In [2]-[5], the problem
was solved by using the tool of K-T condition in optimal control theory, and the optimal control
model for saving energy was built based on train control notch. By analyzing the structure of a
typical section, Jin et al. [6] proposed an optimization method of train operation for saving
energy based on GA. Then Fu [7] put forward a strategy of operating a following train for
saving energy based on the cellular automaton model.
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
33
However, [1]-[6] all focused on a single running train; [7] aimed at the following train in a
train tracking, and did not involve the leading train. In fact, every train running on a real line is
interfered not only by the line conditions, but also by its leading train, whose interference is
represented by the control of the signal system. Unlike that of the line conditions, the
interference from the leading train is dynamic. Compared with the studies on a single train, the
research targeted at both of the leading train and following train is more complicated. With the
help of GA, this paper deals with the leading and the following trains in a train tracking and an
optimum operating strategy is proposed for the two trains, based on the optimal operating model
of trains for saving energy.
II. FOUR-ASPECT FIXED AUTOBLOCK SYSTEM
The working principle of the four-aspect fixed autoblock system is illustrated in fig. 1, with
two features. One is that the signals have four aspects, which can predict the block status of
three blocks forward. The other is that there are three speed grades and the braking distance
from the specified speed to 0 requires two blocks.
Following train
Leading train
Signal
Braking curve
Fig 1. Working principle of the four-aspect fixed autoblock system
III. MODEL
3.1. Description of the problem
Under the control of the signal system, the leading train and the following train depart in
turn. After finishing their trip, they all arrive at the destination and stop. The energy
consumption of the leading train and the following train are Q1 and Q2, respectively. Let f(Xi) =
Qi (i = 1,2), and then the object function is demonstrated as
min( f ( X1 ) + f ( X 2 ))
(1)
The control variables of the function f(Xi) are the train control notch and the train position
when its control notch is changed. Because the control notch could be changed many times
during the whole trip, the variable Xi in (1) should be a two-dimensional matrix as
⎛ l1 ,l2 ,",ln ⎞
X iT = ( x1 ,x2 ,",xn ) = ⎜
⎟
⎝ p1 ,p2 ,",pn ⎠
(2)
Where xj (j=1,2,…,n) is the element of matrix Xi, which include the control notch lj and the
train position pj. Thus, (1) is transformed into
min( f ( X1T ) + f ( X 2T ))
34
(3)
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
3.2. Dynamic equations of the trains
The running trains are affected by the traction, resistances, and braking forces, which result
in different trip time and energy consumption. The dynamic equations of the trains are as
follows:
a=
( f ( v ) − r( v ) − p( s ) ) × 9.8
(4)
1 000
r( v ) = a + bv + cv 2
(5)
B = Bk + Bd = ∑ (ϕ h ∑ K h ) + Bd
(6)
U w ∑ ⎡⎣ ( I p 2 + I p 0 ) Δt ⎤⎦
(7)
T
T
0
0
Q(T ) = ∫ Q y ( t ) dt = ∫
60
dt
Where a is acceleration, in m/s2; f(v), r(v) and p(s) are unit driving force, basic resistance
and gradient resistance, respectively, in N/kN; a, b and c are the coefficients of basic resistance;
B is the common brake force of the train; Bk is air brake force and Bd electric brake force, in kN;
ϕ h is the conversion friction coefficient; ∑Kh is the total conversion brake shoes pressure of the
train, in kN; QT is the total energy consumed by the train, in kW·h; Q y ( t ) is the energy consumed
by the train at time t, in s; T is the trip time, in s; UW is the voltage of train pantograph, in kV; IP0
is the active current of the locomotive devices, in A; and, IP2 is the average active current of the
locomotive under partial load, in A.
3.3. Constraints
3.3.1. Boundary constraints
The boundary constraints of trip time and speed are.
t(0) = 0, t( X ) = T , ⎫⎪
⎬
V(0) = 0, V( X ) = 0, ⎪⎭
(8)
Where t(0) and V(0) are, respectively, the time and speed of the train when departing; and t(X)
and V(X) are, respectively, the time and speed of the train arriving at the destination.
3.3.2. Static speed constraints
The static speed constraints are defined by the line conditions, including the curve speed limits,
the tunnel speed limits, the turnout speed limits, etc., demonstrated as
Vmax_ s
⎧Vmax_ s1 , 0 ≤ s < S1 ,
⎪
⎪Vmax_ s 2 , S1 ≤ s < S2 ,
=⎨
"
⎪
⎪Vmax_ sn , Sn −1 ≤ s < X ,
⎩
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
(9)
35
Where Vmax_si is the static speed limits of the section i (i=1,2,…,n), in km/h; Si is the border
position of section i, in m; and, s is the train position, in m.
3.3.3. Dynamic speed constraints of the following train
The dynamic speed constraints of the following train are caused by the leading train, whose
position restricts the speed of the following train for safety. The constraints are closely related to
the positions of the leading train and the following train, their trip time, and the signal system.
Dynamic speed constraints are very important to the safety of the train tracking. After analyzing
the working principle of the four-aspect autoblock system, we formulate the dynamic speed
constraints of the following train as follows:
Vmax_ d
⎧Vmax_ d 1 , s ≥ S1 , t0 ≤ ts < t1 ,
⎪
⎪Vmax_ d 2 , s ≥ S2 , t1 ≤ t s < t2 ,
=⎨
"
⎪
⎪V
⎩ max_ dn , s ≥ Sn , tn −1 ≤ t s < tn
(10)
Where Vmax_di is the max safe speed of the following train in section i during the period from ti-1
to ti, and ti is the time section border of dynamic speed constraints, which is defined dynamically by
the leading train position.
When the leading train passes every signal, the data of the dynamic constraints will be
recalculated according to the working principle of the four-aspect autoblock system.
3.3.4. Constraints of shifting the train operating state
There are three operating states for a train, i.e., motoring, coasting, and braking. The
shifting rule is listed in tab 1 [8].
Tab 1. Principle of shifting train operating state
Operating states to be shifted
Current
operating state
Motoring
Motoring
○
Coasting
Braking
Coasting
Braking
●
○
●
○
○: need not shfit; ◎: can shfit; ●: cannot shfit
3.4. Object function
This problem is difficult to solve with these constraints mentioned above. In order to simplify
the solving process, we introduce an external penalty function method to transform this problem to
another without constraints. The object function is demonstrated as follows:
min E = min( E1 + E2 )
36
(11)
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
T
X
E1 = ∫ Q y ( t ) dt + α (t( X ) − T ) 2 + β ∫ load ( s ) ds
0
T
E2 = ∫ Q y ( t ) dt + α (t( x ) − T ) 2 +
0
(12)
0
X
β ∫ load ( s ) ds + γ ∫
0
X
0
∫
T
0
load ( s ,t ) dtds,
⎧δ , V( s ) > Vmax_ s , ⎫
load ( s ) = ⎨
⎪
⎩0, else,
⎪
⎬
⎧η , V( s ,t ) > Vmax_ d , ⎪
load ( s ,t ) = ⎨
⎪
⎩0, else.
⎭
(13)
(14)
Where E1 is the object functions of the leading train, E2 is the object functions of the
following train, load(s) is the penalty function of the static speed constraints, load(s,t) is the
penalty function of the dynamic speed constraints, and α, β and γ are the factors of the trip time
error, static overspeed and dynamic overspeed.
IV. GENETIC ALGORITHM
The problem under study is a nonlinear one; thus, we use GA to solve it.
4.1. Chromosomes
4.1.1. Structure of chromosomes
As mentioned above, the control variables in (3) are the train control notch and the
corresponding train position, and they must be included in the genes of the chromosomes. The
structure of chromosomes is illustrated in Fig. 2, where l1i and l2i are the control notches of the
leading train and the following train, respectively, which both include motoring (integer, greater
than 0), coasting (0) and braking (−1); p1i and p2i are the positions of the two trains where their
control notches are changed.
l11p11 l12p12 … l1np1n l21p21 l22p22 … l2np2n
Fig 2. Structure of chromosomes
The length of chromosomes should be as short as possible for improving the searching speed
and convergence of GA. The strategy of dividing ramps into three parts [9] is introduced here. That
is, a long ramp can be divided into three parts. The train adopts the same operating mode as the
previous ramp in the first part. At the beginning of the second part, the train changes the operating
mode to fit the gradient of the second part. In the third part, the train changes the operating mode
again to prepare for the next ramp. The real number coding strategy is also applied for the same
purpose.
4.1.2. Fitness function
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
37
The chromosome is evaluated by its fitness value (must be positive) in GA. The larger the
fitness of a chromosome is, the more likely the chromosome is selected into the next generation.
From (11), the following fit-ness function is derived:
f i = 1 ( E1 + E2 ) ,
(15)
Where fi is the fitness of the chromosome i. The fitness function guarantees two necessary
conditions for the fitness. First, it is positive. Second, the less energy the trains consumes, the bigger
the fitness of the chromosome.
4.2. Operators of GA
4.2.1. Selection operator
Roulette wheel selection based on elitism strategy is adopted in this paper. The probability
of each chromosome in every population is calculated by (16). Refs. [10-11] demonstrates that
the GA with elitism strategy is convergent in the global searching scope with the Markov chain:
pi =
fi
=
N
∑
j =1
fj
fi
f sum
(16)
Where fsum is the total fitness of the population, and pi is the selection probability of
chromosome i.
4.2.2. Crossover operator
According to the characteristics of the problem under study, the two-point parallel
crossover operator is adopted. That is, a pair of parent chromosomes selected randomly is cut
off at two positions. One is located in the leading train genes, and the other in the following
train genes. Then the chromosome sections are exchanged and recombined to create the new
offspring chromosomes.
4.2.3. Mutation operator
A mutation operator based on probability is used in this article. GA mutates the
chromosomes randomly based on the preset mutation probability.
4.3. Parameters of GA
The parameters of GA are listed in tab 2.
Genera-tions
Popula-tions
Chromo-somes
100
1
50
Probability
of crossover
0.95
Tab 2. Parameters of GA
Probability
of variation
0.05
V. SIMULATION
In order to verify the effectiveness of the proposed algorithm, a simulation model was
38
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
created with 1 SS8 locomotive and 18 cars. The calculation process and the parameters used
here were all in conformity to Train Traction Calculation Regulation [12]. The line parameters
are listed in tab 3.
Tab 3. Parameters of the line
Name
Departure stop
Destination
Length
Standard trip time
Signal system
Value
Guangzhou
Guangzhoudong
8 510 m
10 m
Four-aspect fixed automatic block signal
system
The profile of the line is illustrated in fig 3, where a speed-distance curve describes an
excellent driver’s operating strategy in a trip from Guangzhou to Guangzhoudong. The total
energy consumption was 262.09 kW·h, and the running time was 559 s [8].
The data of braking process is calculated and the ramps of the line are simplified before the
beginning of the GA according to Train Traction Calculation Regulation [12].
The data of braking process is calculated and the ramps of the line are simplified before
the beginning of the GA according to Train Traction Calculation Regulation [12].
Through iterative calculation of GA, the speed-distance relation of the leading train and the
following train are shown in fig 4 and fig 5, respectively.
The energy consumption of the leading train is 206.84 kW·h and the trip time is 576 s. The
corresponding data of the following train are 201.79 kW·h and 562 s. Compared with the data of
[8] (single train not in a train tracking, 223.086 kW·h and 603 s), the energy consumption of the
leading train decreased 7.3% and the trip time decreased 4.5%, while the corresponding data of
the following train are 9.5% and 6.8%.
The average fitness of all the generations of GA is illustrated in fig. 6. One can see that the
maximum fitness is achieved at the 65th generation.
VI. CONCLUSIONS
(1) The static speed constraints and the dynamic speed constraints guarantee the safety of
trains while running in a train tracking under the four-aspect autoblock system.
(2) It is feasible to optimize the operating strategy of the leading train and the following
train with the structure of chromosomes and parallel crossover (two points) proposed in this
paper.
(3) Using the real number coding and the strategy of dividing ramps into three parts, the length
of chromosomes is shortened and the speed of convergence is improved.
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
39
(4) The simulation result shows that the proposed method can effectively lower the energy
consumption of the two trains in a following operating in the premise of ensuring safety and
punctuality.
Guangzhoudong
Guangzhou
Fig 3. Profile of the line from Guangzhong to Guangzhoudong
and trajectory of the train driven by an excellent driver
Sped of the leading train (km/h)
80
60
40
20
0
0
2,000
4,000
6,000
8,000
10,000
Position of the leading train (m)
Fig 4. Trajectory of the leading train
Sped of the leading train (km/h)
80
60
40
20
0
0
2,000
4,000
6,000
8,000
10,000
Position of the leading train (m)
Fig 5. Trajectory of the following train
40
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
0.20
Average fitness
0.18
0.16
0.14
0.12
0.10
0.08
0
20
40
60
80
100
Generation number
Fig 6. Relation of GA generation and average fitness
ACKNOWLEDGEMENTS
This paper was supported by the National Science & Technology Pillar Program during the
Eleventh Five-Year Plan Period of China (No.2009BAG12A05).
References
[1]. C.S. Chang, S.S. SIM, Optimizing train movements through coast control using genetic algorithm,
IEEE Proc-Electr. Power Appl., 1997, 144(1): 65-73.
[2]. J.X. Cheng, P. Howlett, Application of critical velocities to the minimization of fuel consumption in
the control of trains, Automatica, 1992, 28(1): 165-169.
[3]. J.X. Cheng, P. Howlett, A note on the calculation of optimal strategies for the minimization of fuel
consumption in the control of trains, IEEE Trans. on Automatic Control, 1993, 38(11): 1730-1734.
[4]. P. Howlett, Optimal strategies for the control of a train, Automatica, 1996, 32(4): 519-532.
[5]. J.X. Cheng, J.S. Cheng, J. Song, et al., Algorithms on optimal driving strategies for train control
problem, In: Proceedings of the 3rd World Congress on Intelligent Control and Automation, Hefei: IEEE
Press, 2000: 3523-3527.
[6]. W.D. Jin, Z.L. Wang, C.W. Li, et al., Study on optimization method of train operation for saving
energy, Journal of the China Railway Society, 1997, 19(6): 58-62 (in Chinese).
[7]. Y.P. Fu, Research on modeling and simulations of train tracking operation and saving energy
optimization [Dissertation], Beijing: Beijing Jiaotong University, 2009 (in Chinese).
[8]. Q. He, Train optimized control based on genetic algorithm and fuzzy expert system [Dissertation],
Chengdu: Southwest Jiaotong University, 2006 (in Chinese).
[9]. Y.S. Li, Z.S. Hou, Study on energy-saving control for train based on genetic algorithm, Journal of
System Simulation, 2007, 19(2): 1-4 (in Chinese).
[10]. Y. Shi, S.L. Yu, Real-coded crossover operator and improved real-coded genetic algorithm, Journal
of Nanjing University of Posts and Telecommunications (Natural Science), 2002, 22(2): 42-45 (in
Chinese).
[11]. D. Lin, M.Q. Li, J.S. Kou, On the convergence of real-coded genetic algorithms, Journal of
Computer Research and Development, 2000, 37(11): 1322-1326 (in Chinese).
[12]. The Ministry of Railways of The People’s Republic of China, TB/T 1407-1998 Train traction
calculation regulation, Beijing: China Railway Publishing House, 1998 (in Chinese)♦
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
41
Y THE ORIENTED APPLICATION
OF WEATHERING STEEL FOR BRIDGE IN VIETNAM
PROF. DR. NGUYEN VIET TRUNG
DR. TRAN VIET HUNG
University of Transport and Communications, Vietnam
Summary: Steel bridges are very sensitive with environmental conditions, thus it is
necessary to protect rust of bridge structure. However, the cost of the anti-rust and
maintenance is especially high. Recently, new steel technologies have rapidly developed and
created many new steel materials by steelmakers such as high-performance steel, weathering
steel. Weathering steel is available anti-rust by itself and low maintenance cost. The exposure
test specimens were attached on Cho Thương Bridge. The results of the test based on
weathering steel (Japanese Industrial Standard, JIS-SMA-W) has showed good corrosion
protection. Thus, the oriented application of weathering steel on the bridges in Vietnam,
where have a long coastline of more than 3,200 km, is studied in this paper.
Key word: Bridge, steel bridge, weathering steel, high performance steel
I. INTRODUCTION
Steel bridges without painting or other protective coatings have been used widely for the
sake of environment protection and low cost (Mathay, W.L., 1993). Recently, weathering steel
bridge has received increased attention because correct applications of weathering steels to
infrastructures are beneficial for reducing maintenance cost. Weathering steel use commonly in
Unites States, named Cor-Ten™, is a group of steel alloys that develop a stable oxidation
requiring no additional coating. Almost weathering steel is ASTM A 588, also known as CorTen™ B, an "improved" alloy. Material is available in rolled shapes and plate conforming to
ASTM A 709, which can be fabricated into various structural shapes, and in coiled sheet
conforming to ASTM A 606, which is formed into roofing and siding panels.
In Japan, the conventional weathering steel specified as Japan Industrial Standard G 3114
SMA (JIS-SMA weathering steel) and advanced weathering steels from three Japanese
steelmakers, are available commercially. When the weathering steel bridges construct, it is
important to judge whether the weathering steels applied to the structure are suited to the local
environment. The Japanese Roads and Bridges Policy Manual states that JIS-SMA weathering
steel can be used without rust controlling surface treatment for bridges in regions where the
deposition rate of air-bornsalt is less than 0.05 mdd (mg/dm2/day) because corrosion loss is
related to deposition rate of air-born salt (Ohya, M. et al., 2009).
A rust layer formed on a low-alloy steel surface is generally considered to be responsible
for protection of the steels against corrosives in atmospheric environment. Weathering steel
possesses high corrosion resistance, approximately twice of that of carbon steels, and therefore
42
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
has been widely used as a structural material in an atmospheric corrosion environment. The
protective ability of the rust layer emerges, the corrosion rate of the weathering steel is not
considerably low and initial un-protective rust contaminates surfaces of itself and other
environing materials. The protective rust layer cannot form in a coastal environment where the
amount of air-borne sea-salt particle is relatively high. These are significant problems for
reducing maintenance task for weathering steel structures (Yamashita et al., 2002). However,
the standard to be applied to advanced weathering steels is not stipulated clearly. Corrosiveness
of atmosphere differs by location, so a corrosion prediction method for both conventional and
advanced weathering steels is needed at the planning and/or design stage to ensure selection of
materials that will enhance structural durability (Ohya, M. et al., 2009).
Almost steel bridges in Vietnam have been constructed long time ago, have not been used
new steel materials and have been down graded serious. Thus, the new steel as weather-resistant
steel is necessary to research for applying in Viet Nam. Three railway bridges in the NorthSouth railway system (N-S railway) have been constructed for this purposing. The results show
an application capacity of new steel materials is good on Vietnam’s condition. Overview of the
development situation of the steel bridge and the ability and potential application of new steel
materials in Vietnam is presented in this paper.
II. THE SITUATION AND DEVELOPMENT OF THE STEEL BRIDGE IN VIETNAM
Viet Nam has along a coastline of 3,260 km across the territories of 24 provinces and cities,
which including 127 urbans and rural districts, 21 towns and 6 large-cities (Ha Long, Vinh,
Hue, Quy Nhon, Nha Trang, Vung Tau). The national highway system runs nearly a coast, and a
railway system runs cross sparsely populated areas with a lot of steel bridges. The rust
protection of the structural steel bridges are necessary maintain within a short period, however,
it have difficult with cost and technology. Therefore, a lot of steel bridges are being
seriously rust in Vietnam (shown in Fig. 1). Steel bridges will be built in the new transport
systems. Thus, it is necessary to research the applied orientation of new steel technology to
bridge in the future.
Fig. 2 shows the concrete bridges of 89% and the steel bridges of 11% in
Vietnam). However, the steel bridges have constructed long time ago. Almost steel bridges have
been constructed since more than 20 years ago, such as Thang Long truss bridge was completed
since 1980, Chuong Duong truss bridge was completed since 1986, etc. (shown in fig. 3). Steel
materials used mostly carbon steel and low alloy steel as shown in table 1. Recently, few steel
bridges have been constructed in Viet Nam such as the Ho truss bridge was completed in 2000;
the Binh cable stayed bridge with main span of 250 m was completed in 2005. This bridge is the
first cable stayed bridge used steel girder in Vietnam with 17 spans of continuous composite
girder, a main span of 260m, effective width of road 22.5 m (4 lanes + 2 sidewalks) and total
length of 1.280 m. Steel girder and steel anchor box used steel of 6.700 tons including SM570,
SM520C, SM490Y, SM490 and SM400A following to Japan Standard-JIS (Matsuno et al.,
2006); the Thuan Phuoc bridge is the first suspension bridge used steel box girder in Vietnam
with main span of 405 m was completed in 2009; the Rao II and the Nhat Tan cable stayed
bridges are constructing in Vietnam.
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
43
Fig. 1. Corrosion of steel bridge in Vietnam
Steel bridge
11% Concrete bridge 89% Fig. 2. Percent of steel bridge in Vietnam (Source: VRA)
Binh bridge
Thang Long bridge
Fig. 3. Typical steel bridge in Vietnam
Chuong Duong bridge
Vietnamese specifications for bridge design and material standards have been changed as
22TCN 18-79, (referred to Russia standard) has been replaced by 22TCN 272-05 (referred to
ASSHTO LRFD 1998). The new specifications for bridge design (22TCN 272-05) has been
officially used on 2005, and materials standard based on ASTM standard. Currently, Vietnam
has no design standard requirements for high-performance steel materials as weathering steel.
Table 1. Characteristics of steel structures were applied in Vietnam
Carbon steel following to Vietnam standard (TCVN 5709: 1993)
Yield strength, fy (N/mm2) with thickness, t (mm)
Symbol
t ≤ 20
20 ≤ t ≤ 40
40 ≤ t ≤ 100
CCT34
220
210
200
CCT38
240
230
220
CCT42
260
250
240
Low alloy steel following to Vietnam standard (TCXDVN 338: 2005)
Yield strength, fy (N/mm2) with thickness, t (mm)
Symbol
t ≤ 20
20 ≤ t ≤ 40
40 ≤ t ≤ 100
09Mn2
310
300
14Mn2
340
330
16MnSi
320
300
290
09Mn2Si
330
310
290
10Mn2Si1
360
350
340
10CrSiNiCu
400*
400*
400*
Note: * γM = 1.1 with maximum thickness of 40 mm
44
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
III. DEVELOPMENT AND CHARACTERISTICS OF WEATHERING STEEL
Weathering steel used for bridges is low alloy steel containing small amounts of corrosion
resistant elements such as copper (Cu), nickel (Ni), and chromium (Cr) (Table 2). During its use
without painting, compact and protective rust having good adhesion forms on steel surfaces,
which suppresses further progress of corrosion to a sufficiently low level. A characteristic of
weathering steel bridge is the rusty layer on weather resistant steel restrains corrosion. The
usage of this steel helps to reduce the total maintenance cost. A consideration of bridge lifetime
when designing, the life cycle cost can be reduced.
Table 2. Ni-advanced weathering steels
Material
Components
3.0% Ni-Cu
2.7% Ni-Cu-Ti
Ni-advanced weathering steel
2.5% Ni-low carbon
1.5% Ni-Mo
1.2% Ni-Cu
1.0% Ni-Cu-Ti
JIS Weathering steel
0.3% Cu – 0.5% Cr
Fig. 4. Stable oxidant layer (Source: Japan bridge associate)
The application of weathering steel to steel bridges began in the 1960s in Japan. The
protective rust does not always form as designed owing mainly to air-borne sea-salt, resulting in
lamellar exfoliation of rust layers. In view of the situation, unpainted use of the weathering steel
for bridges is recommended, without requiring actual measurement of the air-borne sea-salt,
only at locations where the deposition of airborne sea salt is 0.05 mdd (mg/dm2/day) or less,
measured by the method described in JIS Z2381 (Japan Road Association, 2002). This steel
possesses a unique property of suppressing the development of corrosion by a layer of denselyformed fine rust on its own steel surface: the corrosion rate gradually reduces to the level that
causes virtually no damage from an engineering viewpoint, as the layer of the rust grows. Thus,
the coating is not required in the weathering steel bridge, the cost of which therefore can be
much lower than that of a conventional steel bridge. Fig. 4 shows characteristic of 3%-Ni
weathering steel in Japan.
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
45
V. ORIENTED APPLICATIONS OF WEATHERING STEEL BRIDGE IN VIETNAM:
CASE STUDY OF CHO THUONG BRIDGE
5.1. Description of Cho Thuong bridge and items of the investigation
Three bridges using weathering steel on the N-S railway have been constructed in Viet
Nam (Trung and Hung, 2007). It located between Ha Noi capital and Ho Chi Minh city (fig. 5),
was constructed at the 338 km from Ha Noi (about 25 km away from the sea) in Ha Tinh
Province on May 2000. This is a truss bridge with unpainted weathering steel. This bridge
includes 4 simple spans with a length span of 61 m and effective width 4.26 m. Truss structures
used SMA400AW, SMA400AP and BP. Some locations of structural members have been
painted and unpainted as above rails and under rails respectively (fig. 6). We also installed
exposure test pieces, named “Button-test specimen” to investigate the rust every years. The
items and methods of the investigation show in table 3.
Investigation position Cho Thuong bridge HA TINH Province Railway Fig. 5. Location map of Cho Thuong bridge
Fig. 6. General profile of Cho Thuong bridge
Fig. 7. Actual state of structural members
Fig. 8. Arrangement of the exposure test specimens
5.2. Environment conditions of Ha Tinh province
Cho Thuong bridge belongs to Ha Tinh province and the tropical area. Ha Tinh has two
clearly climate seasons: rainy season from August to November with average rainfall over 2000
mm and dry season from December to July of the next year. The dry season is very hot with hotwind stream from Laos.
46
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
Table 3.The items and methods of the investigation
No.
Items
Methods
Appearance
Visual investigation, photos
Rust condition
Comparison with the standard photos of rust
2
Measurement of rust
thickness
Electromagnetic instrument to measure thickness
3
Thickness loss of steel dueto
corrosion after 1,2,3 years by
Button-test
4
Environmental data: flying
salt, temperature, humidity,
wind velocity, sulfur oxide,
for a year
1
Attach exposed Button-tests pecimens to the bridge
Next time: Take them off after 1,2,3 years
Measure the thickness loss and compare the corrosion
of the specimens with it of the bridge itself
Check exiting data from the meteorological authority
or others
Table 4. Rating of air-born salt in Vietnam (QCVN 02: 2009/BXD)
2
Zones
S mmd (mg/m /day)
1
S >4
2
4≥S>2
3
2≥S>0.5
4
0.5≥S>0.4
5
S≤0.4
Regional characteristics
Marine areas and islands
Coastal
Delta
Region far sea and midland
Midland far sea and mountain
Ha Tinh lies in the tropical and monsoon belt, gives a typical tropical climate in southern
and a cold and dry winter in northern. Annual average rainfall from 2,500 mm to 2,650 mm
occurred in rainy season within heavily rainy period of the late August to mid-November.
Especially, the highest rainfall of 3,000 mm occurred in this period. The lower rainfall from
1,000 to 1,500 mm also occurred in the other areas. Dry season occurs from December to July
of the next year with an intensively sunny period of dry, hot and heavily evaporative SouthWest wind blowing from Laos. The temperature from April to mid-August is very hot with
average temperature of 290 C, reaching 40-410 C on some days. The temperature from October
to March next year is cold with the temperature down to 6-70 C (http://ngoaivuhatinh.gov.vn).
The detailed data referred to the QCVN 02: 2009/BXD (fig. 8). The air-born salt on almost
coastal and in-land region in Vietnam is less than 4 mmd (= 0.04 mdd(mg/dm2/day)) (table 4).
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
47
1000
100
10
1
Jan
Feb
Mar
Apr
May
June
July
Aug
Sep
Oct
Nov
1.5
1.4
1.3
1.3
1.4
1.4
1.6
1.4
1.4
1.9
1.8
1.7
Temperature (C‐degree) 17.6
Evaporate (mm)
40
18.1
20.7
24.3
27.8
29.3
29.7
28.7
26.8
24.3
21.4
18.6
43
57
72
87
115
110
98
65
50
48
42
Moisture (%)
91
92.9
91.8
88.3
81.7
76.8
73.9
79.7
86.5
88.7
88.5
88.2
Sunshine (hours)
77
50
76
137
219
206
233
193
161
135
95
82
Rainfall (mm)
97
64
54
74
143
144
112
225
532
765
319
162
Wind speed (m/s)
Dec
Fig. 9. Statistic average environment conditions in Ha Tinh province (QCVN 02: 2009/BXD)
Rust Grade
Average thickness of rust (μm)
4
90.1
Rust Grade
Average thickness of rust (μm)
3
107
Rust Grade
3
Rust Grade
4
Average thickness of rust (μm) 124
Average thickness of rust (μm) 126
Fig. 10. Rust investigation in span P1~P2 (unpainted) for right side of truss (left fig.)
and left side of truss (right fig.)
5.3. Appearance of rust
The rust investigation of the unpainted steel has shown in Figs. 10 and 11. The rust grade
of 4 is estimated at most members of this bridge. The rust grade of 3 is partially shown on the
surface of the diagonals, where contact to strong wind blows. The rust grade estimation of 3 and
4 show the rust state of this bridge is good condition. The average rust thickness of each surface
is around 80 to 140μm and 110 to 190μm for the grade of 4 and 3, respectively. The rust
thickness does not show the thickness loss of steel as rust expands.
48
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
According to the data measured in Japan, the thickness loss is about 30% of the rust
thickness. Thus, it is thought to be around 20 to 40μm and 30 to 60μm in case grade of 4 and 3,
respectively. This is very good condition. As regards the condition of the painted surface of the
lower members was also good (fig. 11). As the results, the unpainted and painted weathering
steel of Cho Thuong bridge is estimated to good condition.
Rust Grade
Average thickness of rust (μm)
4
82.8
Rust Grade
Average thickness of rust (μm)
3
157
Fig. 11. Rust investigation in span P3~A2 (unpainted elements)
Rust Grade
Good
Rust Grade
Average thickness of rust (μm) 371
Average thickness of rust (μm)
Fig. 12. Rust investigation in span P1~P2 (painted elements)
Good
597
5.4. Corrosion estimation of weathering steel
This bridge was constructed since 12 years ago. Thus, the corrosion estimation of this
bridge is limiting because it is necessary long term estimation on the existing bridge and
exposure button test using attachable small test pieces made of weathering steel. We used shortterm exposure testing method to predict long-term corrosion losses that could occur on the
weathering steel bridges. In this study, the corrosion prediction is performed with limited
investigate results in recent years.
Method of predicting loss of thickness of weathering steel due to corrosion is used in this
study (Kihira et al., 2005; Fijii et al., 2008). The penetration curves of weathering steels can be
expressed as follows:
Y = A XB
(1)
Where X is time in years, Y is the penetration (mm), A is the first year corrosion loss
(mm), and B is the index of corrosion rate diminution.
Based on the results of short-term exposure tests, the loss of steel thickness due to
corrosion of SMA weathering steel in the first year of the test (X = 1 year) is the first evaluated
as the environmental corrosiveness index ASMA. The value of ASMA corresponds to the value of
Y (thickness loss due to corrosion). On the other hand, the value of BSMA corresponds to the
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
49
reciprocal of the degree of decrease in corrosion rate due to an increase in the protective action
of the rust. Using the short-term exposure testing results, we tried to find the value of A and to
estimate the corrosion loss after 100 years by means of estimation procedure as shown in fig.
13.
The mean of the loss prediction result is fairly small of 0.32 mm when the distance from
the coast is about 25 km. However, the long-term exposure test is necessary carried out to
estimate the applicability of this steel type with Vietnam conditions.
After 5 years
Estimated loses in 100 years
Fig. 14. The Cho Thuong raiway bridge, Ha Tinh province [25 km from coastal region] – COR-TEN
ASMA = 0.019, BSMA = 0.62, Y = 0.32 mm, (0.23-0.46), [X=100]
B
VI. CONCLUSION
Weathering steel has been used for many structures including bridges because it’s unique
property of preventing rust by rust. Many successful instances have been witnessed with the
reduction of maintenance and control cost. In this paper, we outlined applicability evaluation for
weathering steel and the advantages of applying them to bridges in Vietnam. Through above
analysis and investigation show applicability of weathering steel in Vietnam as follow:
- Oriented application high-performance steel, and weathering steel is very feasible in
Vietnam condition with a long coast-line, 3260km, and narrow width.
- The use of ability restrains corrosion of it-self is very good in the sparsely populated areas
and less maintenance conditions.
- The air-born salt is not too large (most less than 4 mmd at the coastal and inland in
Vietnam) allows better application of these steel structures in Vietnam.
- The investigation of loss corrosion with SMA weathering steel are using at Cho Thuong
bridge since 12 years ago shown good corrosion protection within local environments. The
prediction results for loss corrosion of this steel type within 100 years also showed nonsignificant. However, it is necessary to observe the course by continuing to conduct exposure
tests and to verify the reliability through collecting the data in long-term exposure tests at some
locates and various environments.
Although the applicability large but also in future need more research on cost, construction
technology and especially, a new design standard in order to creating sustainable developing in
50
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
the bridge construction in Vietnam.
References
[1]. Akira, U., Hiroshi, K., Takashi, K.: 3%-Ni Weathering Steel Plate for Uncoated Bridges at High
Airborne Salt Environment, Nippon steel technical report, No. 87, pp 21-23, 2003.
[2]. Miki, C., Homma, K., Tominaga, T.: High strength and high performance steels and their use in
bridge structures, Journal of Constructional Steel Research, Vol. 58, pp. 3–20, 2002.
[3]. Eiki Y.: Application Potential of Weathering Steel Bridges, Steel construction Today & Tomorrow
(JISF-JSSC), Vol. 15, pp. 11-12, 2006.
[4]. Fujii, Y., Tanaka, M., Kihira, H. and Matusoka, K.: Corrosion risk management methods to realize
long-term durability of weathering steel bridges, Nippon Steel Technical Report, No. 387, 2007.
[5]. Kihira, H., Senuma, T., Tanaka, M., Nishioka, K., Fujii Y., and Sakata, Y.: A corrosion prediction
method for weathering steels, Corrosion Science, Vol. 47, pp. 2377–2390, 2005.
[6]. Japan Road Association: Design Specifications for Highway Bridges, Part II Steel Bridges. Tokyo:
Japan, 2002.
[7]. Japanese Society of Steel Construction, “Possibility and new technologies of weathering steel,” 2006
(in Japanese).
[8]. JSSC: Potentials and New Technologies of Weathering Steel Bridges, JSSC Technical Report No. 73,
2006.
[9]. Isamu, K., Kazuyuki, M., Fumimaru, K.: Minimum Maintenance Steel Plates and Their Application
Technologies for Bridge - Life Cycle Cost Reduction Technologies with Environmental Safeguards for
Preserving Social Infrastructure Assets, JFE Technical report, No. 5, pp. 37-44, 2005.
[10]. Kihira, H., Tanabe, K., Kusunoki, T., Takezawa, H., Yasunami, H., Tanaka, M., Matsuoka, K. and
Harada, Y.: Mathematical modeling to predict long-term corrosion loss to occur on weathering steels, J.
Structural Eng./Earthquake Eng., JSCE, No.780/I-70, pp.71-86, 2005.
[11]. Ohya, M., Takebe, M., Adachi, R., Ota, J. and Kitagawa, N.: Applicability of a New Method for
Selecting Weathering Steel for Bridges, Corrosion in Marine and Saltwater Environments 3 214th ECS
Meeting, Vol. 16 (43) pp. 87-99, 2009.
[12]. Yamashita, M. and Uchida, H.: Recent Research and Development in Solving Atmospheric
Corrosion Problems of Steel Industries in Japan, Hyperfine Interactions 139/140: 153–166, 2002.
[13]. Masaaki, U.: Tokyo Bay Bridge - Long-span Bridge Employing Advanced Bridge Materials and
Technologies, Steel construction Today & Tomorrow (JISF-JSSC), Vol. 18, pp. 1-6, 2007.
[14]. Matsuno, K., Nakayama, M., Fujita, K., Yamamoto, Y. and Oyama, A.: Construction report for
“Binhbridge” in Vietnam, IHI engineering review, Vol. 39, No. 2, pp.47-64, 2006.
[15]. Miki, C., Ichikawa, A., Kusunoki, T. and Kawabata, F.: Proposal of New High Performance Steels
for Bridges (BHS500, BHS700), JSCE, No. 738/I-64, pp.1-10, July 2003.
[16]. Trung, N.V. and Hung, T.V: Characteristics of weathering steel and its application in steel structure
construction in vietnam, Journal of Bridge and Road Engineering, Vietnam, Vol. 3, P. 25-28, 2007.
[17]. QCVN 02: 2009/BXD, Vietnam Building Code - Natural Physical & Climatic Data
for Construction, 2009.
[18]. The Japan Iron and Steel Federation, and the Japan Bridge Association: Application of Weathering
Steel to Bridges, 2003.
[19]. Vietnam Ministry of Transport, Specification for bridge design - 22TCN 272-05, 2005.
[20]. Mathay, W.L.: Highway structures design handbook, American Institute of Steel Construction Inc.,
1993.
[21]. Fujino, Y.: Steel Bridges in Japan - Current Circumstances and Future Tasks, Steel construction
Today & Tomorrow (JISF-JSSC), Vol. 15, pp. 1-3, 2006.
[22]. Yamaguchi, E., Nakamura, S., Hirokado, K., Morita, C., Sonoda, Y., Aso, T., Watanabe, H.,
Yamaguchi, K. and Iwatusbo, K.: Performance of weathering steel in bridges in Kyushu-Yamaguchi
region, Doboku GakkaiRonbunshuu A, JSCE, Vol.62, No.2, pp.243-254, 2006♦
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
51
AN IMPROVED ALGORITHM FOR FLUID-STRUCTURE
INTERACTION OF HIGH-SPEED TRAINS UNDER CROSSWIND
TIAN LI
JIYE ZHANG
WEIHUA ZHANG
Traction Power State Key Laboratory,
Southwest Jiaotong University,
Chengdu 610031, China
Summary: Based on the train-track coupling dynamics and high-speed train
aerodynamics, this paper deals with an improved algorithm for fluid-structure interaction of
high-speed trains. In the algorithm, the data communication between fluid solver and
structure solver is avoided by inserting the program of train-track coupling dynamics into
fluid dynamics program, and the relaxation factor concerning the load boundary of the fluidstructure interface is introduced to improve the fluctuation and convergence of aerodynamic
forces. With this method, the fluid-structure dynamics of a high-speed train are simulated
under the condition that the velocity of crosswind is 13.8 m/s and the train speed is 350 km/h.
When the relaxation factor equals 0.5, the fluctuation of aerodynamic forces is lower and its
convergence is faster than in other cases. The side force and lateral displacement of the head
train are compared between off-line simulation and co-simulation. Simulation results show
that the fluid-structure interaction has a significant influence on the aerodynamics and
attitude of the head train under crosswind conditions. In addition, the security indexes of the
head train worsen after the fluid-structure interaction calculation. Therefore, the fluidstructure interaction calculation is necessary for high-speed trains.
Key words: High-speed train; fluid-structure interaction; crosswind; aerodynamics;
relaxation factor.
I. INTRODUCTION
High-speed transportation is the new direction of modern railway transportation [1]. Highspeed train aerodynamics and train-track coupling dynamics, the indispensable parts of highspeed transportation system, are mutually coupled and influenced. The action of aerodynamic
forces will change the running attitude of the train, and consequently the running attitude will
affect the flow field around train. Thus, the aerodynamic forces of the train will change, and the
train system will be in a particular state of coupling vibration under this mutual feedback.
Strong crosswind seriously affects the running security of high-speed trains. The train
derailments and overturns happen because of strong crosswind [2]-[5]. Many researchers [6][11] have analyzed the running security of trains under crosswinds. There are mainly two
calculation methods of high-speed dynamics under crosswind conditions: off-line simulation
method, and co-simulation method. The former is as follows: the aerodynamic forces of the train
under crosswind conditions are calculated first, and then the dynamic responses of the train are
52
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
calculated with the train dynamic model, on which the static aerodynamic forces are loaded.
This method neglects the change of attitude caused by the action of aerodynamic forces, and
does not reflect the essence of the aerodynamics and train dynamics. In the co-simulation, the
high-speed train aerodynamics and train-track dynamics are calculated alternatively during the
interaction; namely, the interaction effect between the high-speed train aerodynamics and traintrack dynamics is considered. Yang et al. [10] and Cui et al. [11] performed a co-simulation
between high-speed train aerodynamics and train dynamics through parameter transfer and
synchronization control, but they neglected the influence of track structures on the system [11].
In Refs. [10]-[11], the train dynamic solver is in a state of either calculating or waiting, which
consumes too much memory and resources. In addition, the information exchange on the
interface from one time step to the next time step in an alternating fashion can easily cause
energy dissipation.
In this paper, a co-simulation algorithm was improved by (1) inserting the program of
train-track dynamics into the program of computational fluid dynamics to avoid the data
communication between fluid and structure solvers, and (2) introducing the relaxation factor to
improve the information exchange on the interface between aerodynamics and train dynamics.
With the improved algorithm, we conducted a co-simulation between aerodynamics and traintrack dynamics for a train running at a speed of 350 km/h under a crosswind of velocity 13.8 m/s,
and analyzed the fluid-structure dynamics of the high-speed train.
II. GOVERNING EQUATIONS
2.1. Governing equations of fluid dynamics
When the high-speed train is running under crosswind, its flow field can be considered as a
three-dimensional transient viscous turbulent flow. When the running speed is lower than 400
km/h, the flow field around the train can be considered as an incompressible flow. The standard kε two-equation model is adopted and the equations of incompressible flow [9] are written as
∂ ( ρφ )
+ div( ρ uφ ) = div ( Γ grad φ ) + S ,
∂t
(1)
Where t is time, ρ is the air density, u is the velocity vector, φ is flow flux, S is the source
term, and Γ is the diffuse coefficient.
2.2. Equations of train-track coupling dynamics
The train-track coupling dynamics mainly include vehicle dynamics, track dynamics, and
wheel-rail contact. It is assumed that the body, bogies, and wheelsets are rigid and that their
elastic deformations can be neglected. The track system is considered as a continuously
distributed spring-damper model with a two-mass (sleeper and ballast) and three-layer (rail–
sleeper–ballast-bed) structure. The equations of the train-track coupling dynamics [9] are
written as
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
53
+ CX + KX = F ,
MX
(2)
Where M, C, and K are the mass, damp, and stiffness matrixes of train-track system,
respectively; X, X , and X are the generalized displacement, velocity, and acceleration vectors
of the system, respectively; and F is the generalized load vector including the rail excitation and
aerodynamic forces load.
III. IMPROVED ALGORITHM FOR FLUID-STRUCTURE INTERACTION
3.1. Technique for solving train-track dynamics equation
Based on the theory of train-track coupling dynamics, the program of train-track coupling
dynamics is written with FORTRAN and verified to be reliable [9].
The train-track dynamics equations are solved with the Zhai's method [12]. Introducing two
integral parameters and ψ, we construct the new explicit integral format as:
⎧
Δt + ( 1 + ψ)X
Δt 2 - ψX
Δt 2 ,
⎪X n +1 = X n + X
n
n
n-1
2
⎨
⎪X
⎩ n +1 = X n + (1+ φ)X n Δt - φX n -1Δt,
(3)
Where Δt is the integral interval, and subscript n means the time iteration step.
The form of (3) at time t=(n+1)Δt is
+ CX + KX = F
MX
n +1
n +1
n +1
n +1
(4)
.
Substituting (4) into (3), we can calculate X
n +1
3.2. Mesh renewing technique
The spring analogy method [12] and the re-mesh method are adopted to renew the mesh.
When the spring analogy method fails, the re-mesh method is adopted. Spring analogy method
is a simple but efficient method among mesh deforming methods. In this method, each edge of
the grid is modeled as a linear tension spring. The spring stiffness for a given edge i-j is taken to
be inversely proportional to the length of the edge as
K ij =
1
1
=
rij || ri - rj ||
(5)
Where rij is the distance between node i and j, and ri is the position vector of node i.
The displacements of grid points is solved by
Ni
∑ K Δr
ij
j
=0
(6)
j
Where Ni is the total number of the nodes connected with node i, and Δrj is the
displacement of node j. The summations are performed over all the edges of quadrilaterals with
54
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
node i as an end point, i = 1,2,…,n.
The new locations of the nodes are determined by
ri = ri + Δri .
(7)
3.3. Solution strategies
Fig 1 shows the co-simulation method of high-speed train fluid-structure interaction. The
process of the method is described as follows: first, calculate the aerodynamic forces of train
under crosswind until the forces reach a relatively steady state, and then the iteration between
aerodynamics and train-track coupling dynamics begins. At each iteration, the transient
aerodynamic forces are transferred from the fluid solver to the train dynamic model, and the
responses of train dynamics are calculated with this model. Then the attitudes are transferred to
the fluid solver and the aerodynamic forces are calculated under the train attitude. In this
method, the state of train dynamic solver is either calculating or waiting, causing waste of
memory and resources. In addition, the load boundary of aerodynamics and dynamics is very
simple, which may cause energy dissipation.
Fig 2 shows an improved co-simulation method of high-speed train fluid-structure
interaction. The fluid-structure solver includes the aerodynamic solver and train-track coupling
dynamic solver, and the latter is inserted into the aerodynamic solver through interface. Thus,
the data communication between fluid and structure solvers is avoided, and the train-track
coupling dynamic solver does not have to wait when the fluid solver is calculating. The
relaxation factor is introduced for renewing the boundary of aerodynamics and dynamics. The
aerodynamic forces loaded to the train-track dynamics model at time i+1 is no more than the
forces at time i, which is predicted by the aerodynamic forces and their velocities at time i.
It is assumed that the aerodynamic forces fi-1 and fi are calculated by the fluid solver at time
i-1 and i, respectively, and the time interval is Δt. As the Δt is very small, the velocity of
aerodynamic forces at time i is
f − fi −1
fi = i
Δt
Aerodynamics
Tr
in-track
coupling dynamics
0
…
1
Forc
i-1
Attitude
Forces
…
1
(8)
Forces
i
Attitude
i-1
…
i+1
Forces
Attitude
i
…
i+1
Fig 1. Co-simulation method of high-speed train fluid-structure interaction
Aerodynamics
Train-track
coupling dynamics
0
…
1
Forces
Attitude
1
i-1
Forces
i
i+1
Forces
Forces
Attitude
Attitude
i-1
i
…
Attitude
i+1
Fig 2. oved co-simulation method of high-speed train fluid-structure interaction
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
55
60 m
90 m
100 m
175 m
Fig 3. matic of computational domain
Thus, the predicted aerodynamic force at time i+1 is
f i +1 = fi + Δt × fi = 2 fi − fi −1 .
(9)
By introducing the relaxation factor λ, the predicted aerodynamic force fi +1 , which is
loaded to the train-track dynamic model at time i+1, is expressed as follows:
fi +1 = (1 − λ ) fi + λ f i +1 = (1 + λ ) fi − λ f i −1 .
(10)
When λ=0, the aerodynamic force loaded to the train-track dynamic model at time i+1
equals fi. Under this condition, the method is similar to the method in [10]-[11].
IV. SIMULATION
4.1. Computational domain and boundary
In this section, we describe the fluid-structure problem of the high-speed train under
crosswind. The schematic of the computational domain is shown in fig 3. The computational
range is 350 m in length, 90 m in width and 60 m in height. The distance from the inlet
boundary to the nose of the head train is 100 m, and the distance from the outlet boundary to the
nose of the tail train is 175 m. Velocity inlet condition and traction-free condition are preset at
the inlet boundary and outflow boundary, respectively. In addition, a no-slip condition and
symmetry condition are specified as the train surface and top boundaries, respectively. The slip
condition is adapted to the wall boundary.
The train consists of three cars, including head train, middle train and tail train, with the
bulge (such as the pantograph) ignored. The iteration time steps of the fluid dynamic and the
train-track dynamic are 2.0×10-3 s and 5.0×10-5 s, respectively. In this paper, calculations are
carried out for a high-speed train with a speed of 350 km/h and the crosswind velocity of
13.8 m/s; that is, the combination velocity equals 98.2 m/s and the yaw angle is 8.08°.
4.2. Influence of relaxation factor
The load boundary of fluid-structure interface, with different values of the relaxation
factor, may affect the energy dissipation. In this section, three different relaxation factors
(λ=0.0, 0.5 and 1.0) are chosen to analyze the effect.
Fig 4 shows the side force and lift force curves calculated with different relaxation factors.
The change curves of forces are almost the same, except some differences in the amplitudes of
side force and lift force with different relaxation factors. The amplitude of side force becomes
larger with an increase in relaxation factor. However, the opposite is true with lift force. One
can see that there are fluctuations of aerodynamic forces.
56
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
-46
λ=0.0
λ=0.5
λ=1.0
Force (kN)
-48
-50
-52
-54
-56
0
3
6
9
12
15
12
15
Time (s)
(a) Side force
-8
λ=0.0
λ=0.5
λ=1.0
-9
Force (kN)
-10
-11
-12
-13
-14
0
3
6
9
Time (s)
(b) Lift force
Fig 4. Side force and lift force with different λ
In order to analyze the influences of aerodynamic forces on the relaxation factor, the
average standard deviation is introduced for evaluating the fluctuation of aerodynamic forces.
The average standard deviation is described as
S=
fj =
1
5
1 n−2
∑ ( fi − fi ) 2 ,
n − 5 i =3
j+2
∑
j = j −2
f j j = 3, 4,..., n − 2,
(11)
(12)
Where n is the total number of time steps, and fi is the aerodynamic force at time i
obtained by the five-point mean method.
Fig 5 shows the average standard deviation of drag force, side force, lift force, roll
moment, pitch moment and yaw moment with different relaxation factors. When the relaxation
factor equals 0.5, the average standard deviation of force or moment is nearly the minimum in
the three cases. That means the fluctuations of aerodynamic forces are relatively smaller.
In addition, another definition of standard deviation of error is introduced for evaluating the
error between the predicted aerodynamic forces and the calculated aerodynamic forces. The
standard deviation of error is described as
S=
1 n −1
∑ ( fi − fi )2 .
n − 2 i=2
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
(13)
57
150
35
Average standard debiation
30
25
Average standard debiation
λ=0.0
λ=0.5
λ=1.0
20
15
10
5
90
60
30
0
Drag force
Side force
λ=0.0
λ=0.5
λ=1.0
120
0
Lift force
Roll moment
Pitch moment
Yaw moment
(a) Forces
(b) Moments
Fig 5. Standard deviation of aerodynamic forces
Fig 6 shows the standard deviation of error of drag force, side force, lift force, roll moment,
pitch moment and yaw moment with different relaxation factors. Similarly, when relaxation
factor equals 0.5, the standard deviation of error of force or moment is nearly the minimum in
the three cases. From Eq. (13), one knows that the predicted aerodynamic forces are more close
to the calculated aerodynamic forces.
By comparing average standard deviation and standard deviation of error of the
aerodynamic forces calculated with different relaxation factors, one can see that the fluctuation
of aerodynamic forces is small and that the predicted aerodynamic forces are closer to the
calculated aerodynamic forces when the relaxation factor equals 0.5.
350
λ=0.0
λ=0.5
λ=1.0
60
40
20
0
Drag force
Side force
Lift force
300
Standard deviation of error
Standard deviation of error
80
λ=0.0
λ=0.5
λ=1.0
250
200
150
100
50
0
Roll moment
Pitch moment
Yaw moment
(a) Forces
(b) Moments
Fig 6. Standard deviation of error of aerodynamic forces
4.3. Fluid-structure interaction dynamics
Fig. 7 shows the pressure distribution of a high-speed train cross-section. The cross-section
is 14 m away from the nose of the head train. We can see that, under the interaction effect, the
pressure of the train windward side becomes larger and the pressure on the lee side becomes
smaller. In addition, the pressure of the train bottom becomes larger because it has turned into
lee side after the roll of the train body.
The side force comparison of the head train is shown in Fig. 8. The side force calculated
with off-line simulation is a steady value; however, the one with co-simulation shows a
fluctuant curve. After considering the interaction effect, the side force becomes larger than
before. The side force is closely related to the pressure difference between the lee side and the
windward side. According to the analysis of pressure distribution around the train cross-section,
as shown in fig. 8, the pressure difference becomes larger after taking the change of train
58
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
attitude into account. Thus, the side force calculated with co-simulation is over 10% more than
that with off-line simulation.
(b) Co-simulation
Fig 7. The pressure distribution of cross-section
(a) Off-line simulation
The train-track coupling dynamic responses to the aerodynamic forces are calculated.
There are some differences in the dynamic responses because of the differences of forces and
moments. Fig 9 shows the lateral displacement comparison of the head train. After we take the
interaction effect into account, the lateral displacement of the head train toward the lee side is
larger than before. However, the variations of lateral displacement are similar as the lateral
displacement mainly depends on the lateral track irregulity.
Tab 1 shows the comparison of train security indexes calculated by different methods.
After considering the interaction effect, the security indexes, including the wheel/rail vertical
force, lateral wheelset force, deraiment, and wheel unloading, become larger.
-40
100
Co-simulation
Off-line simulation
80
Displacement (mm)
Force (kN)
-45
-50
-55
-60
60
40
Co-simulation
Off-line simulation
20
0
3
6
9
12
0
15
0
3
6
9
12
15
Time (s)
Time (s)
Fig 8. Side force comparison of head train
Fig 9. Lateral displacement comparison of head train
Tab 1. Comparison of train security indexes
Method
Wheel/rail vertical force
(kN)
Lateral wheelset
force (kN)
Derailment
Wheel unloading
Off-line
94.69
28.79
0.31
0.63
98.92
32.85
0.35
0.78
Co-simulation
This means that the fluid-structure interaction cannot be neglected in the calculation of
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
59
train security.
V. CONCLUSION
In this paper, an improved algorithm of high-speed train fluid-structure interaction under
crosswind is presented. First, data communication of the fluid solver and structure solver is
avoided by inserting the program of the train-track coupling dynamics into the fluid dynamics
program; second, the load boundary of the fluid-structure interface is improved by introducing
the relaxation factor. For the velocity of crosswind of 13.8 m/s and a running speed of train at
350 km/h, the aerodynamic forces and attitude of the head train are compared via the off-line
simulation and the co-simulation. The comparison shows that the fluid-structure interaction has
a significant influence on the head aerodynamics and attitude, and the security indexes become
larger in the fluid-structure interaction simulation. Thus, the fluid-structure interaction
calculation is necessary for high-speed trains under crosswind conditions.
ACKNOWLEDGEMENTS
The research was supported by the National Natural Science Foundations of China
(Nos.50821063 and 50823004), 973 Program (No.2007CB714701) and the Fundamental
Research Funds for the Central Universities (No.2010XS34).
References
[1]. X.B. Li, Z. Yang, J.Y. Zhang, W.H. Zhang, Aerodynamics properties of high-speed train in strong
wind, Journal of Traffic and Transportation Engineering, 2009, 9(2): 66-73 (in Chinese).
[2]. H.Q. Tian, Study development of train aerodynamics in China, Journal of Traffic and Transportation
Engineering, 2006, 6(1): 1-9 (in Chinese).
[3]. M. Suzuki, K. Tanemoto, T. Maeda, Aerodynamic characteristics of train/vehicles under cross winds,
Journal of Wind Engineering and Industrial Aerodynamics, 2003, 91: 209-218.
[4]. C.J. Baker, J. Jones, F. Lopez-calleja, et al, Measurements of the cross wind forces on trains, Journal
of Wind Engineering and Industrial Aerodynamics, 2004, 92(7): 547-563.
[5]. S.G. Tan, X.B. Li, Z. Yang, J.Y. Zhang, et al, The flow field structure and the aerodynamic
performance of high speed trains running on embankment under cross wind, Rolling Stock, 2008, 46(8):
4-8 (in Chinese).
[6]. B. Dierichs, M. Sima, A. Orellano, et al. Crosswind stability of a high-speed train on a high
embankment, Journal of Rail of Rapid Transit, 2007, 221(2): 205-225.
[7]. Y.G. Wang, K. Chen, Effects of crosswinds on curve negotiation of high-speed power cars, Journal of
Southwest Jiaotong University, 2005, 40(2): 224-227 (in Chinese).
[8]. Y. Song, Z.S. Ren, Research on dynamics performance of high speed trains under strong lateral wind,
Rolling Stock, 2006, 44(10): 4-7 (in Chinese).
[9]. T. Li, J.Y. Zhang, W.H. Zhang, Performance of vehicle-track coupling dynamics under crosswinds,
In: Proceeding of Railway Motor Vehicle Dynamic Simulation, Chengdu, 2010: 317-322 (in Chinese).
[10]. J.Z. Yang, H.Q. Bi, W.M. Zhai, Dynamic analysis of train in cross-winds with the arbitrary
Lagrangian-Eulerian Method, Journal of the China Railway Society, 2009, 31(2): 120-124 (in Chinese).
[11]. T. Cui, W.H. Zhang, Study on safety of train in side wind with changing attitudes, Journal of the
China Railway Society, 2010, 32(5): 25-29 (in Chinese).
[12]. W.M. Zhai, Two simple fast integration methods for large-scale dynamic problems in engineering,
International Journal for Numerical Methods in Engineering, 1996, 39(24): 4199-4214♦
60
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
NEW APPROACHES TO THE CHOICE OF ARCHITECTURE
FOR A SUPERVISORY SYSTEM OF GAZ TRANSPORTATION
BOGDANOV N.K
ZAMYTSKIKHY P.V
KHADEEV A.S
State Technical University – MADI,
Moscow, Russia
Summary: The paper presents the advantages of centralized processing of operational
information technology systems in supervisory control of gas transport. We consider the
technical aspects of data transmission between the equipment automation of gas mains,
central control point (DP) and DP slave branches. The article is published with the financial
support of the Government of the Russian Federation (Russian Ministry of Education) as part
of the project under the Contract № 13.G25.31.0064 on October 22, 2010.
Keywords: Dispatch and transport the gas control system, a hierarchical approach, data
transfer protocols.
Beginning in 2010, JSC "AtlanticTransgasSystem" participates in a number of projects to
create supervisory systems transport natural gas, which created or redeveloped technical
infrastructure automation and communication, as well as articulated by the customers job to
cause formation of a new design trend in architecture dispatching systems. The traditional
approach to building a control system of gas transportation companies across the region was a
hierarchical approach "bottom up" when the control towers (DP) of individual industrial
branches (linear production controls gas pipelines-Linepipe Operation Center) was concentrated
to a maximum of information technology as the linear part of MG, compressor shops, gas
distribution stations, etc., as well as data from all support systems-electrical, electro-chemical
protection, communications and security. Further, each of DP higher level of management
receiving a subset of the technological information available to the DP slave, plus a set of
aggregate parameters and data reporting manual input [1]. This concept has several advantages,
including:
1. The creation of enterprise control system could be implemented in stages over several
years at the expense of major repairs.
2. Control systems not subject to special requirements for the communication channelsData collection was carried out with grassroots systems over dedicated copper cables within the
prom.
The site; with remote control systems-anti-interference, with the use of protocols [2].
Failure to transfer data to the upper level was not considered as critical as branch manager is
still in full control of all process and support systems in their area of responsibility and could
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
61
report to the central control room over the phone. However, the construction of systems "bottom
up" has its drawbacks, as it is impossible to ignore the fact of providing facilities MG,
especially newly constructed, in principle more reliable and more than three orders of high speed fiber-optic lines (FOCL). It features fiber-optic, as well as the investment nature of most
of today's development projects and reconstruction of the supervisory system can be considered
an alternative to the traditional layered architecture.
The concept of a centralized architecture
Centralized supervisory control (see fig 1) provides the following direction of information flow.
Fig 1. Information flowscontrol and management
1. Process data from the controllers installed at the sites of the linear part of the MG (DCS
and RTU-unit, as well as ACS GDS) go directly to the CDP process control. Thus the data are
collected as a process of gas transport, as well as state support systems.
2. Manager of branches observe the state of objects in the MG and its neighboring areas of
responsibility, using the actual remote terminal access control system CDP. They can also
generate reports for yourself - using a common portal. In Fig. 1 within the systems of automatic
control, of course, comes to the control operator of the shop. However, some technological
information, which characterizes the condition of the equipment in time, goes to the CDP
process control.- Data from the gas metering stations, similar to the compressor-nym shops,
arrive simultaneously at a local point of data collection and control system in the CDP.
3. Commands closures Manager submits CDP controlling MG in general. Staff offices shall be
notified. Teams that require changes in the operational conditions are fed to the level of branches in
the form of traffic control tasks. Achieving the targets set automatically controlled by the CDP.
4. Reporting data on supply, use and loss of gas in the area of responsibility of the branch can
be formed automatically in the CDP process control, but must be confirmed by the responsible
person (manager) of the branch, which can also, if you have the justification to adjust.
What are the benefits and risks is a central passage? First, the decline in the number of
program - technical complex control system. This eliminates the problem of PCS service
62
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
branches, which are often on the ground there are no qualified personnel. Disclaimer of data
storage and processing at the branch level and eliminates the need to synchronize both content
and structure of object databases, SCADA - systems - the basic elements of control systems.
And if synchronization valuesin principle not a problem (only time - consuming configuration
information Interface), the automatic or even automatic synchronization of database structures is
almost never realized, that leaves room for defects associated with the "human factor".
Secondly, the centralization of facilities allows for a more professional management of their
operation - place in the overall data center (DPC) of the organization and use the same highly
reliable means of electricity, air conditioning, etc. It becomes easier to apply uniform security
policy, both in the field of access to equipment/data, and reduce the vulnerability of the system for example, to conduct regular backups. Thirdly, the centralized processing of information
makes it easier to apply the same algorithms to model data sets of different branches. Thus, if
the planning phase costs of FER in the coming period costing process needs its own gas by a
group of centralized payment, the owner of a detailed plan for transport/distribution of gas, and
the actual values, for example, made of the commodity transport, or loss of gas can be
calculated in accordance with the approved method of price-decentralized.
Access Managers of branches to a centralized system will allow them to also see the results of
the simulation MG, often carried out for all the gas company as a whole and has a much smaller sense
for the plot area of responsibility of a separate branch. Direct access to the DCS/ RTU - CDP devices
of the control system also allows you to implement software control of MG and leak detection.
Centralize the management of risks associated with possible failures of hardware and
software, and communication channels are rarely evaluated with a formal precision. Reduction
in computer technology allows almost every project, set the task of creating redundant systems,
with 100% redundancy of servers, workstations, network equipment and cable lines within
buildings (areas). For dispatch systems across the region and also back above control towers,
geographically remote from the main. This, as well as duplication of data channels through the
use of satellite channels, let’s talk about disaster recovery systems established, large-scale
failure are possible, but again only because of the "human factor".
Another objection to the centralization of information is 100% of the impossibility of one
man-Manager CDP effectively taken a wealth of information and adequately reacts to it. To
ensure basic ergonomics PCS CDP - screen form, which works manager must contain, of
course, only the most important parameters, and some generalized symptoms, on the other hand,
when an abnormal situation. Manager will receive all of the possible amount of information on
specific object opening the appropriate detailed mimic. The transfer of information we consider
separately the two technical aspects of information interactions in a centralized architecture: a
unified communication protocol that is ideal for centralized control system, and the mechanism
of dispatching jobs. Any person professionally connected with the industrial automation, sign,
or at least heard about the technology of communication OPC, have received very widespread in
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
63
the world and in Russia over the past 15 years. In principle, the transfer of air traffic control
tasks can be viewed as some internal company e-mail, with the following features:
- For each task should be fixed status, in which it is currently (created, approved, the work
performed, etc.), and time of each assignment status;
- A job can contain both text-only instructions and a link to a formal setting, a target which
is to be achieved (see fig 2).
- One supervisory task can be associated with others.
Fig 2. Targetsettingin the controltask
CONCLUSION
Considered by the example of the "center-branch" approach can be more successfully
applied in the case of three or more levels of supervisory control-remote access to a centralized
system, engineers can get a replacement unit, operating without the linear sections MY COP, or
managers of individual pipelines. The availability of reliable and high-speed communication
channels, modern communication technology, supported by specialized software tools supervisory
control, open up the possibility of increasing the reliability and convenience of the new systems
under tight control of their budgets as at the implementation stage, and subsequent maintenance.
References
[1]. Berner, LI, Bogdanov, NK, Kovalev AA, Integrated automation solutions for gas transportation and
gas companies of OAO "Gazprom" / / Gas Industry. № 7, 2007. - C. 38-43.Two.
[2]. Zeldin, M., Kovalev AA, Implementation of joint information in the automation of dispatching management
(for example, projects for OAO "Gazprom") / / Industr ial process control and controllers. № 6, 2008.Three.
[3]. BogdanovA.N., KiselevA.OPC, Unified Architecture: changes in the popular technology information
exchange with the engineer's point of view / / Modern automation technology. № 3, 2010. - C. 82-87♦
64
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
IMPROVEMENT QUALITY OF CONGESTION CONTROLLER
IN ATM NETWORK BY METHOD USING NEURAL NETWORK
TRAN XUAN TRUONG
Telecommunication Engineering Department,
UTC
Summary: This paper presents method of congestion control using neurons at the
switching nodes in the ATM network to improve quality of controlled system. Neural network
have a important special role in the prediction the congestion status for the controller, which
based on that it can determine the limited rate at the moment of the switches. This is
foundation to control the growing traffic source on the network. The experimental simulations
have demonstrated the effectiveness of this method when compared with conventional
methods.
I. INTRODUCTION
As you known, the ATM network is a network based on cell switching technology (small
and fixed length packets, 53 byte) [2]. This is the information network that can provide multiple
services, speed from a few hundred Kbps to Mbps. On the other hand, the ATM Forum has five
current services for ATM networks but only available bit rate ABR and UBR services is
designed to manage the application busrt type data. ABR service is trusted to manage better the
data flow, because it allows the switching monitoring network behavior and feedback of
relevant information by RM cells to the source. The source will adjust its data rate based on the
received information. The task of this traffic management is necessary, because this is the
essence problem for network control and distribution efficiency, negotiation about quality of
service QoS. Congestion control is one of the most important task in the traffic management. It
is a dynamic problem that can not be resolved by conventional static solutions. Typically, only
bit explicit forward congestion indication EFCI is used for congestion control purposes. In
addition, studies have shown that explicit rate ER can provide better and faster quality, the
researchers gave different directions in computing the explicit rate ER as NIST ER [4] and
EPRCA [6] techniques. However, these techniques also present limitations when applied in the
switching nodes. This is reason we offer solutions to solve network congestions effectively by
neurons.
This paper presented a congestion control solution for available bit rate services (ABR) in
the ATM network using Neural network. This is a congestion control method indicated explicit
rate (ER) by Neural network, which uses Artificial Neural Network to determine the explicit
rate for the ABR service. ER is defined as maximum allowed cell rate for source in the next
cycle. Neural networks used in this technique will monitor the buffer queue status of the switch
and make the suitable ER value carried by resource management cells (RM) to source for
adjustment their speed. We perform comparing the proposed algorithm and some known
algorithms based on process system modeling and simulation. Since we can see the advantages
and disadvantages of this intelligent control method.
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
65
II. CONGESTION CONTROL SOLUTION BY METHOD OF THE NEURAL
INDICATION EXPLICIT RATE IN ATM NETWORK
The proposed switching congestion control is explicit rate technique indicated by neurons.
The main purpose in this design is made Neural Network to create congestion control process
simply but effectively. The ordinary congestion control studies can not meet control quality in
ATM networks. Several other researchers have proposed the use of Neural Network [5], [7], but
they focused on the using Neural networks to forecast traffic patterns. So this section, the author
gives a complete algorithm used Neural network to monitor network status and determine how
much bandwidth can be seized for.
As mentioned, this technique comply with the specification by the ATM forum, so basic
operation is similar to the other switching diagram (see fig. 1). The main difference is way that
how the explicit rate values measuring made are. Neural Network will monitor the queue status
in the switching buffer for each fixed time period (N). After N received ABR cells, Neural
Network should value suitable ER based on the reading information about the queue length for
the current and the next time period. When a backward resource management cell (BRM) go to
the switch, the calculated values are compared with ER values in BRM cells. If the calculated
value is smaller than the current value of ER, the ER field in BRM will be updated. Conversely,
if the calculated value is greater than the current value of ER, the ER field in BRM cell will be
guaranteed.
Our algorithm has two inputs required to bring the neural network. They are the current and
pass queue length. Two input are essential in providing sufficient information about the status of
the neural network. Because, from the information about current queue status, Neural network
can learn queue occupancy and the level of congestion. Besides, Neural network compares the
current queue length with the previous queue length to get information about the buffer change
rate. This is a good measured value to known the network congestion ability. If the network is
not congestion, this is a good indicator for the network to achieve the completely application of
network resources.
FRM
Source
BRM
Switch
ATM
Queue
FRM
BRM
Switch
ATM
Queue
Neural network
congestion
control
Neural network
congestion
control
ANN
ANN
FRM
Dentination
BRM
Fig.1. The model of the proposed Neural network control technique
66
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
The Neural network output is the rate adjustment factor (RCF) which can be used by the
virtual circuit (VC) based on current traffic conditions. The value of this coefficient ranges from
-1 to 1; In there, the value is less than 0 displays the network facing congestion or have the
potential to be network congestion, so the source has to transmit at rate lower than the current
cell rate. Otherwise the RCF value is greater than 0 displays all links that can be used
completely, and the speed of the source can be increased to match the traffic of the suffer
network. With the RCF provided by the network neurons, explicit rate ER for a new cycle is
calculated by:
ERnext = Min(Link Rate, max(0, ERcurrent + RCF x Link Rate))
(1)
All switches will ensure an value of ER and updated in each N received ABR cells. For
each cell passing through switches, the switch examines the ER field of BRM cells. If the ER
field value of BRM cell is greater than explicit value that switches ensure, the ER field in BRM
cell is updated to ER ensured value of the switch. If the value in the BRM cell is smaller than do
not perform any action.
For behaviour at the source, they will adjust its data rate based on the value of this ER with
the help of congestion indication bit (CI) and no increase indication bit (NI) in the same cell.
The allowed cell rate of the source in each control cycle is defined based on the bit values of NI
and CI in the resource management cell BRM. When the NI bit is not set (NI = 0), the are
allowed cell rate at the source is defined in value:
⎡ACR= min(ER, ACR+ RIFx PCR,PCR) if CI = 0
⎢ACR= min(ER, ACR - RDFx ACR)
if CI = 1
⎣
(2)
When NI was established (NI = 1), the allowed cell rate at the source is determined by:
⎡ACR = min(ER, ACR)
⎢ACR = min(ER, ACR - RDF x ACR)
⎣
III. CONSTRUCTION NEURAL
CONGESTION CONTROLLER
NETWORK
USING
if CI = 0
if CI = 1
IN
THE
(3)
PROPOSED
As pointed out in Section 2, the neural network is a very important role in predicting the
network congestion level based on the measure the switch queue value between two consecutive
time. On that basis, at a time, Neural network will provide a ER value to switches through the
RCF from the neural network output. Therefore, the development of Neural network to
undertake its role in the congestion control is essential.
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
67
Inputs
Class 1
p1(k)
IW1,1
1
6
6x1
n2(k)
LW2,1
6x1
6x1
IW1,2
1
a1(k)
n1(k)
6x1
p2(k)
Class 2
a2(k)
12x1
12x6
12x1
b2
12
12x1
b1
6x1
a1(k) = logsig(IW1,1. p1(k) + IW1,2. p2(k) + b1)
a2(k) = logsig(LW2,1. a1(k) + b2)
Class 3
a2(k)
12x1
1
LW3,2
5x12
Class 4
a3(k)
n3(k)
5x1
5x1
1
b3
5x1
5
a3(k) = tansig(LW3,2. a2(k) +b3)
LW4,3
n4(k)
1x5
b4
1x1
y(k)
1x1
Output
1
y(k) = purelin(LW4,3. a3(k) +b4)
Fig. 2. The structure of neural network used in the study
Based on the Neural network knowledge, the design skills to build the network neurons, we
have built a network neurons completely to apply in the proposed congestion control problem.
Neural network are designed that has two inputs for receiving the corresponding value of the
queue length at the previous and the present time, separated by a control cycle. The distance
between that time was the distance of two cell RM. The obtained neural network output
parameter is ratio RCF to help switches decide it’s explicit rate ER at next time.
Neural network architecture consists of four layers and described in Fig.2, in which the
input layer (layer 1) has 6 neurons; Layer 2 has 12 neurons; layer 3 consists of 5 neurons and
the output class (class 4) has 1 neuron. The neurons in the first and second layers used the
transfer function as logsig, neurons in layer 3 used the transfer function as tansig, the remaining
neuron in the output layer used transfer function as purelin form. The set of weight matrix
including a threshold, class 2 and class 3 weight matrix (net.b {1}, {2} net.b, net.b {3}, {4}
net.b), input weight of the two inputs to the network’s first layer (net.IW {1.1}; net.IW {1, 2}),
connections weight between layers (net.LW {2.1}, net . IW {3.2}, {4.3} net.LW) are given
detail in [3].
68
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
IV. BUILDING THE SIMULATION MODEL TO VERIFY THE PROPOSED
METHODOLOGY
DABR-A
SABR-A
DABR-B
SABR-B
S1
S2
Link 155Mbps
DABR-C
SABR-C
SABR-D
SCBR
DABR-D
DCBR
Fig. 3. Configure an ATM network using simulated and compared
With the Neural network has been designed in Section 3 and the algorithm presented in
Section 2 above, we have used broadband communication network ATM as shown in Fig.3 to
perform this simulation. The network consists of four virtual channels (VC) bring available bit
rate service (ABR), called group A, B, C and D. Besides, it contains four VC carrying the
service constant bit rate (CBR) named group X. The network has two switching nodes and are
connected by 155 Mbps links. The ABR source is set at the rate of 31 Mbps, 93 Mbps, 155
Mbps, 217 Mbps and 279Mbps respectively while the CBR source is fixed at the rate of 31
Mbps for all simulated cases. The CBR source has main function to occupy a bandwidth certain
amount in the link and also to observe its impact to the network configuration. For switches,
they have time slot of 155 Mbps with buffer size of 900 cells. There are two reasons for
choosing the small buffer size. First of all, this is the design trend to keep the switching buffer
size to be smallest. In addition, the smaller buffer size can make us observed congestion control
scheme when the source has a minimum rate entirely possible.
The link between broadband terminal equipment (BTE) and switches is set to 2 km with
rate of 155 Mbps. The links between the switches is also 155 Mbps but the distance between
them is 10 km. Simulation time for each run is 500 ms, while the received cells per receiver and
the lost cell number in each switch are recorded. In this simulation, only the measure about
quality of service (QoS) is observed, which is characteristic of the control rate range and queue
in the switch node with the neural network controller. Since we can calculate the throughput of
the switches.
Besides the proposed explicit rate neural network congestion control algorithm, the
simulation is repeated using the algorithm that is accepted by the ATM Forum, which is a
Enhanced Proportional Rate Control Algorithm (EPRCA) [6].
V. THE OBTAINED RESULTS FROM SIMULATION AND DISCUSSION
To obtain the necessary results from the simulation system, we undertook to write the
simulation program based on Matlab software version 7.8 in 2009. The traffic type can be
selected as the source with Poisson distribution or two-state Markov distribution, the other
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
69
S
o
u
rceR
a
te[M
b
p
s]
hypotheses as shown in section 4.
To conduct a survey of control rate range for resources, we have obtained the results
presented in figures fig.4, fig.5 and fig.6. The results showed that control method by Neural
network has source control rate range wider and larger than the conventional control methods.
The rate range of proposed method has values from 5Mbps up to 220 Mbps, whereas the
conventional method only from 10 Mbps to 200 Mbps. This parameter also reflects partly the
high sensitivity of the control method using Neural network. This result due to the congestion
predictability quite well of the neural network in controller, the more accurate forecast, the more
accurate value of ER in the switches, which the ABR sources the can adjust their rate to the
network accurately based on these reference rate.
250
225
200
175
150
125
100
75
50
25
0
0
50
100
150
200
250
300
Time [ms]
350
400
450
500
S
o
u
rc
eR
a
te[M
b
p
s
]
Fig. 4. A range of control rate in method using Neural network
250
225
200
175
150
125
100
75
50
25
0
0
50
100
150
200
250
300
Time [ms ]
350
400
450
500
Fig. 5. A range of control rate in the conventional method
Neural Control Method
250
Conventional Method
225
SourceRate[M
bps]
200
175
150
125
100
75
50
25
0
0
50
100
150
200
250
300
Time [ms]
350
400
450
500
Fig. 6. Compare a range of control rate in two methods
The queue properties of the switching nodes using neural control technique as fig.7 shows
that the buffer level is often consumed at high levels but concentrated in the safety of queue
overflow problem and the queue’s cycle characteristics. This feature demonstrates, the ability
using network resources to be effective and therefore the switches throughput will increase
significantly when compared with other control methods. At the same time it can avoid the cell
loss phenomenon due to full buffer queue, so it will contribute to reducing the rate of cell loss
compared to many other control methods. Meanwhile, for the conventional control method, the
variation of queue length always permanents in a dangerous level, in which the switch will
overflow buffer and the throughput ability in the buffer also be slowly. This decreases the
switch throughput and be easy to cause the cell loss due to the full queue, which affects directly
the quality of service of the applications.
70
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
900
800
Q
u
e
u
eL
e
n
g
th[c
e
lls
]
700
600
500
400
300
Neural Control Method
200
Conv entional Method
100
0
0
50
100
150
200
250
300
Time [ms]
350
400
450
500
Fig. 7. Characteristics of queue in the switch under different control techniques
320
T
h
r
o
u
g
h
p
u
t[
M
p
b
s
]
300
280
260
240
Neural Control Method
Conv entional Method
220
200
0
31
93
155
Sourc e Rate [Mpbs ]
217
279
Fig. 8. Throughput of switches in the control methods
For the throughput of the switches, we can observe the results obtained from the fig.8. The
switching throughput of the methods changed according to the rate, the Neural network control
method achieves high throughput (300 Mbps), meanwhile, conventional control scheme as
EPRCA has lower level (about 275 Mbps). This property also shows the traffic congestion
control ability be better with conventional methods, because the throughput is the total traffic to
be served by the switching node so the traffic control as efficiently the traffic through the
switches as many, also means higher throughput. Furthermore this will avoid the network
collapse phenomenon (a throughput be reduced in minimum, negligible value) due to the traffic
retransmitted too much from the sources because the sender - receiver process be failed.
VI. CONCLUSION
In this paper, we have implemented the congestion control method using Neural network
applied for the ATM network. It is demonstrated that the proposed control algorithm is better
when compared to the scheme switching using conventional control methods, because it has the
quality of service better. It also has the simplicity and efficiency, because the simulations have
shown that the proposed algorithm has high network throughput, the high link utilization and
low cell loss rate. To have an expected result because the neural network can predict accurately
about congestion in the high volatility and unpredictable traffic environment of ATM network.
References
[1]. Bùi Công Cường, Nguyễn Doãn Phước, Hệ mờ mạng nơ ron & ứng dụng, Nhà xuất bản Khoa học kỹ thuật,
2001.
[2]. Nguyễn Hữu Thanh, Tổng quan về kỹ thuật mạng B-ISDN, Nhà xuất bản Khoa học Kỹ thuật, 2007.
[3]. Trần Xuân Trường, “Nghiên cứu giải pháp chống nghẽn trong mạng thông tin băng rộng ATM sử dụng
mạng nơ ron”, Đề tài nghiên cứu khoa học cấp Bộ, nghiệm thu năm 2010.
[4]. Golmie, N., Chang, Y và Su. D, “ NIST ER Switch Mechanism”, ATM Forum/95-695.
[5]. N. J. H. Kotze and C. K. Pauw (1997), “The use of Neural Networks in ATM”, Proceeding of the
Communications and Signal Processing, COMSIG '97, South African, Page(s): 115-119.
[6]. L. Robert (1994), Enhanced PRCA (Proportional Rate Control Algorithm), ATM forum 94-0735 R1,
August 1994.
[7]. A. Tarraf, I. HabiB and T. Saadawi (1995), “Congestion Control Mechanism for ATM Networks Using
Neural Networks”, IEEE International Conference on Communications, ICC '95 Seattle, Volume: 1 Page(s):
206-210♦
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
71
ENVIRONMENTAL INDICATORS FOR SUSTANABLE
DEVELOPMENT URBAN TRANSPORT PLANNING IN VIETNAM
MSC. VU KIM HUNG
Institute of Transport Planning and Management
University of Transport and Communications
Summary: The research aims at selecting environmental indicators for the planning of
sustainable development of urban transport in Vietnam. The author used the methodology of
systematic analysis, methods of literature review on environmental indicators in the
development of urban transport in the world and DPSIR model to assess the necesity to
consider the environmental indicators in the process of urban transport planning. The article
also described the selection of the 13 environmental indicator based on 4 criteria for
protection of natural resources and environment.
Key words: Environmental indicator; environmental criteria, urban transport planning.
I. INTRODUCTIONS
In general, big cities in Vietnam are facing with challenges in developing urban transport.
Problems of urban transport has become particularly severe, in particular, traffic jams are
becoming more frequent and extended over almost the city. Next is the phenomenon
of flooding due to poor sewerage during the heavy and long-lasting rains. On the other hand, the
increase means the number of private property led to the demand for fossil fuel.
Increasing greenhouse gases and air environment, dust, noise, vibration due to operation of
urban transport is
becoming
increasingly serious,
threatened ecosystems, urban
green space narrowing leads to loss of ecological balance.
In fact, urban planning and urban transport planning have been developped, but there is no
standard for developping the plan. Almost planning of urban transport development are lack
of sustainability, lack of strategic vision and un-friendly with the environment.
Law of Environment in 2005 has already required to assess the environmental impact
strategies for strategic projects, master plans and planning of transportation. However, this
work still has a lot of inadequacy in the coordination between the transport planning and
environmental impact assessment strategies.
The above-mentioned problems show the necessity to incorporate environmental
indicators into the development process of urban transport which must be taken into
consideration from the planning stage. By the law of Vietnam shall all urban transport
projects be taken into construction only after being approved in the steps for urban
transportation planning.
The objective of this study is to identify and select indicators of environmental conditions
suitable for urban transport development in Vietnam towards the goal of sustainable
development.
However, in this study, the author do not quantify the selection indicators and general
72
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
assessment of interaction with the target of economic
Indicators
and social indicators.
In order to see the relationship between
environmental indicator with the objective of urban
Criteria
transport development,
the
environmental
indicator for the urban transport planning is defined in
the tree of criteria identifying (see figure above).
Objective
Definitions
Sustainable Development in the report of the
United Nation Commission on Environment and
Vision
Development (United Nations, 1987) [2] is defined as
follows:
"Sustainable
development means development that meets current needs but without prejudice to the ability to
meet the needs of future generations".
Urban Transport is a system including the interaction of such three subjects as
transport infrastructure (roads, transit pointa, bus stops/transitions), transportation and traffic
control systems in order to the satisfy the mobility demand of human (passenger x km) and
freight (ton x km) [1].
Indicators of environmental sustainability in urban transport has been studied in the
world such as planning policy for urban transport development of the European Union [3]; In
the report of the Ministry of Environment and Energy, USA about “Indicators and performance
measures for Transportation, Environment and Sustainability in North America”[4]; In the
conference of OECD about Environmentally Sustanable Trasport (EST) in Vienna “Futures,
Strategies and Best practice” [5]; And in a Guidebook for Performance-Based Transportation
Planning of transportation research Board (TRB) [8].
II. METHODOLOGY
This study is carried out based on the following methodology:
Author used DPSIR model to assess the status of the urban environment in Vietnam. The
author selected urban transport in Hanoi as the object of study and assessment, suggesting
that environmental indicator should be considered during the planning process of
urban transport development.
DPSIR model inherited by the
European Environment
Agency (EEA) is
one
of
the basic frame, chain facilities for general information, using the indicator with the different
categories (driving force - Pressure -state of Environment - Impact - Response)
(UNEP / RIVM, 1994) [6]; (RIVM / UNEP,1995) [7]. This
model
is similar
to the
model PSR framework (OECD, 1993 Organization for Economic Cooperation Development)
but with two (D) item is the driving force and effect (I). DPSIR is the first letter of the year from
the English language:
Driving Forces
the local environment are
transport development.
being
(D) is the general motivation factors are affecting
considered. For
example, sustainable urban
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
73
Pressure (P): The pressure of dynamic factors on the
environment and alter
the current
status
of
environmental pollution. For
example: Increased
emissions, dust, noise, vibration, increased greenhouse
gas,…
P
S
D
State (S): The state of the environment at a
time or time period. For example, the current stateof air
(SO2, NOx, TSP, VOC), noise,
vibration,
loss
of biodiversity ....
R
I
Impact (I): The impact of negative or positive to the economic situation, social and human
health. Example: Due to increased pollution of traffic leading to increased incidence of
disease, ...
Response (R): integrated solutions to improve environmental quality.
From the study of environmental indicator in the world and the lack of environmental
indicators in transport development planning of Hanoi. Author identified and selected
appropriate environmental indicators for the planning process of urban transport development in
Vietnam.
III. RESULTS OF RESEARCH
Through the study authors go to some of the results as follows:
Environment indicator is understood as a functional description of environmental issues
and resource development planning in urban traffic, the data base to be able to conclude the
planning impacts the quality the environment.
Sustainable urban transport environment means developing an urban transport system to
meet the transportation needs of the market and people travel now and in the future, saving land
use restrictions energy consumption and reduce emissions and waste within the limits
of absorption of the medium, with the mode of travelaffordable alternative to reduce traffic
congestion.
It is easy to see the lack of system of environmental indicators in urban transport
development planning processes. The author used the DPSIR model for analysis and evaluation.
Driving force factors selected by the author is to develop sustainable urban transport
and environment is
expressed through
the secondary driving
force is
to
develop
infrastructure, vehicle development and traffic control. From The Driving force of urban
transport growth above will cause Pressure on the environment such as the land use, energy
use and emissions in transport, traffic congestion…. leading to State environmental quality,
expressed through changes of parameters such as SO2, NOx, VOC, TSP, PM10, Pb and noise,
greenhouse
gas
and
landscape. The
Impact
of environmental
pollution are
analyzed through the economic
damage
and
social impact
on public
health.
With the impact that requires,
Response is the integrated
solution
to improve
the environmental quality of
urban
transport cause, that
is solutions and
policies to
achieve development objectives of urban traffic and minimize impacts environment, is shown in
the image below.
74
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
Urban transport development
- Infrastructure development
- Vehicle development
- Traffic control
Driving Force (D)
- Land use
- Energy
- Emission and congestion
Pressure (P)
- Transport emission
- Greenhouse gases
- Landscape
- Land use for transport policy
- Air Quality Management
- Energy quality management
- Transport demand management
Response (R)
State (S)
- Public health
- Economy damage
Impact (I)
Through the DPSIR model shows that environmantal indicators are important toward the
urban
transport system for
sustainable
development. Starting from
the
perspective (vision) for the development of each country, urban or community that is
sustainable development, defined in the Brundtland report in 1987 at the United Nations as:
"development to meet present needs without compromising the ability to meet the needs
of future generations".
To generalize the sustainable development of urban transport can be understood as:
“urban transport development to meet current trafic demands without compromising the
ability to meet the transport needs of future generations”.
Regarding the demand for urban transport in the sustainable development of urban
areas there are many ways define different but come from a "model for the metabolism of the
settlements of man (human stetllement's metabolism model), Newman and Kenworthy (1999)
suggested that the path reaching a state of sustainable development, each city needs
to achieve three objectives: 1- there is adeveloping economy, efficiency and environment
friendly, 2 - have a rich culture, harmony and charming, and 3 - have a transportation system for
sustainable urban transport [1].
Inherit the results of different studies, the authors [1] has compiled an objective
framework for sustainable development of urban transport system consists of 4 levels. The
highest
level is
the vision
of
a sustainable urban development, such
as the
Brundtland concept, followed
by the objectives of
sustainable urban
development, as
defined by Newman and Kenworthy (1999). Three is the level of the concept of an
urban transport system's sustainability and
members Albert Speer (1993) is a
transportation system to achieve the following objectives: 1 - Traffic smooth, 2-traffic safety, 3 environmental transportation friendly, 4 - transport to promote economic development.
Number 4 levels are 13 criterias to be achieved to have a transportation system for sustainable
development including 4 criteria for traffic movement; 2 criteria ensure traffic safety; 4 criteria
for protecting natural resources and emvironment and 3 criteria for improving economy of the
city and region.
Based on the target framework above, in terms of the environment in urban
transport shall be four criteria on environment and natural resources of concern are: 1INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
75
reduction air emissions criteria; 2- reduction fuel and energy use in transport ; 3-criteria for land
use for urban transport and 4- criteria to save space and landscape. Based on the criterias
above, the authors selected 13 environmental indicator for the sustainable development of urban
transportin as shown in the picture below:
Sustanable Urban Development
Leve l 1: Vision
Sustanable Environment
To achieve a strong and compact
Economy within urban area
To ensure mobility for all
transport demand
Economically Viable&
Efficient
To achieve a Sustanable Urban
Transport System
To ensure safety of all
transport movements
To ensure equality in using
transport properties
To reduce accident
frequency
To increase the number of
modal choises
To reduce accident
severity
Socially compatible
To achieve a rich, harmonise and
compact culture within urban area
To protect natural resources
and Environment
To reduce air pollutions
To reduce tatal
transport cost
To reduce traffic noise
To increase economic
productivity and
efficiency
To increase productivity
and efficiency of transport
supply
To save energy
consumption in transport
To increase capacity of
transport supply
To save space consumption
for transport
(1)-Emission per vehicle
(ton/vehicle)
(2)-Emission per each route
(ton/km)
(3)-Tons of greenhouse
gases generated
(4)-The level of noise
(dB/km)
(5)-Fine dust (ton PM10/km)
Ghi chú:
(6)-Fuel consumption per
VMT (km/vehicle)
(7)- Fuel consumption per
PMT (km/passage)
(8)- Fuel consumption per
freight (km/ton goods)
[1];
To improve economy of the
city and region
(9)-Percent of urban
land change in urban
transport (%)
(10)- percent of trees in
street (%)
Level 2: Objective for
Sustanable Urban
Development
Level 3: Objective for
Sustanable Urban
transport
Development
Level 4: Criteria for
Sustanable Urban
transport
Development
To improve economic
attractiveness
(11)-Percent of change
in Public
transportation per
other mode (km/km).
(12)-Percent of vehicles
using alternative fules
(%).
(13)-Percent use of
non-motorized modes
per all trips (%)
Level 5:
Indicator for
Sustanable Urban
transport
Development
Author's choice
1. Environmental criteria related to traffic emissions: The objective of this criterion is to
reduce emissions on vehicles and on roads. To achieve this criterion we need to control
and quantify the emissions of each of the planning compared with the current state plan and is
expressed through the indicator of (1) (2); (3) (4) (5).
76
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
2. Environmental criteria related to reduced fuel use in transport and energy: The
objective of this criterion is rational use of energy in the transport of passengers and
freight. Thus, for transport planning of urban, we need to identify targets for the rate of fuel
consumption per unit of transport is made or tons of kilometers of each mode of transport urban
centers. Only
when
the energy
cost per
unit
of urban transport is
reasonable pollution reduction can be expressed through the indicator of (6) (7) (8).
3. Environmental criteria related to rational use of urban land and environmental
landscape, is reflected by the indicator of (9) (10).
4. Environmental criteria related to alternative modes of travel: The objective of this
criterion reduced the number of personal motor vehicles, increase the percentage of public
transport modes and non-motorized modes that
will make reduce emissions
from transportation activities as
well
as reduce energy demand in the transport of
passengers and freight, expressed through the indicator of (11), (12) and (13).
Above are 13 environmental indicator selected by authors towards an urban transport
system of
environmental sustainability. These
indicators can
be reviewed
and
evaluated directly or indirectly in the process of urban transport planning.
IV. CONCLUSION
The author has provided with the definition of sustainable urban transport in environment.
By DPSIR model approach, the author has identified environmental problems caused
by transport development in urban in big cities of Vietnam, indicating that regardless
of the environmental indicator from step urban transport planning . From general research in the
world, author have chosen 13 environmental indicators based on 4 general criteria
for sustainable development and environmental resources.
References
[1]. Khuat Viet Hung (2006), Traffic Management in Motorcycle Dependent Cities, Doctoral Dissertation,
Darmstadt University of Technology.
[2]. United Nations (1987), Report of world commission on environment and development, Development and
International Economic Co-operation, Brundtland.
[3]. Commission of the European Communities (CEC) (1995), The Citizens Network. European Commission
Green Paper (COM 95), Brussels.
[4]. German Marshall Fund Fellowship - GMF (2000), Indicators and performance measures for Transportation,
Environment and Sustainability in North America, Ministry of Environment and Energy, USA.
[5]. OECD (2000), Futures, Strategies and Best practice. OECD Conference Environmentally Sustanable
Trasport (EST), Vienna.
[6]. RIVM/UNEP (1995), Scanning the global environment: A framework and methodology for UNEP's
reporting functions. UNEP Environment Assessment Technical Report 95-01, Nairobi, Kenya.
[7]. UNEP/RIVM (1994), An Overview of Environmental Indicators: State of the art and perspectives. UNEP
Environment Assessment Technical Report 94-01, Nairobi, Kenya.
[8]. TRB (2000), A Guidebook for Performance-Based Transportation Planning. National Academy Press
Washington, D.C. USA♦
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
77
ALGORITHM OF VIRTUAL TRAINING COMPLEX DESIGNING
FOR PERSONNEL RETRAINING ON PETROCHEMICAL
ENTERPRISE
BARINOV K. A
KRASNYANSKIY M. N
MALAMUT A. J
OSTROUKH A. V
State Technical University – MADI,
Moscow, Russia
Summary: The algorithm of design of a virtual simulator that is not tied to a specific
discipline is developed. We propose a sequence of development stages of a virtual simulator in
accordance with the proposed algorithm, in order to increase productivity and ensure high
quality simulator created.
This work was supported by the Government of the Russian Federation (Russian Ministry
of Education) as part of the project under the Contract № 13.G25.31.0064 on October 22,
2010.
Keywords: Virtual simulator, algorithm, training.
In the study of most subjects it is necessary to support indissoluble connection between
theoretical analysis and experimental research. In the traditional method of training
experimental research can be conducted in normal laboratory conditions [2]. Current rate of
development of distance education has put before the developers the problem of unification of
virtual simulators (VS) creation methods. Solution to the problem is possible with common
algorithm for the creation of computer simulator that is not focused on any particular discipline.
Modern computer technologies allow automating the process of theoretical knowledge
gaining, as it has no significant difficulties in implementing the corresponding software. More
difficult in terms of automation is the process of practical skills development [2], as well as
learning new methodologies used in the study.
Virtual simulator (VS) designing
For the practical skills gaining virtual simulators can be applied. Simulators can be created
with the help of specialized computer systems (constructors) or by a combination of various
software tools. Currently, constructors of computer simulators, as independent software packages,
allow you to create simulators almost of any subject from scratch, are not widespread. Either way,
regardless of a method of simulator creation to be used, it is necessary to add here some algorithm
of virtual simulator designing to automate the process and try to consider maximum number of
components and featuresin it [3], [4].
In a number of works attempts to develop and formalize the algorithm of virtual simulator
78
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
designinghave been made, but they were focused only on a small number of industries: nuclear,
petroleum, railway and electrical engineering. Also, besides this disadvantage - arrow focus, it
should be noted that there was no multi ser mode provided in a considered number of works.
Among other restrictions,an orientation of virtual simulators on operators retraining can be
mentioned. In this work we propose extended, yet universal (not tied to any particular
discipline) scheme of the algorithm, which eliminates the above disadvantages (partially). The
scheme of the algorithm of designing of a virtual simulator for operators of chemical and
technological systemswas taken as the base [1].
The process of computer simulator designing, which is an intelligent system itself, requires
several tasks to be solved for providing following characteristics of a virtual simulator:
•
High degree of similarity between model and real object;
•
Possibility of working out the specified actions;
•
Ensuring high efficiency for short duration training courses;
•
To support for multi-user mode;
•
To meet common psychological and pedagogical requirements;
• Short time of adaptation during the transition of human activities from the simulator to
the real object;
•
Opportunity to make updates and changes without additional costs.
The algorithm of a virtual simulator designing can be represented as a functional diagram
in IDEF0 notation.
Sequence of development stages of VS
The inverse decomposition allows us to represent considered algorithm stepwise.
Fig 2. Development stages of VS
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
79
Fig 1. Functional diagram of the algorithm of a virtual simulator designing
The first stageis analysis of the subject area. The main steps are:
• Reproduction of complete process flow chart for simulator being created (all that
involved in the experiment is to be modeled according to its characteristics; it is a
fundamentally important moment, since a completely modeled process allows correctly and
fully to work out the required tasks and gain skills);
• Forming of suggestions concerning information technologies and methods of
simulator implementation to be used;
•
Discussion of acommon interface, design and style of simulator.
The second stage is modeling of structure of VS. The main steps are:
• Development of mathematical models for all devices and processes in accordance with
the technological scheme (the creationof a mathematical model should start from its creation for
the ideal process);
•
80
Refinement and accordanceof actions to perform work for each component (module)
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
of simulator;
•
Programming of application logic;
• Bringing the training materials, needed for a particular experiment, in accordance with
the requirements,developed on the first stage, to its formalized description.
The third stage is setting of VS. The main steps are:
• Introduction the required perturbing factors into the mathematical model, if necessary,
to adoptthe model to the real process, considering differential acceleration of the particular
processes, as long as for the observation and execution of several processes, depending on the
discipline, considerable time resource is required (precisely adjusted mathematical model
increases the efficiency of training due to its identity to the real process);
•
Designing ofthe tools for training control and management.
The fourth stage is packaging and testing of VS. The main steps are:
•
Stepwise treatment of formalized material;
•
Connection of the modules in a single complex;
•
Testing and validation of execution logic;
•
Adjustmentof the simulator for its future exploitation.
Conclusion
Simulator designing with observance of division of processes and stages makes it possible
to significantly reduce development time, to improve the quality and reliability, to simplify the
process of maintenance, updating and support of virtual simulators. Especially whena virtual
simulator is created by a team, such approach gives the opportunity to adapt already developed
technology (algorithm) for creating simulators on other disciplines.
References
[1]. V.A. Nemtinov, S.V. Karpushkin, V.G. Mokrozub [i dr.]. Metody i algoritmysozdaniyavirtual'nyhmodeleihimiko-tehnologicheskihsistem: monografiya/M-voobr. inauki RF, GOUVPO «Tamb. gos. tehn.
un-t». Tambov: Izdatel'skiidom TGU im. G.R. Derjavina, 2011. 282 s.
[2]. Dmitriev V.M., Gandja T.V.. Zadachipostroeniya i konfiguraciyakomp'yuternyhtrenajerov //
Distancionnoeobrazovanie, innovacii i konkurentosposobnost': Materialyregional'noinauchnometodicheskoikonferencii. –Tomsk: TGU sistemupravleniya i radioelektroniki, 2004 – S. 85-86.
[3]. S.K. Gupta, D.K. Anand, J. Brough, M. Schwartz, and R. Kavetsky. Training in Virtual Environments.
A Safe,сost-effective, and engaging approach to training. University of Maryland. 2008.
[4]. Robert J. Seidel, Paul R. Chatelier, Virtual Reality, Training's Future? Perspectives on Virtual
Reality and Related Emerging Technologies. New York. 1997. 232 s♦
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
81
NI DOPING EFFECTS YBA2FE3O8+W
XIAOYU GUAN
YONG ZHAO
XIAOQIU JIA
Southwest Jiaotong University,
Chengdu 610031, China
Summary: By doping Ni into YBa2Fe3O8+w (YBFO) system, we obtained the phase
YBa2Fe3-xNixO8+w (YBFNO, x=0, 0.05, 0.10, 0.15, 0.30, 0.50, 1.00). This paper discusses
the changes in crystal structural, resistivity and magnetoresistivity (MR) of YBFO samples due
to the incorporation of transition metal Ni. The results show that Ni substitution for partial Fe
in YBFO does not substantially transform the structure of parent phase, but results in tiny
changes in the lattice parameters. The YBFO crystal with Ni doped is semiconducting.
Key words: YBa2Fe3O8+w; Ni doping; crystal structure; magnetoresistivity.
I. INTRODUCTION
Compound YBa2Fe3O8+w (YBFO) is an antiferromagnet (AFM) with Neel temperatures
(TN) around 660 K [1]-[2]. The crystal structure of YBFO is triple perovskite as it is formed by
merging Y and Ba, and by disordering of the oxygen atoms and vacancies. The oxygen
stoichiometry 8+w varies in a narrow range of −0.2＜w＜+0.1 [3]. The crystal is the only
analogue of YBa2Cu3O7 (YBCO) where Cu is fully replaced by another transition metal. There
is a wide miscibility gap in two phases as iron is unable to adopt a square-planar coordination.
Fe3+ is stabilized in the environment of one octahedral and two square-pyramidal per unit cell,
where YBFO composition is generated. YBFO and YBCO differ in their arrangements and the
number of oxygen atoms on the basal plane of their unit cell. When w>0, chemical analysis
conducted to locate the oxygen in excess of the eight atoms per formula unit showed that the
sites available for such oxygen were those in the Y layer at 0, 0, 1/2, assigned as O(4) [4]. The
structure diversification of introducing Fe into YBCO has been widely studied in an attempt to
detect possible correlations between structural modifications due to the presence of iron and the
lack of superconductivity.
Ref [4] discussed the nucleus and magnetic structure of YBFO, samples of which were
prepared from liquid-mixed citrate precursors. The cooperative magnetic structure for all
compositions was based on a larger cell related to the nuclear cell by the transformation matrix
[1,−1,0/1,1,0/0,0,2] [4]. The tetragonal crystal structure of YBa2Fe3O8 became orthorhombic
once oxygen vacancies were included into the iron coordination octahedron [4]. The
antiferromagnetic ordering of the iron magnetic moments was below 630–670 K, indicating that
the variation in oxygen content was too small to significantly affect the magnetic exchange
interactions.
82
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
Huang et al. [5] investigated the accommodation of Co in the oxygen-saturated solidsolution phase YBa2(Fe1-zCoz)3O8+w using techniques of powder X-ray and neutron diffraction,
and 57Fe Mossbauer spectroscopy. In the nominal composition range 0.00≤z≤1.00, the solidsolution limit under syntheses 950℃ in 1 bar O2 was z=0.47(5) [5]. No symmetric change in the
nuclear and magnetic structures was observed due to the Co substitution. The Co atoms were
distributed evenly over the two sites, which were square-pyramidally and octahedrally
coordinated for w=0. The value of w was viewed as nearly constant and Co adopted valence
close to 3.00.
At present, the research on doping transition metals into Fe position in YBFO system, such
as Ni and Mn, has not been reported. This paper investigates the Ni doping effects on crystal
structure, resistivity and magnetoresistivity (MR) of YBFO system.
II. EXPERIMENTAL PROGRAM
Conventional sol-gel method was followed to prepare samples YBa2Fe3-xNixO8+w (x=0,
0.05, 0.1, 0.15, 0.3, 0.5, 1, for short YBFNO).
Step 1: Weighed high-purity stoichiometric Y2O3, Ba(CH3COO)2, Ni(CH3COO)24H2O and
Fe powders.
Step 2: Raw materials were uniformly and slowly dissolved into acetic solution with
continuous stirring during a gradually heating process. The heating process continued until the
solution became dry.
Step 3: Grinded the xerogel for 60 min to gain the well proportioned powder and then
pressed the powder into tablets.
Step 4: Sintered at 1 100 °C in the atmosphere for 350 h and cooled in the furnace.
1 100 oC
o
100 C/h
100 h
6h
500 oC
Cool with furnace
50 h
Atmosphere
100 oC/h
Room temperature
Time (h)
1 100 oC
Temperature (oC)
Temperature (oC)
To improve the sample homogeneity, intermediate grinding and palletizing must be
ensured. The sintering processes are shown in Fig. 1. A Philips X’Pert MRD diffractometer with
Cu-Ka radiation recorded the θ-2θ XRD patterns, which characterized the phase purity of the
series samples. Each sample was in single phase with tetragonal structure. The resistancetemperature curves were obtained by PPMS (physical property measurement system).
100 oC/h
100 h
Cool with furnace
Atmosphere
Room temperature
Time (h)
(a) The first fire
b) The second and third fire
Fig 1. Heat processes of YBFNO samples
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
83
103
Ba2Fe2O5
BaFe2O4
(1) Y2BaNiO5
(2)
NiO
226
109
310
106
310
116
213
(2)
210
(2)
(2)
006
020
005 113
100
Intensity (counts)
x=1
111
112
(1)
(1)
x=0.5
O
x=0.3
x=0.15
x=0.1
x=0.05
O
O
x=0
20
30
40
Fig 2.
50
60
2 patterns
(°)
XRD
of YBFNO
70
80
III. RESULTS AND DISCUSSION
3.1. XRD test and analysis
Typical XRD θ-2θ patterns for YBFNO are shown in fig 2. Similarities between spectra of
YBa2Fe3-xNixO8+w (x=0.05, 0.1, 0.15, 0.3, 0.5, 1) and YBFO (x=0) reveal that incorporating a
small amount of Ni into the YBFO lattice will not substantially transform tetragonal structure of
parent phase since only tiny changes of the lattice parameters are observed. In fig 2, the
intensity of the (100) peaks gradually become smaller, meaning that the tendency towards a-axis
preferential orientation is weakened as Ni content increases. Impurities peaks of Ba2Fe2O5 and
BaFe2O4 can occasionally be detected, indicating that they are close to YBFO in phase graph.
Only a few impurities peaks of Ni oxides are found. This is because most of Ni is incorporated
into octahedral or square-pyramidal environments in YBFO lattice in the substitution for Fe. In
addition, in samples of x＞0.3, the typical peaks intensity of objective products are not
prominent and the impurity peaks such as Ba2Fe2O5, BaFe2O4, Y2BaNiO5 and NiO become
stronger. When x≤0.3, the intensity of characteristic peaks in YBFNO spectra dominates even
though there are a few negligible impurities. As a result, we presume that the solid-solution
limit of Ni in YBFO is in the vicinity of x=0.3 under the condition of adequate oxygen
atmosphere.
Peaks for a series of samples of YBFNO in fig 3 shift slightly to the right so as to conform a
strong dependence of peak position on the content and radius of the Ni2+ incorporated into YBFO
system.
84
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
x=1.00
Intensity (counts)
x=0.50
x=0.30
x=0.15
x=0.10
x=0.05
x=0.00
46.0
2θ (°)
46.5
47.0
Fig 3. The enlarged diffraction peaks (006) and (200) in YBFNO
For instance, the lattice parameters of YBFNO gradually become smaller as Ni content
increases. However, if Ni enters in YBFO crystal and does not substitute Fe, the lattice
parameters of YBFNO will be larger than those of the parent phase, and the new crystal will
expand. Thus, we are convinced with the partial substitution of Fe by Ni in octahedral and
square-pyramidal coordinations of YBFO crystal. The lattice parameters change because the
radius of Ni ion is smaller than that of Fe. And the oxygen content in the unit cell is nearly a
constant when oxygen is adequate.
Tab 1 lists the values of the lattice parameters versus Ni content within its solid-solution
limit. These data are obtained by a least-squares fit to a large number of the Bragg diffraction.
Tab 1. The calculated unit cell parameters of YBFNO
Ni-content
Unit cell parameter on the x
axis
Unit cell parameter on the z
axis
Volume of unit cell
0.00
3.916 45
11.813 68
181.20
0.05
3.916 38
11.813 58
181.19
0.10
3.913 02
11.802 57
180.72
0.15
3.910 62
11.795 40
180.39
0.30
3.910 20
11.793 44
180.32
3.2. Resistance analysis and MR of YBFNO
Fig 4 shows curves in logarithmic scale between resistance and temperature of YBFNO
samples in 0T and 1T magnetic fields. Samples of YBFNO in the two magnetic fields remain
semiconducting like YBFO. It was reported that the resistivity function of LaBa2FeO8.3 (150
K＜T＜300 K) acted as Arrhenius-type behavior with an activation energy of about 0.2 eV [6].
Elzubair et al. [7] formulated an evident variable-range hopping (VRH) process:
ρ = ρ∞ exp (T0 T )
14
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
(1)
85
Where T0 is characteristic temperature related to the density of state at the Fermi level. The
VRH process was suitable for describing the resistivity of RBa2Fe3O8+w (R=La, Nd, Sm, Eu, Gd)
when measurement temperature range was increased to 620 K [8]. We find that the VRH
process also fits for YBFO when measurement range was 10–300 K, and there was a direct link
between the resistivity and temperature. Comparing fig 4(b) with fig 4(a), we notice limited
effects of application field on the resistivity of YBFNO.
To compare the density of states at the Fermi level, we investigated the relationship
between the resistivity and the Ni content in YBFNO at 300 K. The curves in fig 5, as a whole,
are very similar to two curves in 0 T and 1 T magnetic fields at 300 K, except for x=0.15. It is
shown that except for x=0.15 at 300 K, the application field has limited effects on the resistivity.
Moreover, when Ni content is 0, 0.15, or 1, the resistivity of samples in 0 T and 1 T fields at
300 K is noticeably different from other samples, as is probably related to the onset of spatial
charge or magnetic ordering.
Fig 6 shows the relationship between the resistance and the temperature for different Nicontent samples in 1 T field. With the decrease in temperature, the crystals experience
Insulator-Metal (IM) phase transition.
20
20
15
x=0.00
x=0.05
x=0.10
x=0.15
x=0.30
x=0.50
x=1.00
10
5
0
256 K
0.25
123.46 K
LogeR (Ω)
LogeR (Ω)
15
10
5
66.64 K
0.30
256 K
0
0.35
0.25
123.46 K
0.30
x=0.00
x=0.05
x=0.10
x=0.15
x=0.30
x=0.50
x=1.00
66.64 K
0.35
T -1/4 (K-1/4)
T -1/4 (K-1/4)
(b)
(a)
Fig 4. Representative logarithmic resistance of YBa2Fe3-xNixO8+w in 0 T (a) and 1 T (b) fields.
The continuous line is a fit to a VRH process. For convenience, only several lines were given
20
(×103Ω·cm)
16
0T
1T
12
8
4
0
0.0
0.2
0.4
0.6
0.8
1.0
x
Fig 5. The relationship between the resistivity and Ni content in YBFNO at 300 K
86
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
The temperatures of this phase transition TIM (nearly 110 K) are much lower than the Neel
transition temperature, TN (nearly 660 K for pure YBFO). This is because TIM is closely relevant
to the factors such as synthesizing conditions, crystal boundary, and magnet ordering.
Moreover, the oxygen content in crystal also plays a role in the low TIM.
6
600
4
x=0.00
x=0.05
x=0.10
x=0.15
x=0.30
x=0.50
x=1.00
2
MR (%)
R (×108Ω)
x=0.00
x=0.05
x=0.10
x=0.15
x=0.30
x=0.50
x=1.00
400
200
0
−200
0
100
200
300
100
T (K)
200
300
T (K)
Fig 6. The relationship between the resistance
and the temperature of YBFNO in 1 T field
Fig 7. The relationship between the MR and the
temperature of YBFNO samples in 0 T field
Fig 7 shows the relationship between the magnetoresistivity (MR) and the temperature. MR
is obtained by
ρ = ρ∞ exp (T0 T )
14
(2)
Where ρ 0 and ρ H are resistivities in 0 T and 1 T fields, respectively.
The different intensity peaks corresponding to x=0.1, 0.15, 0.3, and 0.5 appear within a lowtemperature range of 90–130 K and a high-temperature range of 240–300 K.
This phenomenon can be attributed to the effects of Colossal Magneto resistance (CMR).
The maximum positive MR value at x=0.1 reaches nearly 600%. Other CMR peaks in the lowtemperature range are flat and small, because of the phase transition between ParamagneticMetal (PM) and Ferromagnetic-Insulator (FI), instead of the one between PM and
Ferromagnetic-Metal (FM). Impurities such as Ba2Fe2O5 and BaFe2O4 are rarely seen in the
synthesized products. In addition, phase transitions probably exist between PM and FM at
x=0, 0.05, and 1 within the both temperature ranges.
IV. ENDING REMARKS
We used the wet chemical method to prepare YBFNO samples. A small amount of Ni does
not tremendously transform the space structure of YBFO system. The main reason for the
decrease of lattice parameters is that Ni has smaller ionic radius than Fe. In YBFNO, Ni can
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
87
arbitrarily substitute partial Fe in the parent phase. YBFNO samples exhibit the semiconducting
feature like YBFO. The VRH process is suitable to describe the resistivity of YBFNO samples.
ACKNOWLEDGEMENTS
The work was supported by Funds from Center University of the Education Ministry
(No. SWJTU09ZT24).
References
[1]. M.E. Massalami, A. Elzubair, H. Ibrahim, et al., Structural and magnetic properties of YBa2Fe3O7+x Febased 1-2-3 ceramic oxide, Physica C, 1991, 183(1-3): 143-148.
[2]. Q. Huang, P. Karen, V.L. Karen, et al., Neutron-powder-diffraction study of the nuclear and
magnetic structures of YBa2Fe3O8 at room temperature, Phys. Rev. B: Condensed Matter, 1992, 45(17):
9611-9619.
[3]. Q. Huang, P. Karen, V.L. Karen, et al., Neutron-powder-diffraction study of the nuclear and magnetic
structures of the substitution compound (Y1-xCax)Ba2Fe3O8+y (x=0.05,0.10,0.20), Phys. Rev. B: Condensed
Matter, 1994, 49(5): 3465-3472.
[4]. P. Karen, A. Kjekshus, Q. Huang, et al., Neutron powder diffraction study of nuclear and magnetic
structures of oxidized and reduced YBa2Fe3O8+w, Journal of Solid State Chemistry, 2003, 174(1): 87-95.
[5]. Q.Z. Huang, V.L. Karen, A. Santoro, et al. , Substitution of Co3+ in YBa2Fe3O8, Journal of Solid State
Chemistry, 2003, 172(1): 73-80.
[6]. A. Elzubair, M. ElMassalami, Thermally activated electronic hopping and oxygen nonstoichiometry in
the Perovskite LaBa2Fe3Ox, Physica B, 1996, 225(1-2): 53-62.
[7]. N.F. Mott, E.A. Davis, Electronic Processes in Non-Crystalline Materials, Oxford: Clarendon Press, 1971.
[8]. A. Elzubair, M.E Massalami, P.H. Domingues, On the structure and magnetic properties of the series
RBa2Fe3O8+x (R=La,Nd,Sm,Gd), Physica B, 1999, 271(1-4): 284-293♦
88
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
ECONOMIC COMPETITIVE EVALUATION
OF HIGH-SPEED LINES
PH.D. KURENKOV P.V
Department "Transport business" MIIT
POLYAEVA T.I
Department "Economics and Logistics in the transport"
SamGUPS
Motion with speeds above 250 km/h at the moment is a priority for the strategic
development of rail transport from 23 countries worldwide. High-Speed Lines (HSL), which are
considered globally competitive in relation to air transport for distances up to 700 km, in last
years have tended to the length with increasing speed. To Russia, even where in the European
part of the way a distance between cities with a population of five hundred thousand to half a
million people are at the length of 700 km the problem of building and recovery of such
"extended" HSL has particular actual.
The first HSL in the world had a length of 570 km (Tokyo - Osaka), and in Europe, 470 km
(Paris - Lyon). At these days, the longest in the world of HSL is constructed in China between
Beijing and Shanghai (1,320 km), the presumed value of $ 32 billion. High-speed passenger
traffic (HST) on its scheduled non-stop, because of what will be achieved a high-speed route. At
the moment, night trains on this route also followed without any interruption. Primarily the
demographic situation serves in China. Beijing (19.7 million people) and Shanghai (24.6 million
people) are the largest metropolitan China. Accordingly, the passenger traffic flow between the
Chinese capitals greater than that between the Russian cities.
Table 1. Dynamics of carriage passengers by HSL
in Europe (1994 - 2004) according to the UIC/3, p. 31/
Year
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
France
Pass-km Growth
(millions)
(%)
21,9
21,4
- 2,3
24,8
15,9
27,2
9,7
30,6
12,5
32,2
5,2
34,7
7,8
37,4
7,8
39,
6,7
39,6
-0,8
41,5
4,9
Germany
Pass-km Growth
(millions)
(%)
8,2
8,7
6,1
8,9
2,3
9,3
4,5
10,2
9,7
11,6
13,7
13,9
19,8
15,5
11,5
15,3
-1,3
17,5
14,4
19,6
12,0
Italy
Pass-km
Growth
(millions)
(%)
0,8
1,1
37,5
1,3
18,2
2,4
84,6
3,6
50,0
4,4
22,2
5,1
15,9
6,8
33,3
7,1
4,4
7,4
4,7
7,9
6,6
Spain
Pass-km
Growth
(millions)
(%)
0,9
1,2
33,3
1,3
8,3
1,5
15,4
1,5
0
1,7
13,3
2,2
29,4
2,4
9,1
2,5
4,2
2,5
0
2,8
9,9
Rest of Europe
Pass-km
Growth
(millions)
(%)
0,3
0,4
33,3
1,4
250,0
2
42,9
2,7
35,0
2,8
3,7
3,5
25,0
3,8
8,6
4
5,3
4,1
2,5
4,1
0
Europe
Pass-km
Growth
(millions)
(%)
32,1
32,8
2,2
37,7
19,9
42,4
12,5
48,6
14,6
52,7
8,4
59,4
12,7
65,9
10,9
68,8
4,4
71,1
3,4
75,9
6,8
In France, a prerequisite for the construction of SCM also served as a congestion of railway
line Paris - Lyon (410 km). Design of the line, called the South-East, was completed by
1981. Reduced travel time (about 3 hours) caused an unprecedented increase in the number of
passengers carried. It was a commercial success that has inspired many countries to expand or
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
89
built a high-speed rail network.
High-speed rail is more profitable than usual, not counting the cost of constructing
infrastructure. Reason - the fact that a lot of operating costs such as a staff have a fixed cost per
hour, while the income from ticket based on the distance. Passengers also pay more for a high
speed per mile. Thus, the ratio of operational revenue/cost more for high-speed systems.
Fig 1. The cost of construction of 1 km of the new HSL /3, p.25/
At the moment we have an obvious contest in the development of high technology between
the European and Asian countries. And if previously the undisputed leader in this area were the
Europeans, in the long term for the year 2025 in Asia will be built - 21 452 km of HSL, while in
Europe will be built 17 565 km high-speed lines. Major Trends in construction and operation of
BCM in the world are displayed in fig. 1.
Obvious results the acceptance actively implemented introduce in the context of
continental China program for the construction of high-speed lines. As of 2007, the ShenyangQinhuangdao line pro ¬ pulling over 400 km of secondary and technical train speed was 197.1
km/h. From ¬ covering August 2, 2008 ¬ HST Beijing-Tianjin line length of 117 km, calculated
at a maximum speed of 350 km/h, reduced the time the train between the two cities from 70 to
30 minutes. The highest value recorded ¬ Noah technical speed to the mainland in 2009
amounted to 236 km/h. This was achieved in the family rides CRH, developing a maximum
speed of over 300 km/h / 4 /.
14000
12000
10000
8000
planned lines
6000
lines under construction
4000
operating lines
2000
0
China Taiwan India
Iran
Japan Saudi North Turkey
Arabia Korea
Fig 2. Planned, built and operated by HSL Asia km /5/
90
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
At the moment, the longest in the world of SCM has built in China between Beijing and
Shanghai (1,320 km), the presumed value of $ 32 billion.
6000
5000
4000
3000
2000
planned lines
lines under construction
1000
operating lines
0
Fig 3. Planned, built and operated by HSL Europe, km
Currently in Europe the process of connecting high-speed systems of different countries
into one network TEN-T, which is part of the sustainable spatial development of the European
continent. The situation here is exactly the opposite of China. HSL connects densely populated
areas, such as Corridor № 6 MAF Lyon (1,5 million people) - Turin (0.9 million people) Trieste (0.2 million people) - Ljubljana (0.3 million people) - Budapest (1, 7 million) in length
about 1500 km. The average length of high-speed lines in Europe is 300 km, which a priori
ensures competitiveness and a high return on rail transport. The average distance between the
cities of one million people in Europe is 200 km. Therefore, competition between operators of
high-speed lines and airlines on intra-European market low-kosters rather difficult.
Now in Europe passengers preference high-speed trains near 60 to 85% of. And, for
example, on the lines of the Paris-Brussels-Cologne and Frankfurt are completely displaced
airlines. True, the success of HSL, not least ensured by an intensive development of intermodal
transport, where the airline and railroad companies are moving from direct competition to
cooperation, giving passengers the opportunity to travel on one ticket. In this case, the mediumand long-haul travel by air require high-speed traffic on the initial and final sections of the path
to deliver passengers to the airport and back.
Many operators do not think that more speed achievement is main. In most cases, an
increase of frequency of trains is more important in terms of attractiveness for passages as, for
example, in the UK.
Currently, there are a lot of the concepts of "high-speed traffic" (IRR) on a rolling stock,
and on infrastructure (road, signaling, centralization and blocking, electrification and power
supply, etc.), as well as on the compatibility of rolling stock and the infrastructure.
In terms of the Directive 96/48/E performance enough to cover the entire rail infrastructure,
capable to provide high-speed services. However, in practice the speed is not always an
indicator that can clearly delineate the boundary between the speed and high-speed traffic. In
many countries, the commercial rate is limited by factors such as passage of a densely populated
urban areas (to be slowing down to reduce noise levels and the risk of accidents), or because of
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
91
the use of tunnels and viaducts (where the speed limit is set to 160-180 km / h from for security
reasons).
In Europe, there are two types of countries - France and Spain have lower cost of
construction of 1 km road than Germany, Italy and Belgium. The reason is not only geography,
but also the construction of the path used in the construction of the HSL. So in France,
construction costs are minimized by eliminating the tunnels and viaducts. Due to the fact that in
France the HSL is used only for passenger traffic, slope may be 3.5%, while for the mixed type
- only 1-1.5%.
Table 2. The cost of operating the infrastructure 1 km single-track section
of HSL in Europe according to UIC in 2008 / 3, p.26 /
Belgium
142 km
Rolling stock
13 834
43,7%
Energy
2576
8,1%
Alarm
3248
10,3%
Telecommunications 1197
3,8%
Other costs
10821
34,2%
Total
31683
100%
France
2638 km
19140 67,3%
4210 14,8%
5070 17,8%
0
0
0
0
28420 100%
Italy
Spain
492 km
949 km
5941 46,0% 13531 40,4%
2455 19%
2986
8,9%
4522 35%
8654 25,9%
0
0
5637 16,8%
0
0
2650
7,9%
12919 100% 33 457 100%
In all cases, the cost of operating the way has a share from 40% to 67% of total
expenditure, while the share of only 10-35% of signaling, which is less than the usual ways (1545%).
With respect to interaction with the traditional railway network of HSL appropriate to
classify as follows:
- Mixed traffic (new or reconstructed line). In this case high-speed trains also go to the
usual lines. Mixed traffic is dominated mainly by German, Spanish and Italian high-speed lines,
as well as in the UK, Austria, Switzerland, the Netherlands, Belgium, Finland and China.
- Only high-speed traffic with access to a regular network (mostly new line). In this case,
high-speed lines runs only specialized rolling stock, which often comes at the same time and on
a regular network, especially in the areas of hubs and in the final sections of routes. Typical
examples are the French line.
- Stand-alone high-speed network communications (only new lines). Here, the transfer of
rolling stock between conventional and high-speed networks is not provided. An example is a
network of high-speed lines in Japan.
- Mixed traffic on a standard European track 1,435 mm with access to the local track by
having a rolling of the wheel sets. The Spain railways of Spain's gauge of 1668mm (narrowgauge lines at 1000 mm). An example of this is the high-speed rail system in Spain AVE, a
country with advanced technologies in the gauge change. Automatic points change wheelsets
first, second and third generation networks combine various gauges. Daily through the points of
change wheelsets are about 60 trains. Since 1969, these items are served over a quarter of a
million
trains.
92
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
At the moment, ready to introduce automatic gauge change system of the 4th
generation. Unlike previous systems, which apply only to passenger traffic, the new system
would:
- Automatic change of gauge passenger and freight trains;
- The use of rolling stock for different manufacturers;
- The ability to change the track width of mixed trains, that means trains with different
types of cars and manufacturers.
The differences between the systems can be displayed graphically HS systems (see Fig.
4.) For the adopted distance of 250 km length, which is the usual train passes for 2 - 2.5
hours. Time frame for which the same distance will be high-speed trains of different types of
HSL depends on the technical configuration, the values of slope, the number of arrows,
viaducts, etc.
Geographical expansion and range requirements for the product in use meaning ¬ increase
in the number of types and series of high-speed trains. Most important trends in their
development are:
• Achieving a maximum velocity ¬ velocity of 350 km/ h, and (above all in Europe) to
ensure exploitation compatibility with infrastructure of the railways in various countries every ;
• An increase in passenger capacity, while maintaining an adequate level of comfort;
• Reducing energy consumption as energy the manufacture and in use;
• Use of common construction platform to accelerate development and reduce costs in
manufacturing;
• A high level of modularity and interchangeability for reduction of costs.
Fig 4. The position of different systems of SCM and conventional rail
in relation to time intervals and distance / 3, s.22 /
After 2012 it is expected that the Asian demand for new rolling stock will decline due to
the completion of the program complete replacement of the Japanese high-speed trains of first
generation. In contrast, in Western Europe the market will continue to grow as the replacement
program trains first generation in France and Germany are in the initial stage. Another
confirmation of this tendency is a program may be for, which announced in the CIS countries,
South America and North ¬ of South Africa / 1, 2 /.
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
93
Table 4. Types of high-speed trains in the world /3, p.27/
Country
France
Germany
Italy
Spain
China
Japan
Type of Train
Year
TGV Reseau
TGV Duplex
Thalys
ICE-1
ICE-2
ICE-3
ICE 3 Polyc.
ICE-T
ETR 500
ETR 480
AVE
CRH380A
CRH380AL
500
700
N700
1992
1997
1996
1990
1996
2001
2001
1999
1996
1997
1992
2010
2010
1995
1997
2007
Seats
number
377
510
377
627
368
415
404
357
590
480
329
494
1027
1323
571/1323
1323
Serviced distance
(km)
495
525
445
500
400
420
420
360
360
288
470
420
420
550
550
550
Maximal speed
300/320
300/320
300/320
280
280
330
330
230
300
250
300
350/380
350/380
320
285
300
Of particular importance is the account of expenses for operation and maintenance of
trains. In the world of every country that has so far developed its own HST type of rolling stock,
the most responsible realities of life and needs (see table. 4.).
Table 5. HSL technology in Europe: Comparing the costs
of operation and maintenance according to UIC (2008)/3, p.28/
Country
Type of
Train
Operating costs
For train
For
For seat(mln)
seat
km
Maintenance costs
For train
(mln)
For seat
For seat-km
TGV
17,0
45,902 0,0927
1,6
4,244
0,008
Reseau
France
TGV
20,8
40,784 0,0776
1,6
3,137
0,005
Duplex
Thalys
24,8
65,782 0,1478
1,9
5,039
0,011
ICE-1
38,9
62,041 0,1240
3,1
4,944
0,009
ICE-2
26,0
70,652 0,1766
1,4
3,804
0,009
ICE-3
17,9
43,132 0,1026
1,6
3,855
0,009
Germany
ICE 3
20,4
50,495 0,1212
1,7
4,207
0,010
Polyc.
ICE-T
15,5
43,417 0,1206
1,8
5,052
0,014
ETR 500
34,1
57,796 0,1605
4,0
6,779
0,018
Italy
ETR 480
21,1
43,958 0,1526
3,2
6,666
0,023
Spain
AVE
23,7
72,036 0,1532
2,9
8,814
0,018
Maintenance costs depend on the distance between the depot and the terminus, the average
waiting times at the depot, the time of shunting. Greater share of the cost of maintenance costs
take on staff salaries. The flow rate depends on the technology used by the operator. The
average cost per seat (see Table. 5.) Account for 53 €. For AFL 500 km long, considering the
94
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
duty cycle trains 100%, the average operating cost per passenger ranges from 41,3 € (2000) for
the TGV Duplex, up to 93 € for the ICE-2 in Germany.
In Russia, is an issue of competitiveness and profitability of high-speed projects in most of
the territory. Long distances between major conurbations and sparsely populated suggest only a
network version of HSL. And the main principle is to reach the largest possible number of
densely populated areas. This principle applies to the example of the proposed route Moscow Nizhny Novgorod - Kazan (with a branch in Samara) - Ekaterinburg. Distance along the branch
Samara - Kazan - Nizhny Novgorod - Moscow 1300 km, and the branch Samara - Penza Moscow 1030 km, much less with the high cost per km of new road.
The population of Kazan, $ 1.1 million, Nizhny Novgorod - 1.3 million people. In major cities
at an alternate route home: Syzran - 0.2 million people in Penza - 0.5 million people in Ryazan 0.5 million people.
Primarily based on return line is the proximity of highways to a greater number of
consumers, so the advantage is on the side of more extended, but more densely populated route.
If you continue to branch in Samara to Saratov and Astrakhan, or through Samara to
Orenburg is possible to connect the Volga and Southern Urals into a single high-speed
network. In addition, the route Samara - Orenburg is a perspective of international importance
and is indirectly linked to China's ambitious plans for the organization of high-speed traffic
through Beijing - London. The most likely way for Transeuroasian route lies on the line Beijing
- Hohhot - Hoto - Bao - Tou - Uday - Urumqi - Friendship - Aktogai - Astana and then through
Russia en route Orenburg - Samara - Penza - Moscow, where it is possible to extend the line in
the directions of Brest - Kosice - Budapest - Rome, Warsaw - Prague - Munich and so on.
Network organizing principle of HSL in Russia will intensify passenger traffic and
economic activity not only in the direction of the center - periphery, but also between regional
industrial and economical centers.
Economic evaluation of infrastructure upgrades, construction of new facilities, changes in
technology, transport processes and other activities for the organization of the HST is the
following: cost accounting HST; economic evaluation of changes in the schedule and timetable
of the freight train, options for local work stations and areas in the implementation of the HSL,
the options interaction of urban, trunk, air and river passenger transport options for the highspeed lines: the construction of special ramps, construction of new railway lines, the upgrading
of existing lines under the HSL.
References
[1]. Высокоскоростное движение: что впереди? // Железные дороги мира. – 2009. - № 7. – С. 9-20. - Rus.
[2]. Высокоскоростные сообщения: частота важнее скорости // Железные дороги мира. – 2010. - №
3. – С. 9-22. - Rus.
[3]. Rus G., Barron I., Nash C, Economic Analysis of High Speed Rail in Europe // Fundacion BBVA.
– 2009. – P. 140.
[4]. Материалы сайта http: // rbcdaily.ru
[5]. Материалы сайта www.hsrail.ru♦
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
95
CALCULATION PROCESS OF SETTLEMENT - TRANSITION
PILE NET FOR BRIDGE APPROACH EMBANKMENTS
PhD. VU THE SON
University of Transport
and Communications
MA. NGUYEN VAN SUU
Transport Engineering Design
Joint Stock Incorporated South
MA. TRAN HUU BANG
Sai Gon Construction Quality Control
Joint Stock Company
Summary: Based on the analysis of load transfer structure in pile net, this report
summarizes the calculation content of pile net solution in the settlement transition treatment at
bridge approach embankments on soft soil. Calculation content includes: Determine the length
of pile net, calculate the vertical forces on the pile caps, the extreme load capacity of piles,
settlement of pile and geosynthetic. The calculation example is applied for the Abutment A2Binh Loi Bridge of Tan Son Nhat - Binh Loi - Outer Ring Road Project with the necessary Pile
net length to ensure both stability and the slide is 24m. The calculated pile lengths vary from
22m to 12m, from the abutment adjoinment to the end of the pile net. Corresponding
settlement after 15 years remaining at the end of the pile net is 17.5cm.
I. INTRODUCTION
The pavement of bridge approach embankments is often settled, cracked and broken at the
start and the end of the approach road. In Vietnam, as a result, vehicles cannot run smoothly and
cannot reach high-speed after an operating time of load transfer platform. This affects the
operation efficiency of roads. There are many solutions of soft soil treatment in approach road
with high efficiency such as: Prefabricated vertical drain, Sand drain, Deep cement mixing, Pile
slab [8]. However, these solutions have not completely solved the above matters.
It is important to consider, evaluate and find a solution of soft soil treatment for approach
roads that ensures the smooth run of vehicles and maximizing the operation capacity of roads.
Based on the analysis of behavior of the pile net and soil, as well as related documents,
foreign and domestic specifications; this paper proposes calculation content of settlementtransition pile net. The calculation content of pile net solution is illustrated through the example
of Abutment A2 at Binh Loi Bridge in Tan Son Nhat - Binh Loi - Outer Ring road Project.
96
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
II. PILE NET SOLUTION FOR SETTLEMENT TRANSITION
Geosynthetic combining with the reinforced concrete piles with variable lengths is known
as settlement - transition pile net as shown in figure 1 and figure 1.
Figure 2. Typical cross section of Pile net solution
Piles adjacent to abutment are calculated for fully length restrained in hard soil layer. The
length of remaining pile increases from pile net to abutment position, so settlement is also
reduced correspondently to create a settlement transition area.
Figure 2. Profile of Pile net solution of settlement transition
In some countries, geosynthetic has been placed on piles (reinforced concrete piles, deep
cement mixing) with a constant length to treat the bridge approach road.
III. CALCULATION CONTENT
Based on the analysis of load transfer in Pile Net, report proposes calculation content of
pile net solution for settlement transition as following figure 3:
Figure 3. Calculation content of pile net solution for settlement transition
3.1. Determine the length for Pile Net
Assess the stability of pile net vertical road method based on circular slip through Bishop
GeoStudio 2004 software to audit [3]. Based on the sliding characteristics of the arc length
round to define the length for pile net
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
97
3.2. Load transfer mechanism in Pile Net
It is important to consider the load transfer mechanism in Pile Net from the interaction
between the piles, geosynthetic and soil. This is the basis issue for calculation of settlementtransition pile net.
Considering the load distributing on Pile Net as figure 3 [7]:
Figure 4. Load transfer structure in Pile Net
With: Ws - Surcharge effect on embankment (kN/m2); W1 - Weight of embankment
between the adjacent pile caps (kN); W2 - Weight of embankment on pile caps (kN); Wt - The
vertical load effect on the strengthened reinforcement between the adjacent pile caps (kN/m2);
P’c - The vertical stress on pile caps (kN/m2); Trp - The tensile load in the reinforcement (kN/m);
τ - Shear resistance of soil (kN/m2); H - Height of embankment (m); a - The size of the pile
caps (m); S - The spacing between the adjacent piles (m).
Strengthened reinforcement combine with good soil layer cover above, has effect
transferring the embankment load between piles to pile caps. Blocks of embankment will be
moved down and settlement much because the below soil layer of Strengthened reinforcement is
very weak. Transition is reduced by shearing resistance of embankment on the pile cap. Shear
strength of embankment on the pile cap reduce the pressure on Strengthened reinforcement but
it make increase the pressure on the pile cap.
To calculate Pile net, It is necessary to calculate the value of vertical pressure on the pile
cap. The order of calculation is as following:
- Define the average vertical tress at the base of the embankment follow formula [6]:
σ 'v = DL.f fs + Ws .f q
(1)
With: DL - Dead load effect on piles (T/m2); ffs - The partial factor for soil unit weight
(refer to table [6]); fq - The partial load factor for surcharge (refer to table [6]).
The vertical stress on the pile caps is determined as following formula [6]:
2
⎡ C .a ⎤
Pc = ⎢ c ⎥ .σ 'v
⎣ H ⎦
'
(2)
With: Cc - The arc coefficient.
98
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
3.3. Calculation of the vertical forces on the piles cap
Calculation of the vertical forces on the piles cap are determining load (Dead load and live
load) on piles cap. The vertical forces on the piles cap calculated follow formula [5]:
Q a = Pc' .a 2
(3)
3.4. Calculation of total pile capacity
Total pile capacity include is resistance of pile (Skin friction and end bearing).
- Extreme Load of pile according to soil is calculated by the formula [3]:
Q u = Q p + Qs
(4)
With: Qp - Resistance of pile cap (ton); Qs - Resistance of pile side (ton).
- Safety factor about Load ability of Pile is calculated as following [2]:
Fs =
Qu
Qa
(5)
3.5. Calculation of Pile settlement
Method of settlement analysis of pile groups in Pile Net similar to methods used in
common pile foundation [2].
Settlement estimation method is according to Guideline at Section VI of survey design
process of embankment on soft soil layer 22 TCN 262-2000 [1].
3.6. Calculation of reinforcement
Reinforcement is located below embankment to increase the strength of embankment,
avoid embankment being destructed by excessive deformation or sliding cutting into below soft
soil layer.
According to British Standard BS 8006:1995 when designing strengthen reinforcement
in this case according to the theoretical of flexible cable system [6].
a. According to longitudinal Direction of embankment
Considering calculation diagram as figure 5:
Figure 5. Calculation diagram of tension with longitudinal direction of embankment
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
99
With: d - Diameter of pile (m); Trp - Tension of reinforment with longitudinal direction of
embankment (kN/m).
Tension Trp in reinforcement is determined as follow:
Trp =
WT (s - a)
1
1+
2a
6ε
(6)
With: ε - The elongation of strengthened reinforcement according to rolling direction.
Direction along the embankment length, maximum tension is necessary strength to move
up the vertical load on the pile cap Trp.
Tr = Trp
(7)
b. Horizontal direction of embankment
Considering Calculation diagram of strengthened reinforcement as figure 6:
Figure 6. Calculation diagram of tension with horizontal direction of embankment
With: Tds - Inner Tension of reinforcement with horizontal direction of embankment
(kN/m); Le - Minimum length of anchor of reinforcement with horizontal direction of
embankment (m); Lb - The reinforcement bond length needed beyond the outer row of piles
across the width of the embankment (m); Lp - The horizontal distance between the outer edge of
the outside pile cap and the footing of embankment (m).
Tension Tds of strengthened reinforcement need to determine as following:
Tds = 0.5K a (f fs .γ.H + 2f q .Ws )H
(8)
With: Ka - The coefficient of active soil pressure; γ - The unit weight of the embankment
material (kN/m3).
According to the horizontal width of embankment, the biggest tension must be the total of
necessary strength to transfer the vertical load on the pile cap Trp and necessary strength to resist
the horizontal thrust Tds.
100
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
Tr = Trp + Tds
(9)
IV. EXAMPLE
4.1. Data for calculation
- The physic-mechanical property of this layer at Abutment A2 - Binh Loi bridge of Tan
Son Nhat - Binh Loi - Outer Ring Road Project [4].
- Unit weight of embankment Material [3]: γađ = 2.3 T/m3, γsb = 2.0 T/m3, γs = 1.8 T/m3, γc
= 2.4 T/m3;
- Data for calculation of reinforcement [3]: a = 1m, Bemb = 29m; n=1.5; Hfill =3.69m;
φcv = 300, Ws = 1.18 T/m2.
4.2. Calculation results
Summary results is displayed as Table 2 and Settlement is shown on diagram 7 [3]:
Figure 7. Settlement transition of pile net
Remark: according to diagram, it shows that the settlement increases from the position
close pier to the end of pile net. For transiting settlement, calculate the residule settlement after
15 years of normal embankment section same with the settlement of the last pile when starting
at pile net.
Calculation results of reinforcement is showed in table 1 [3]:
Table 1. Calculation results of reinforcement
Item
Units
Results
The tensile load in the reinforcement
kN/m
Trp =
22
The tensile load in the reinforcement (prevention of lateral sliding)
kN/m
Tds =
208
Maximum tension in reinforcement
kN/m
Ttot=
257
Extreme tensile stability of reinforcement
kN/m
Tult=
55
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
101
Remark: Based on the calculated results, the appropriate type of geosynthetic (according to
specification of manufacturers) must have the biggest pull-out strength along longitudinal
direction of Tult = 55 kN/m and along horizontal direction of Ttot = 257kN/m.
Table 1. Summary of calculation result about Pile Net of Settlement Transition
Row No
1
2
3
4
5
6
7
8
9
10
11
12
13
Distance from
abutment (m)
1.2
3
4.8
6.6
8.4
10.2
12
13.8
15.6
17.4
19.2
21
22.8
Height
embankment
(m)
3.69
3.64
3.58
3.52
3.47
3.41
3.36
3.31
3.26
3.21
3.16 3.10
3.05
Pile length (m)
22
22
22
20
20
20
18
18
16
16
14
14
12
Spacing (m)
1.2
1.8
1.8
1.8
1.8
1.8
1.8
1.8
1.8
1.8
1.8
1.8
1.8
Skin
(ton)
friction
65.1
65.1
65.1
54.9
54.9
54.9
46.2
46.2
39.1
39.1
28.3 28.3
17.4
End
(ton)
bearing
27.5
27.5
27.5
26.2
26.2
26.2
16.7
16.7
5.3
5.3
5.3
5.3
5.3
Total
pile
capacity (ton)
92.6
92.6
92.6
81.1
81.1
81.1
62.9
62.9
44.4
44.4
33.5 33.5
22.7
The vertical
stress on the
pile
caps
(ton/m2)
24.2
24.2
24.2
23.3
23.3
23.3
22.5
22.5
21.9
21.9
21.4 21.4
20.8
The vertical
forces on the
pile caps (ton)
24.2
24.2
24.2
23.3
23.3
23.3
22.5
22.5
21.9
21.9
21.4 21.4
20.8
Safety factor
3.8
3.8
3.8
3.5
3.5
3.5
2.8
2.8
2
2
1.6
1.6
1.1
Total
settlement(cm)
0
0
0
1.3
1.3
1.3
2
2
5.6
5.6
17.5 17.4
26.0
Settlement
after 15 year
(cm)
0
0
0
1.3
1.3
1.3
2
2
5.4
5.3
14.2 14.1
17.5
Consolidation
(%)
0
0
0
100
100
100
100
100
96
96
81
81
67
Remark: To calculate the settlement transition at abutment A2 - Binh Loi Bridge, piles
adjacent to abutment are calculated for fully length restrained in hard soil layer. The length of
remaining piles increases from the starting position of pile net to abutment (Lpile = 12 ÷22m),
therefore the settlement is also correspondently reduced (S15year = 17.5÷0 cm).
102
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
V. CONCLUSION
This paper proposes the calculation content of pile net solution in the settlement transition
of bridge approach embankment including: Determine the length of pile net, calculate the
forces distribution on piles cap, the extreme load capacity of piles, settlement of piles and
geosynthetic. The length of piles decreases from the starting position of bridge approach road to
abutment to solve problem of settlement transition. Piles adjacent to abutment are calculated for
fully length restrained in hard soil layer. The length of piles at the end of the pile net is
calculated so that the remaining settlement after 15 years is equal to that value of common
embankment section to create the area of settlement transition.
The application is for the Abutment A2- Binh Loi Bridge of Tan Son Nhat - Binh Loi Outer Ring Road Project. Pile lengths vary from 22m to 12m from the abutment to the end of
pile net. Corresponding settlement after 15 years remaining at the end of pile net is 17.5 cm. The
selected geosynthetic should be able to bear the biggest pull-out strength longitudinally of Tult
= 55 kN/m and horizontally of Ttot = 257 kN/m.
References
[1]. Ministry of Transport. 22 TCN 262 - 2000 Standard. The publisher transport and communication, Ha
Noi, 2001.
[2]. Ministry of Construct. Pile foundation - TCXD 205 - 1998 Standard.
[3]. Nguyen Van Suu. Applied research of pile net solution in the settlement transition treatment at
approach road embankment on soft soil. Master thesis of scientific and technical, University of Transport
and Communications, 2011.
[4]. Tedisouth. The report geological of Tan Son Nhat - Binh Loi - Outer Ring Road Project. Technical
design, 2009.
[5]. Braja M. Das. Principles of Foundation engineering, Third Edition. California State University,
Sacramento.
[6]. British Standard BS 8006: 1995. Code of Practice for Strengthened/Reinforced Soils and Other Fills.
British Standard Institution, London.
[7]. Rutugandha Gangakhedkarga. “Geosynthetic Reinforced Pile Supported Embankment”. Master of
Engineering, University of Florida, 2004.
[8]. Anand J. Puppala. Recommendations for design, construction, and maintenance of bridge slabs.
Synthesis Report, the University of Texas, 2009♦
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
103
ADVANCES IN DESIGN THEORIES
OF HIGH-SPEED RAILWAY BALLASTLESS TRACKS
XUEYI LIU
PINGRUI ZHAO
FENG DAI
Southwest Jiaotong University,
Chengdu 610031, China
Summary: The design theories of the ballastless track in the world are reviewed in
comparison with the innovative research achievements of high-speed railway ballastless track
in China. The calculation methods and parameters concerning train load, thermal effect, and
foundation deformation of high-speed railway ballastless track, together with the structural
design methods are summarized. Finally, some suggestions on the future work are provided.
Key words: High-speed railway; ballastless track; design theory.
I. INTRODUCTION
Structure forms and design theories of ballastless tracks vary across the world due to the
different development backgrounds. In Japan, the slab track was typically laid on the solid
foundation such as a bridge or tunnel at first, and then gradually developed to the soil subgrade
afterwards. It adopts the unit design that takes into account the effect of train load. The German
ballastless track was first laid on the soil subgrade and then on the foundation of bridges and
tunnels. Its continuous structure involves the consideration of thermal effects. The early
ballastless track in China was mainly laid in tunnels with the chief concern being the influence
of train load. With the increasing application of ballastless track, a relatively general design
theory and a structural system have been gradually formed after the innovative research with
high-speed railway ballastless track.
This paper reviews the calculation methods and parameters as well as the structure design
procedures, and briefly introduces the advance in the design theories, of ballastless track based
on the innovative research achievements in China. Finally, some suggestions on the future work
are provided, including fatigue properties under the coupling action of train and temperature
load, durability, long-term dynamic properties, and maintenance mechanics of the ballastless
track.
II. OVERVIEW OF BALLASTLESS TRACK DESIGN THEORIES
In the design of Japanese slab track, the train load effect is a primary concern. Using the elastic
design method, the security during the manufacturing, hoisting, and constructing of the slab track is
104
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
maximized. As seriously damaged CA mortar at the slab corner and the slab warping caused by
temperature gradients emerged, the uneven support caused by warping is considered in the analysis
[1]. In the baseplate design, in accordance with the limit state method, the train load and the
subgrade’s uneven settlement are considered together with the influence of weather conditions,
concrete contraction, and construction.
German developed its ballastless track by borrowing the design concept and method of
pavement engineering [2]. Most has longitudinally continuous structure, and temperature load and
concrete contraction are the main factors to be considered in the design. The reinforcement is
located near the neutral axis and does not bear the train load. The effect of train load and
temperature gradient is resisted by the rupture strength of the concrete.
In China, the early monolithic roadbed track, whose structure design mainly considers the
train load, was applied in the tunnels with good foundation condition and little temperature
variation. The structural design of the Suining-Chongqing railway took into account the effect
of uneven foundation deformation and temperature load [3]-[4]. Following systematic research
on the ballastless track, the design theory based on the allowable stress method was created with
full consideration of train load, temperature, and foundation deformation effect.
In general, the design theory of ballastless track in different country was relevant to its own
construction environment and structure evolution. The design theory proposed in different
periods could meet the construction requirements for different types of ballastless track.
III. CALCULATION OF TRAIN LOAD STRESS
The track supports the train load and guides the vehicle operation. The calculation of train
load stress must be considered in the ballastless track design. The elastic foundation beam
model [5]-[6] is mainly used for calculation of the load stress in the traditional track structure.
The model can be solved using the multilayer composite beam theory on the elastic foundation
[7]-[10] according to the complexity and analysis requirement of the track structure. In
Germany, however, the Eisenmann theory [11]-[13] was adopted to calculate the stress of the
rail structure under the train load. In this theory, rail is regarded as an infinite beam on the
elastic foundation to calculate the support reaction of the fastener; the multilayer structure is
translated into a monolayer one according to the connection status of the structural layer, and
then the internal force and displacement of the converted monolayer structure under the action
of fastener force is calculated using the infinite beam on the elastic foundation and Westgaard’s
stress function.
To sum up, the main components are treated as flexural members in the train load design of
ballastless track in China and Japan. This is because the ballastless track design was originally
developed based on the traditional design methods for ballast track that put an emphasis on
simulation of the force properties of main components and the generality of analysis method. In
Germany, however, the design theory and parameters selection of ballastless track were
developed from the experience of highway concrete pavement design; thus, its structural
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
105
difference in ballastless track can also be attributed to heritance of the traditional design theory.
In accordance with the structural characteristics that the rail and the sleeper are crosssupported on the elastic foundation in the ballast track, the cross beam model on the elastic
foundation [14-15] was developed on the basis of the elastic foundation beam model, and can
also be used for the stress calculation of the ballastless track [16] once the values of the model
parameters are determined. Thanks to the development of the computing technology, the solid
finite element model [17-19] can be employed to obtain the particular stress state inside the
ballastless track structure.
As the major supporting structure of the ballastless track, the track slab (or bed slab) and
baseplate (or supporting layer), whose deflections under the train load are far smaller than their
thicknesses, have a far smaller size in the vertical direction than in the longitudinal or lateral
direction. This feature conforms to the structural characteristics of the elastic plate.
Consequently, the elastic plate [20] is generally adopted for simulation and analysis of the
supporting structure of ballastless track. The rail, a slender structure, is reasonably simulated by
the beam model, while the fastener and the intermediate elastic layer, as well as the foundation
below, are simulated with different kinds of springs. As a result, a beam-plate model of
ballastless track on elastic foundation [21]-[23] is built as shown in fig 1.
ERJR
Es, hs
P
P
Eb, hb
Kf
Ki
kRD
Notes: ERJR is the flexural rigidity of rail, where ER is the
modulus of elasticity of rail, and JR the moment of inertia of
rail; Es and hs are the modulus of elasticity and thickness of
track slab, respectively; Eb and hb are the modulus of
elasticity and thickness of baseplate, respectively; Kf is the
rigidity of fastener; Ki is the rigidity of intermediate elastic
layer; kRD is the rigidity of foundation below; and, P is the
train load.
Fig 1. The elastic foundation beam-plate model of ballastless track
The load stress of the track slab (or bed slab) and the baseplate (or supporting layer) in the
longitudinal and lateral directions can be obtained by exerting a vertical train load on the rail.
This avoids the calculation in the longitudinal and lateral directions separately in the multilayer
elastic foundation beam model. Moreover, the computational accuracy [7] is higher than that via
the composite beam model or the cross beam model, and the computing workload is less than
that via the solid finite element model.
The design wheel load of the Japanese slab track takes into consideration the wheel load
variation due to wheel tread damage and tolerates three times the static wheel load. In fatigue
checking, the allowable wheel load is 1.45 times the static wheel load. On the basis of the
106
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
allowable value of the derailment coefficient, the design lateral force was determined, and the
lateral force for fatigue checking takes half of the design lateral force. In the Germany
ballastless track design, the load takes the UIC71 with the dynamic coefficient of 1.5 and the
unbalance loading coefficient of 1.2. In China, the dynamic coefficient is based on the results of
dynamic tests and simulation calculations of the ballastless track, and the design wheel load can
be three times the static wheel load. Based on the design parameters and operation conditions of
the ballastless track on the passenger dedicated line (PDL) in China, the coupling dynamics of
train and track system is applied to the statistic analysis. Considering the construction and
maintenance conditions of the ballastless track in China, it is suggested that the constant effect
train load be up to 1.5 times the static load [24].
The Winkler foundation is used to support the ballastless track, and the diameter of the
bearing plate has a significant influence on the foundation coefficient. The smaller the diameter,
the larger the foundation coefficient [1]. However, when the diameter D is not less than 76 cm,
the change in the diameter has little influence on the foundation coefficient. As for the
ballastless track, the supporting area of the track slab or the supporting layer is relatively large.
Thus, for simplicity, the trial value of the bearing plate with a diameter of 76 cm, namely k76,
can be used for calculations. When the subgrade compaction capacity is represented by the
deformation modulus, the layered elastic system mechanics [3] can be applied to analyze the
displacement of the subgrade surface with the even load of the rigid bearing plate; thus,
deducing the supporting rigidity of the subgrade surface [25].
Within every bearing layer of the ballastless track, the substructure is generally weaker
than the upper structure, and may readily crack under the train load if plain concrete or cement
stabilized materials are applied. Once cracking, the bending moment is not readily transferred at
the crack location, resulting in a reduction in the entire rigidity and the modulus of elasticity.
Therefore, the reduced elastic modulus is used for calculation [26]. As for the reinforced
concrete structure, the reinforcement is helpful to improve the flexural rigidity of the structural
layers. However, due to the possible cracking, the transmission of the bending moment at the
cracked location may be weakened. Consequently, only the concrete elastic modulus is used for
calculation, without consideration of the influence of the reinforcement and crack.
IV. CALCULATION OF TEMPERATURE STRESS
The ballastless track is exposed to the atmosphere. With changes in external temperature,
the temperature in every structural layer will vary. Once the deformation of the ballastless track
due to the changing temperature is restrained, the temperature stress will occur inside the
structure. The ambient temperature variation with an effect on the ballastless track includes the
yearly temperature variation and daily temperature variation. In addition, the contraction of
concrete will cause distortion, which is equivalent to decreasing the temperature load acting on
the concrete.
The design of the continuous ballastless tracks represented by Rheda, Züblin, and Bögl in
Germany attach great importance to the temperature effect. In order to limit the width of the
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
107
temperature cracks within the admissible range and maintain the state of incomplete cracks [27],
the ratio of reinforcement in the slab should reach 0.8%–0.9%, according to the German
Ballastless Track Design Specification. As a result, the width of cracks is limited within 0.5 mm.
From the viewpoint that the sum of the minimum stress of the reinforcement with the slab
cracking and the bending stress increment under the dynamic load must be less than the
reinforcement fatigue limit to guarantee the service life, it is supposed that the longitudinal ratio
of the reinforcement must be larger than 1.0%, so as to meet the demands of crack width and
service life.
The Japanese slab track design adopts unit structure, and temperature variation has little
influence on the track slab. Thus, temperature effect is not considered in the design.
Nevertheless, warping displacement of the track slab is found in tests, where the track slab is in
a state of being incompletely supported. Therefore, to address the variation properties of the
track slab due to temperature, a series of theoretical and experimental research has been
conducted [4].
As for the continuous slab structure, under the action of concrete contraction and
decreasing temperature, concrete may easily crack, causing a stress redistribution of the
reinforcement and concrete inside the slab. In order to guarantee security and utility, it is
necessary to control the reinforcement stress and crack width.
The continuous slab shows different stress and variation properties at various tension
stages. Before the concrete cracks, the concrete deformation is coordinated by the
reinforcement. When the tensile stress of the concrete reaches its tensile strength, it will crack
and stop working, which leads to the bond damage adjacent to the cracks. At this moment, the
plain section hypothesis does not fit any more, and the reinforcement at the crack location bears
all the axial force. When the axial force increases to the yield strength of the reinforcement, the
concrete is cracked severely without bearing the tension. All the axial force is born by the
reinforcement, such that the reinforcement yielding becomes the limiting condition of the slab in
tension. The cracking axial force of the continuous slab depends on the tensile strength of the
concrete and the sectional area of the slab. The amount of reinforcement has little influence on
the cracking axial force, while the ultimate bearing capacity completely depends on the yield
strength and the area of the reinforcement. In order to avoid cracking, the minimum ratio of
reinforcement of the continuous slab should be specified.
The cracking in the continuous slab go through two phases: incomplete cracking and
complete cracking. At the stage of incomplete cracking, the amount of cracks increases with the
increasing load, and the maximum crack width remains basically unchanged. At the stage of
complete cracking, the number of cracks remains unchanged, while its width increases with the
increasing load. In order to limit the crack width, the cracking should be controlled at the stage
of incomplete cracking. In the cases of incomplete cracking, the maximum temperature force
inside the slab depends on the tensile strength and the sectional area of the concrete. The
temperature force calculated with the design tensile strength is regarded as the common
temperature force (main force). And the temperature force calculated with the standard tensile
strength is taken as the maximum temperature force for checking in design. Refs. [28-29]
108
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
elaborated the different expressions of fracture interval, cracking width and reinforcement stress
at different stages of cracking, and the relevant design measures have been put forward.
As for the unit bed slab structure, the temperature force of the slab is influenced by the
longitudinal resistance of the fastener at the top and the frictional resistance at the bottom, as
well as the displacement limitation of the convex plate. The classification of the unitary and
continuous structure depends on whether the temperature force leads to the full-section
cracking of the slab.
V. CALCULATION OF WARPING STRESS
The external environment will affect the temperature and humidity of the concrete slab.
The influence of external environment gradually weakens with the depth from slab surface. The
uneven distribution of temperature and humidity inside the slab leads to its warping
deformation. When the deformation is restrained by the bottom friction, dead load, stop blocks,
and train load, the warping stress occurs.
According to the German railway code, it is hypothesized that the slab in the vertical
direction has a linear temperature gradient of 50 ℃/m. In the temperature field test of the
ballastless track on Suining-Chongqing railway, the temperature gradient [30] of the track
before laid is about 52.6–68.4 ℃/m and the temperature gradient of the slab track in the
longitudinal direction on the Jialingjiang bridge is approximately 40–80 ℃/m [31], with a large
dispersion, but all larger than that of 50 ℃/m in Germany.
In terms of geography and climate conditions, China has severely cold areas, cold areas,
and temperate areas. Referring to the recommended value of the temperature gradient in the
field of highway pavement, in consideration of the structure characteristics of the ballastless
track, we advise that the maximum positive temperature gradient of the uppermost structure of
the ballastless track in China be 80–85 ℃/m, 85–90 ℃/m and 90–95 ℃/m for severe cold area,
cold area and temperate area, respectively, and that the temperature gradient distribute linearly
in the vertical direction. The effect of temperature gradient can be neglected in the substructure.
The negative temperature gradient can be half the maximum positive temperature gradient.
According to the statistical data about the temperature and the temperature gradient
variation in Germany, studies have been conducted to analyze the slab stress state under the
action of the temperature gradient, especially the slab with smaller lateral size whose warping
deformation is not restrained completely. The calculation model with discontinuous supporting
was utilized to calculate the warping stress [32] under the action of dead load and temperature
gradient.
The warping stress and displacement of the slab track in different constraint conditions
were analyzed by finite element theory. The results show that the stronger the restraint acting on
the track slab, the more the warping deformation is resisted, and the closer the warping stress in
the slab track to that of an infinite slab. The restraints acting on the track slab include the track
dead load, the restraint of the continuous long rails, and the train load acting on the rails.
Because of the large supporting coefficient in the ballastless track supporting system, the
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
109
loading restraint of the track slab, due to the limitation of the loading magnitude and position,
shows many differences. For convenience, no matter for the unitary or the continuous structure,
the warping stress of the ballastless track in the longitudinal or lateral direction is calculated in
accordance with the infinite slab.
VI. CALCULATION OF FOUNDATION DEFORMATION EFFECT
Ballastless track will be influenced remarkably by the large rigidity of the track slab or the
bed slab once uneven deformation occurs at the foundation.
In the Japanese slab track base design, the maximum settlement displacement (δ) occurs at
the mid-point and at the ends of the baseplate with the half-wave sinusoid of δ=20 mm at the
service and fatigue state, together with that of δ=30 mm at the ultimate state. Based on the
deformation relevance, the rigidities at different locations of the settlement area with an
interval of 5 m are calculated to ensure the settlement of the baseplate under the dead load
reaches the designed uneven settlement. Then the additional bending moment [33] due to
foundation deformation of the baseplate is calculated. Germany has a concept of “zero
settlement” that uneven settlement must not occur. Thus, there is no need to consider the
uneven settlement effect in design. Although high-speed railways have developed rapidly in
China, uneven settlement is also inevitable at the subgrade-bridge transitional sections and
high embankment. In order to ensure the proper operation of ballastless track, the influence of
uneven settlement of foundation should be considered in the design of ballastless track in
China.
Because of the large rigidity of ballastless track, when there is uneven settlement, the slab
will have the same deformation as the foundation, which can be viewed as a forced
displacement of the slab structure. In this case, the bending moment of the slab under the action
of foundation deformation equals to the product of its flexural rigidity and the uneven
deformation curvature.
VII. DESIGN OF BALLASTLESS TRACK STRUCTURE
The bearing structures of the ballastless track mainly include plain concrete, reinforced
concrete, and prestressed reinforced concrete. The plain concrete structure is usually applied to
the tunnels with good foundation and small ambient temperature variation. In this case, the slab
will not crack [34] under the action of train load and environmental factors. Under the common
foundation conditions, the slab may readily crack with the influences of foundation deformation,
train load, and environmental factors. Thus, it is necessary to add reinforcements to limit the
crack development. As for the continuous reinforcement concrete slab, because the temperature
stress is the main influencing factor, reinforcements are laid near the neutral axis to limit the
crack width and crack interval of the track slab. For the sections with severely weak
foundations, the bending moment in the slab is usually large. In order to limit the crack width,
we need to thicken the slab or improve the foundation, which results in high costs. In that case,
placing reinforcements in top and bottom layers can help the track slab bear more bending
110
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
moment [35-36]. To limit the crack width and improve the structure durability, steel fiber
concrete has been increasingly used in the ballastless track structure [37-38]. In cold areas,
prestressed reinforced concrete structure is often adopted for decreasing the freezing injury.
The allowable stress method and the ultimate state method are generally utilized in the
concrete structure design. As the Japanese track slab was designed as reinforced concrete
structure originally, the allowable stress method is adopted provided that the track slab concrete
under the action of bending moment conforms to the hypothesis of plane mechanism, while the
tensile stress of the concrete in the tension zone is negligible. The allowable stress of the
reinforcement depending on the repeated loading times varies with different design wheel load
and structure types. In the cold areas, anti-freezing measures should be taken. Considering
factors such as construction and costs, the prestressed reinforced concrete structure [39]
designed by partial limit state theory is applied.
For the German ballastless tracks like Rheda and Züblin, the longitudinal reinforcements
are placed in the continuous slab for the purpose of controlling the crack types and width. The
width of the slab is determined by the Westergaard’s stress equations and the allowable
compressive stress of the subgrade surface. Determination of the slab thickness follows the
principle that the stress caused by temperature gradient and load is less than the flexural strength
of the slab concrete. The supporting layer is composed of plain concrete probably with cracks or
is the hydraulic supporting layer structure. The load stress should be checked within the
permissible limit to determine the modulus of elasticity of the supporting layer.
Ballastless track, laid on the elastic foundation under the long-term repeated action of train
load and environmental change, is of band structure distinct from the structures like bridge and
building. In order to ensure its high accuracy and high stability, the rail structure is required to
work in an elastic condition under the action of train load and surrounding factors. Therefore,
we suggest that the ballastless track structure design adopts the allowable stress method for the
innovative research of high speed railway in China.
During the design, it is assumed that every plane cross-section remains a plane under the
action of the bending moment. The normal stress of the concrete in the compression zone takes
a triangle pattern, the tensile strength of the concrete in the tension zone is neglected for the
reinforced concrete components, and the normal stress of the concrete in the tension zone also
takes a triangle pattern for prestressed reinforced concrete components. Under the action of
axial force, the temperature stress of the continuous ballastless track, which may cause cracking,
is resisted by the reinforcement, and the plane assumption is invalid.
Because the method for calculating the load effect, especially under the action of bending
moment is different from the calculation model for structure design, a correction factor is
introduced to eliminate the difference in the obtained results. The specific design flow for the
ballastless track structure is shown in fig. 2.
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
111
As for the unit ballastless track, the reinforcement is mainly based on the load bending
moment and the effect temperature force is negligible. For the continuous ballastless track,
concrete contraction and temperature decreases are the main factors influencing the reinforcement.
The load combinations for the different kinds of ballast-less track on diverse foundations are listed
in table 1.
The daily temperature has a periodic variation, leading to a periodic variation in temperature
stress and warping stress. Nevertheless, the maximum temperature gradient and the maximum
temperature force do not appear everyday. Especially for the continuous ballastless track, the
maximum temperature force only occurs at the critical state when a new crack appears. When the
crack is stabilized, the temperature force is mostly less than the maximum temperature tension.
Therefore, the maximum temperature tension is unlikely to appear in the continuous ballastless
track, and it can be regarded as a kind of load combination and checked independently.
The subgrade of PDLs is required to be designed and constructed under the concept of
“zero settlement”. However, uneven settlement is easy to occur at the transitional section
between subgrade and the structures such as bridge, tunnel or culvert. And the probability of the
uneven settlement within a small range occurring to common sections is quite low. Therefore,
the uneven settlement of subgrade should be combined as an additional force with a low
probability of occurrence. Under the train load, a bridge has bending deformation which
coincides with the train load. Consequently, the bridge bending deformation should be
combined as the main force the same as the train load.
Structure functional design
Tentative
structure size
Foundation deformation
bending moment
structure coefficient
Internal force
calculation
Tentative
reinforcement
Train load bending
moment structure
coefficient
Temperature gradient
bending moment
structure coefficient
Train load bending moment
Temperature gradient
Foundation deformation
bending moment
Temperature force
Structure
coefficient
calculation
Load stress σ
σ<[σ]
Allowable stress [σ]
Reinforcement
optimization
Finish design
Fig 2. Design flow for the ballastless track structure
112
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
Table 1. Suggested load combinations for different types of ballastless tracks
Load combination
Type
Unit
ballastless
track
Continuous
ballastless
track
Main force
Main force
+additional force
Main force
Main force
+additional force
On subgrade
On bridge
In tunnel
3MTrain
3MTrain
3MTrain
1.5MTrain+M∆T
1.5MTrain+M∆T+ MD
-
1.5MTrain+M∆T+ MD
-
-
3MTrain
3MTrain
3MTrain
FT, max
FT, max
FT, max
1.5MTrain+M∆T+FT
1.5MTrain+M∆T+ MD +FT+ FD
1.5MTrain +FT
1.5MTrain+M∆T+MD+FT
1.5MTrain+M∆T+ MD +FT+FB+FD
-
B
Note: MTrain denotes the bending moment caused by train load, M∆T denotes the bending moment
caused by temperature gradient, MD denotes the bending moment caused by foundation deformation,
FT,max denotes the maximum temperature force, FT denotes the temperature force, FD denotes the axial
force caused by foundation deformation, and FB denotes the braking force.
According to the load combinations shown in Table 1, one should decide whether the edge
stress of the supporting layer in the ballastless track exceeds its cracking stress. If the edge
stress is lower than the cracking stress, then the supporting layer of the concrete will not crack,
and reinforcement is unnecessary or should be placed in accordance with the structure. If the
edge stress is higher than the cracking stress, then the supporting layer will crack. Especially for
the continuous ballastless track, full-section cracking is likely to occur. At the moment, all the
concrete in the tension zone at the cracking location under the action of bending moment stops
working, and all the tension is resisted by the reinforcement. Nevertheless, the concrete between
two cracks in the tension zone is still functioning, which leads to the variation of sectional
flexural rigidity and neutral axial. The flexural rigidity of the cracked slab is decreased sharply.
A thinner slab with a higher concrete grade will have a smaller flexural rigidity after cracked.
The flexural rigidity used for the load stress calculation is the one with the supporting
layer’s full section sharing the stress. However, during the design the concrete in the tension
zone is supposedly out of operation completely, from which some error will occur and there is a
need for revision.
According to the elastic foundation beam theory, the bending moment of the foundation
beam under the concentrated load (train load) is directly related to the coefficient of elasticity of
the foundation and the flexural rigidity of the foundation beam. The bending moment of the
foundation beam is directly proportional to the 1/4 power of its flexural rigidity. Similarly, the
bending moments of the slab caused by temperature gradient and foundation deformation are
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
113
directly proportional to its flexural rigidity. Thus, the correction factor of the bending moment
caused by train load, temperature gradient and foundation deformation can be obtained.
VIII. CONCLUSIONS AND SUGGESTIONS
Ballastless track, with the merits of good ride comfort, high stability and little maintenance,
has become the main type of the rail structure. The design concepts of the ballastless track are
different in different countries: the factors considered in design and the calculation methods
vary greatly with each other. This paper has summarized and analyzed calculation and design
methods for ballastless track in the world. Based on the re-innovation research results of the
ballastless track in China, relatively general design concepts and methods for ballastless track
were put forward tentatively, which guided the design of ballastless track on the SuiningChongqing test section, the Wuhan-Guangzhou passenger dedicated line, the Lanzhou-Urumchi
No.2 double line, as well as the reference diagram design of the slab track and the double block
track.
Although, a type of structure with little maintenance, the ballastless track has many
conditions during operation. Therefore, the design theory of ballastless track still needs further
study. The future work may involve the following:
(1) Research on the fatigue properties under the coupling action of train and temperature
load. Train load and temperature load are two kinds of loads repeatedly acting on the ballastless
track. The statistical characteristics of train load and temperature load, especially the fatigue
properties under different loads and their coupling action, should be studied to provide a basis
for predicting the fatigue life of ballastless track.
(2) Research on the durability of ballastless track. The ballastless track is a composite
structure composed of many kinds of materials. Under the combined action of environment and
train load, different kinds of materials have different degradation curves, and the damage in one
component will influence the durability of the whole structure. Therefore, a systematic method
should be applied to the durability research of ballastless track, to realize a design concept of
little maintenance.
(3) Research on long-term dynamic properties. Under the long-term combined action of
train load induced vibration and natural environment, the function of the components of
ballastless track is likely to degrade gradually. Consequently, the dynamic characteristics of
ballastless track, and the safety and stability of train will be influenced. An analysis model of
ballastless track with damage should be established for research on the long-term dynamic
properties of ballastless track.
(4) Research on maintenance mechanics. At the beginning of the construction of ballastless
track, some conditions have already occurred because of the errors in design and construction.
Thus far, there is lack of a systematic, intensive study on the causes of diseases and the
114
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
countermeasures. The intensive study on disease mechanism, maintenance standard,
maintenance time, maintenance method, and the influence of maintenance on track and train
will lay a good foundation for the maintenance work of ballastless track.
ACKNOWLEDGEMENTS
This work was supported by the National Natural Science Foundation of China
(No. 51008258), and the Fundamental Research Funds for the Central Universities
(No. SWJTU09BR038).
References
[1]. Y.X. Wei, Y.J. Qiu, Research on the subgrade surface stiffness of ballastless track of high-speed
railway, Journal of Railway Engineering Society, 2010(7): 15-20 (in Chinese).
[2]. L. Bernhard, Developments in road pavement construction and railway track technology for a
sustainable surface transport infrastructure, In: The Emerging Frontiers of Transportation and
Development in China, Chengdu, 2009.
[3]. D.Z. Zheng, D.C. Feng, Mechanics of Layered Elastic System, Harbin: Harbin Institute of
Technology Press, 2001: 126-133 (in Chinese).
[4]. Ando Satoshi, Loading-deformation characteristics of slab track with temperature variation, Railway
Summary Report, 1989, 10(3): 2-9.
[5]. Q.C. Wan, Y.G. Lu, Stress Calculation of New Sub-rail Foundations, Beijing: China Railway
Publishing House, 1987: 15-45 (in Chinese).
[6]. C. Esveld, Modern Railway Track, 2nd ed., Zaltbommel: MRT Productions, 2001: 71-73.
[7]. Y.X. Liu, X.F. Chen, Two analytical methods of slab track structures. Urban Mass
Transit 2007(6) 32-34,66 (in Chinese).
[8]. C.Y. Qi, Research of soil subgrade slab track structures strength calculation, Railway Standard
Design, 2006(2): 26-28 (in Chinese).
[9]. L. Sun, Double-block ballastless track design of Wu-Guang passenger railway, Railway Standard
Design, 2006(Sup.): 155-158 (in Chinese)
[10]. S. Yoshihiko, New Track Mechanics, trans. Y. Xu, Beijing: China Railway Publishing House, 2001:
238-240 (in Chinese).
[11]. J. Eisenmann, G. Leykauf, Slab track for railways, In: Concrete Calendar 2000, Berlin: Verlag Ernst
& Sohn, 2000: 10-15 (in German).
[12]. J.J. Fan, Modern Railway Track, Beijing: China Railway Publishing House 2000: 43-51 (in
Chinese).
[13]. J. Eisenmann, The dimension of a type of slab track, Railway Engineering, 1991, 42(3): 116-118,
120-122, 124 (in German).
[14]. Q.D. Yue F.X. Jiang, X.G. Li, A study on mechanical model of two-level crossing superposed beam
system for section track structure, Railway Standard Design, 2002(12): 1-4 (in Chinese).
[15]. S.Q. Li, S.J. Duan, J.W. Wang, et al., Track structure model of elastic supported grillage girder,
Engineering Mechanics, 2001(Sup.): 508-513 (in Chinese).
[16]. J.W. Wang, S.J. Duan, S.Q. Li, High-speed railway slab track structure static analysis, Engineering
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
115
Mechanics, 2002(Sup.): 290-293 (in Chinese).
[17]. C.X. Liu, W.M. Zhai, Elementary research on problem of slab’s intensity by means of finite element
analysis, Journal of Railway Engineering Society, 2001(1): 24-26 (in Chinese).
[18]. L. Gao, M.N. Ma, D.M. Wang, Mechanic characteristic research of ballastless track structures on
bridge with linear motion actuator delivery system, Railway Standard Design, 2007(7): 5-7 (in Chinese).
[19]. C.X. Li, C.H. Li, Z.H. Kou, Mechanics analysis of slab track on soil foundation, Railway
Engineering, 2005(7): 88-90 (in Chinese).
[20]. P.R. Zhao, X.Y. Liu, Dynamic characteristic analysis and parameter study of slab track, Railway
Engineering, 2004(5): 48-50 (in Chinese).
[21]. P.R. Zhao, Y.A. Zhang, X.Y. Liu, et al., Beam-plate model on the elastic foundation of ballastless
track, China Railway Science, 2009(3): 1-4 (in Chinese).
[22]. J. Ren, Preliminary research and design of slab track, Railway Standard Design, 1996(6): 25-28 (in
Chinese).
[23]. Q.C. Wang, Design and Construction of Slab Track, Chengdu: Southwest Jiaotong University Press,
2002: 52-56 (in Chinese).
[24]. C.B. Cai, P. Xu, Dynamic analysis of key design parameters for ballastless track of high-speed
railway, Journal of Southwest Jiaotong University, 2010, 45(4): 493-497 (in Chinese).
[25]. P. Xu, C.B. Cai, Theoretical calculation of the supporting stiffness of the subgrade surface in
ballastless track and its application, China Railway Science, 2010(1): 21-25 (in Chinese).
[26]. P.R. ZHAO, X.Y. LIU, Reduced elastic modulus of cracked support layer in twin-block ballastless
track, Journal of Southwest Jiaotong University, 2008, 43(4): 459-464 (in Chinese).
[27]. Rail. ONE GnmH, Rheda 2000 Crack width calculation according to DIN 1045-1, Ingolstaedter, 2005.
[28]. P.R. Zhao, X.Y. Liu, Thermal stress calculation method research of continuous slab, Railway
Standard Design, 2008(10): 6-8 (in Chinese).
[29]. P.R. ZHAO, X.Y. LIU, Thermal stress calculation method and parameter analysis of continuous slab,
Railway Engineering, 2008(11): 81-85 (in Chinese).
[30]. S.R. Wang, Study on Temperature Stress for Ballastless Track Slab [Dissertation]. Chengdu:
Southwest Jiaotong University, 2007 (in Chinese).
[31]. R.S. Yang, Experimental report of track static test on Jialing river bridge, Chengdu: Southwest
Jiaotong University, 2007: 21-25 (in Chinese).
[32]. J. Eisenmann, G. Leykauf, Concrete Carriageways, Munchen: Ernst&Sohn, 2003: 22-59 (in
German).
[33]. Japan Railway Construction Public Corporation Morioka Branch Office, Design Calculation of RC
Subgrade En-gineering, t=300 with Detail Design of Northeast Trunk Line, RC Subgrade Engineering,
Japan, 1999.
[33]. A. Faeh, M. Gloor, T. Gerber, Optimised ballastless track systems, In: IABSE Symposium,
Antwerp, Belgium, August 2003: 52-60.
[34]. C. Esveld, V. Markine, Use of expanded polystyrene (EPS) sub-base in railway track design, In:
IABSE Symposium, Antwerp, Belgium, August 2003: 14-15.
[35]. J.M. Zwarthoed, V.L. Markine, C. Esveld, Slab track design: flexural stiffness versus soil improvement,
In: Proceedings of Rail-Tech Europe 2001 Conference, Utrecht, April 2001: 1-22.
[36]. V. Henke, H. Falkner, SFRC for jointless railway tracks, In: IABSE Symposium, Antwerp, Belgium,
August 2003: 46-51.
[37]. T. Takahashi, F. Sekine, T. Horiike, et al., Study on the applicability of short fiber reinforced
concrete to precase concrete slabs for slab track, Quarterly Report of RTRI, 2008, 49(1): 40-46.
[38]. Japan Railway Construction Public Corporation Morioka Branch Office, Design Calculation of
Northeast Trunk Line, Slab Design of PRC Structure A-55C, Japan, 1999♦
116
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
AN EFFICIENT METHOD
FOR VIETNAM LICENSE PLATE LOCATION
MAI VINH DU
DUOQIAN MIAO
RUIZHI WANG
Department of Computer Science
and Technology Tongji University,
Shanghai 201804, China
LE HUNG LAN
Electrical & Electronic Department,
The University of Transport
and Communications,
Hanoi, Vietnam
Summary: In the Automatic License Plate Recognition (ALPR) system, the License Plate
Location (LPL) is the key step before the final recognition. This paper proposed a LPL
algorithm for Vietnam license plates, which combined pre-processing, morphology open &
dilation operation, measure properties of image regions to find candidates and then finding
the license plate angle & rotating license plate. The proposed algorithm consists of three main
modules: Pre-processing (convert RGB image to grayscale image, image binarization use
Otsu method), License plate candidates location (morphology open to remove noises &
dilation operation, measure properties of image regions to find candidates), License plate
exactly location (finding the license plate angle & rotating license plate, cut exactly license
plate region). Experiments have been implemented on 350 Vietnam license plate images taken
from various scenes, including diverse angles, different lightening conditions (night and day),
reflected light and the dynamic conditions. The efficiency of processing of the proposed
algorithm is improved and average rate of accuracy of the license plate locating is 98.64%.
Keywords: License plate location, mathematic morphology, image converting, Ostu
method, image rotating, image measuring.
I. INTRODUCTION
Automatic License Plate Recognition (ALPR) is an important part of Intelligent
Transportation System (ITS) and can be utilized in many ITS applications such as: traffic
management system, traffic information system, traffic volume control system, none-stop toll
collection system & one-stop toll collection system, information traveling system, weigh in
motion system, car-park system, and so on. Usually, an ALPR system consists of three parts: 1)
license plate location, 2) character segmentation, 3) character recognition. Among these three
parts, the most important and basic is to license plate location step, which directly affects
system’s overall accuracy. There are many techniques have been proposed to address vehicle
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
117
license plate location since the 90’s decade, and a number of techniques have been proposed,
such as the methods base on histogram and mathematical morphology [1-9], wavelets transform
[10], texture feature analysis [11], edge detection analysis [12], color feature [13], region-based
segmentation [14]. But most previous works have in some way restricted their working
conditions, such as limiting them to indoor scenes, fixed illumination, prescribed driveways,
and these methods only implemented for their image data base, so it may be not good for
Vietnam license plate.
In this paper, we had considered for the specific characteristics of Vietnam license plates,
we had implemented for the vehicle images, which obtained from the actual system, these
vehicle images are very different background, different illumination, different license angles,
different sizes and types of license plates, different dimensions from camera to vehicles,
different light conditions in Vietnam environment. This paper put forward a method for the
Vietnam license plate location, the proposed algorithm consists of three main modules: Preprocessing (convert RGB image to grayscale image, image binarization use Otsu method),
License plate candidates location (morphology open & dilation operation, measure properties of
image regions to find candidates), License plate exactly location (finding the license plate angle
& rotating license plate, cut exactly license plate region). The rest of the paper is organized as
follows: section 2 introduce Feature of Vietnam license plates, section 3 shows The Algorithm
of Vietnam license plate location, section 4 shows the experiment results and section 5 are the
conclusions and lasts are acknowledements and the references.
II. FEATURE OF VIETNAM LICENSE PLATES
Based on the "Circular 06/2009/TT-BCA-C11 regulations on vehicle registration issued by
the Ministry of Public Security" issued on 11/03/2009, according to that, if we divide types of
license plate in color, the license plates (LP) of Vietnam have four types: Type 1 - the license
plate with white background and black characters & numerals, Type 2 - the license plate with
blue background and white characters & numerals, Type 3 - the license plate with red
background and white characters & numerals, Type 4 - the license plate with the yellow
background and black characters & numerals. In this paper, we only consider to locate for Type
1, this category has large number, over 90% of vehicles in Vietnam, which the license plate
with white background and black characters & numerals. Numerals are from 0 to 9 and
characters use a serial symbol is one of the 15 digits F, H, K, L, M, N, P, R, S, T, U , V, X, Y,
Z. This type also divided in two subtype (one-row LP and two-row LP), the shaps and
dimensions of this type such as table 1.
Table 1. Shaps and dimensions of Vietnam license plates of Type 1
Type
Shape
Dimension
One-row
LP
Two-row
LP
118
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
III. THE ALGORITHM FOR VIETNAM LICENSE PLATE LOCATION
3.1. Flowchart of the proposed algorithm
This paper combined pre-processing, morphological operations (open operation, dilate
operation) [1-9], measure properties of image regions to find candidates, last is cut exactly
license plate region, the flowchart such as figure 1.
Fig 1. Flowchart of the proposed algorithm for Vietnam License Plate Location
3.2. Pre - Processing
The performance of the ALPR system is heavily dependent on the environmental set-up,
such as the quality of cameras, light conditions, and other forms of controlling the environment.
Since the images of license plate are acquired from the various complicated background, e.g.
contrast, illumination, blurring, dirty, sizes and weather conditions, it is difficult for LPL work
especially when the acquired images are with thick shades and in low contrast or bad quality [7,
11, 12], etc. So the vehicle images pre-processing before LPL work in the ALPR system is
necessary.
3.2.1. Convert RGB image to Grayscale image
First one must obtain the values of its red, green, and blue (RGB) primaries in linear
intensity encoding, by gamma expansion. Then, add together 30% of the Red value, 59% of the
Green value, and 11% of the Blue value (these weights depend on the exact choice of the RGB
primaries). We converts RGB values to grayscale values by forming a weighted sum of the R,
G, and B components, with 8 bits per pixel, which allows 256 different intensities. The effective
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
119
luminance of a pixel is calculated by the formula 1. Figure 2(a) is origin RGB image, the output
grayscale image is figure 2(b).
(1)
(a)
(b)
Fig 2. a) Origin RGB image; b) Grayscale image
3.2.2. Adjust grayscale image intensity
We should adjust old intensity values of grayscale image to new values, the image is
weighted toward lower (darker) output values. Intensity adjustment is image enhancement
techniques are used to improve an image, where "improve" is sometimes defined objectively,
and sometimes subjectively. For intensity images, the n bins of the histogram are each half-open
intervals of width A/(n-1), the pth bin is the half-open interval.
(2)
Where x is the intensity value. The scale factor A depends on the image class, A is 255 if
the intensity image of class uint8. Figure 3(a) is histogram of grayscale image, the figure 3(b) is
histogram of adjusted grayscale image intensity.
(a)
(b)
Fig 3. a) Histogram of grayscale image; b) Histogram of adjusted grayscale image intensity
3.2.3. Image binarization using Otsu Method
Now, we convert adjusted image grayscale image in figure 4(a) to binary image based on
threshold, we threshold a gray-level image to binary image, a simple but effective tool to
separate objects from background flowing formula.
(3)
Where, T is some global threshold. Global image threshold using Otsu's method,
computing a global threshold (level) that can be used to convert an intensity image to a binary
image, level is a normalized intensity value that lies in the range [0, 1], which chooses the
threshold to minimize the intraclass variance of the black and white pixels. Otsu Method120
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
Principle - Algorithm as following:
•
Define the within-class variance:
(4)
Where
[0, L-1] the range of intensity level,
is the variance of the pixel in the background
(below threshold,
is the variance of the pixel in the foreground, above threshold).
•
Easier for the between-class variance:
(5)
By simplifying
(6)
Where
•
Easier using simple recurrence relations
(7)
(8)
(9)
(10)
From σ2 and μ, we define optimal threshold T, and then we convert to binary image
following formula (3). Figure 4(b) is image after convert to binary image, based on threshold
(Otsu's method).
(a)
(b)
Fig 4. a) Adjusted image grayscale image, b) Binary image based on threshold (Otsu's method)
3.3. License Plate Candidates Location
3.3.1. Morphological opening operation
The opening serves in image processing as a basic workhorse of morphological noise
removal. Morphological opening [3] used to remove small objects (connected component) from
the foreground (usually taken as the dark pixels) of an image, placing them in the background.
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
121
The opening of A by B is obtained by the erosion of A by B, followed by dilation of the
resulting image by B.
(11)
Where
and
denote erosion and dilation. We use this method for binary image to
improve the binary image, we removes objects binary image based only on their pixel area
from a binary image all connected components (objects) that have fewer than 2000 pixels
(removed noises, which are not license candidates), producing another binary image. The basic
steps of this method are 1) determine the connected components, 2) compute the area of each
component, 3) remove small objects. After implement we have figure 5(a).
(a)
(b)
Fig 5. a) After morphologically opening to remove noises, b) After morphologically dilation by a SE
3.3.2. Morphology dilation operation
Dilation operation used to increase objects in the image, definition for binary images, the
positions where a given structure element fits. The dilation of A by the structuring element (SE)
B is defined by.

(12)
Dilate is a function also known by the names "grow", "bolden", and "expand". It turns on
pixels which were near pixels that were on originally, thereby thickening the items in the image.
Algorithm of dilate operation is automatically takes advantage of the decomposition of a
structuring element object (if it exists), also when performing binary dilation with a structuring
element object that has a decomposition, it is automatically uses binary image packing to speed
up the dilation. Figure 5(b) is image after dilate operation.
3.3.3. Measure properties of cadidate regions (connected components)
From binary image obtained in the figure 5(b), we will locate license plate candidate by
two steps as following.
Step 1 - Convert numeric values to logical: After convert figure 4 (b) to logical values, we
have a logical image where is a logical array that is the same size as figure 5(a).
Step 2 - Measure properties of image regions (connected components): We measures a set
of properties for each connected component (object) in the binary image (the image is a logical
array). The fields of the structure array denote different properties for each region, properties
122
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
can be a comma-separated list of strings, a cell array containing strings, the actual number of
pixels in the region measured by shape measurements. The smallest rectangle containing the
region, a 1-by-Q *2 vector, where Q is the number of image dimensions, where is in the form [x
y] and specifies the upper-left corner of the bounding box, is in the form [x width y width] and
specifies the width of the smallest rectangle containing the region along each dimension.
Selecting the license plates from the candidates, we chosen area is the deepest region in the
frame which has properties of the license plate area (min area of LP and max area of LP, max
ratio and min ratio between height and width of LP, max height and min height, max width and
min width). And then, finding the components which are at the depth "depth", set the
coordinates of the supposed license plate region. After that a license plate candidate is cut from
theses coordinates such as figure 6.
Fig 6. License plate cadidate obtained by measure properties of cadidate regions
3.4. License plate exactly location
3.4.1. Finding the LP angle & rotating LP
In fact, the license plates usually not perpendicular in the images, so we need determine the
angle of the plate, and then if detected an angle we will rotating LP.
Finding the LP angle: The image contains parallel lines or at least one large line. Applying
the Radon transform on an image f(x,y) for a given set of angles can be thought of as computing
the projection of the image along the given angles. The result is a new image R(θ,ρ), this is
depicted in figure 7. This can be written mathematically by defining
(13)
After that, the Radon transform can be written as:
(14)
Where δ(.) is the Dirac delta function. There are two distinct Radon transforms (the source
can either be a single point or it can be a array of sources). In case a array of sources, the source
and sensor contrapment is rotated about the center of the object. For each angle, the density of
the matter the rays from the source passes through is accumulated at the sensor. This is repeated
for a given set of angels, usually from ε [0,180]. The angel 180 is not included since the result
would be identical to the angel 0, as figure 7. The Radon transform is a mapping from the
Cartesian rectangular coordinates (x,y) to a distance, also known as polar coordinates. Radon
transform algorithm includes some step as following: The Radon transform of an image is the
sum of the Radon transforms of each individual pixel. The algorithm first divides pixels in the
image into four subpixels and projects each subpixel separately. Each subpixel's contribution is
proportionally split into the two nearest bins, according to the distance between the projected
location and the bin centers. If the subpixel projection hits the center point of a bin, the bin on
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
123
the axes gets the full value of the subpixel, or one-fourth the value of the pixel. If the subpixel
projection hits the border between two bins, the subpixel value is split evenly between the bins.
Fig 7. Radon transform to find LP angle
After finding the LP angle, we have a image with lines as figure 8(a).
(a)
(b)
Fig 8. a) Image after found angle of LP, b) Image after rotating
Rotating LP: Since angle of license plate determined by Radon transform, we can
implement rotating this image, using the interpolation method specified by bilinear
interpolation, after rotating, we obtained figure 8(b).
3.4.2. Cut exactly license plate region
We find the coordinates of the LP rectangle inside the image, find contours for the sum of
the lines and of the columns in the image. Find contours are find the x-coordinates respectively
of the first point at the left and the first point at the right of the vector which is superior or equal
to the average of the vector. This permits to delimit the plate eliminating noises around it. We
rotate the image using the interpolation method specified by bilinear interpolation. We have a
four-element position vector [xmin ymin width height] that specifies the size and position of the
crop rectangle. After cut exactly license plate region, we have the result as figure 9.
Fig 9. Exactly license plate region
IV. EXPERIMENT RESULTS
We implemented experiment with PC Intel(R) Core(TM)2 Duo CPU T7250 @2.00GHz,
RAM 1.00 GB, Windows Vista Version 6.1 (Build 7600) 32-bit Operating System and
MATLAB Version 7.8.0.347 (R2009a). We implemented test for 350 vehicle images, which
obtained from the actual system, these vehicle images are very different background, different
illumination, different license angles, different size and type of license plates, different
dimensions from camera to vehicles, different light conditions in Vietnam environment. There
124
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
are two type of the size of the images in RGB true-color image was tested (800x600 pixels and
768x288 pixels). The results of the our method and the previous works show in table 2. The
average rate of accuracy is 98.64%, comparision with previous methods, our results are more
axactly, some examples of the experiments show in figure 10.
Table 2. Experiment results of our method and previous works
Experiment results of our method
Previous works
Average
Type
Total
Rate of
Accuracy Accuracy Accuracy
Accuracy of
rate of
of LP
image accuracy
of [2]
of [3]
of [4]
[6]
accuracy
One-row
200
98.97%
LP
98.64%
96.50%
97.00%
96.00%
83.50%
Two-row
LP
150
98.30%
Fig 10. Some examples of our experiment results
V. CONCLUSIONS
This paper put forward a LPL algorithm for Vietnam license plates, the proposed algorithm
consists of three main modules: Pre-processing (convert RGB image to grayscale image, image
binarization use Otsu method), License plate candidates location (morphology open & dilation
operation, measure properties of image regions to find candidates), License plate exactly
location (finding the license plate angle & rotating license plate, cut exactly license plate
region). Experiments have been conducted for this algorithm, which implemented test for 350
vehicle images, which obtained from the actual system, these vehicle images are very different
background, different illumination, different license angles, different size and type of license
plates, different dimensions from camera to vehicles, different light conditions in Vietnam
environment. The algorithm can quickly and correctly detect the region of license plate and the
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
125
average rate of accuracy is 98.64%, comparision with previous methods, our results are more
axactly. From the result of the experiment, we can see the proposed approach is robust. But
there are still some images failed in the experiment, our algorithm still needs further research.
VI. ACKNOWLEDEMENTS
This work was supported by National Natural Science Foundation of China (Serial No.
60970061, 61075056). And supported by Research team of Electrical & Electronic Department,
The University of Transport and Communications, Hanoi, Vietnam.
References
[1]. Dubey, P., “Heuristic approach for license plate detection,” in Annual IEEE INDICON, pp. 366-370,
2005.
[2]. Mahini, H.; Kasaei, S.; Dorri, F., “An Efficient Features - Based License Plate Localization
Method,” in 18th International Conference on Pattern Recognition, vol. 2, pp. 841-844, 2006.
[3]. W. K. I. L Wanniarachchi, D. U. J. Sonnadara and M. K. Jayananda, “Detection of License Plates of
Vehicles,” in Proceedings of the Technical Sessions, vol. 23, pp. 13-18, 2007.
[4]. Humayun K. Sulehria, Ye Zhang, Danish Irfan, “Mathematical Morphology Methodology for
Extraction of Vehicle Number Plates,” in INTERNATIONAL JOURNAL OF COMPUTERS , vol. 1, pp.
69-73, 2007.
[5]. Gisu Heo, Minwoo Kim, Insook Jung, Duk-Ryong Lee, II-Seok Oh, “Extraction of Car License Plate
Regions Using Line Grouping and Edge Density Methods,” in International Symposium on Information
Technology Convergence, vol. , pp. 37-42, 2007.
[6]. Faradji, F., Rezaie, A.H., Ziaratban, M., “A Morphological-Based License Plate Location,” in IEEE
International Conference on Image Processing, vol. 1, pp. 57-60, 2007.
[7]. Haiqi Huang, Ming Gu, Hongyang Chao, “An Efficient Method of License Plate Location in NaturalScene Image,” Fifth International Conference on Fuzzy Systems and Knowledge Discovery, vol. 4, pp.
15-19, 2008.
[8]. Hui Li, Pan Gao, Dongxiu Wang, Wangming Chen, “Study on Mathematical morphology based
vehicle license plate location,” in 2nd International Conference on Power Electronics and Intelligent
Transportation System (PEITS), vol. 2, pp. 116-119, 2009.
[9]. Wei Pan, Rong An, “Morphology and auto -correlation based method of fast locating vehicle license
plate,” in 2nd International Conference on Advanced Computer Control (ICACC), vol. 3, pp. 116-119,
2010.
[10]. Feng Yang; Fan Yang, “Detecting license plate based on top-hat transform and wavelet transform,”
in International Conference on Audio, Language and Image Processing, vol. , pp. 998-1003, 2008.
[11]. Jun Kong; Xinyue Liu; Yinghua Lu; Xiaofeng Zhou, “A novel license plate localization method
based on textural feature analysis,” in Fifth IEEE International Symposium on Signal Processing and
Information Technology, vol., pp. 275-279, 2005.
[12]. Xing Yang; Chaochao Huang; Hua Yang;, “Research on Adaptive Preprocessing License Plate
Location,” in The 9th International Conference for Young Computer Scientists, vol. , pp. 764-768, 2008.
[13]. Xiaoqian Liu; Weiqiang Wang; Tingshao Zhu;, “A Method Based on Character Edge Color for
Quick Locating Vehicle License Plate,” in 20th International Conference on Pattern Recognition (ICPR),
vol. , pp. 3232-3235, 2009.
[14]. Wenjing Jia, Huaifeng Zhang, Xiangjian He, “Region-based license plate detection,” in Journal of
Network and Computer Applications, vol.30 , pp. 1324-1333, 2006♦
126
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
ANALYSIS ON SETTING AIRSHAFT AT MID-TUNNEL
TO REDUCE TRANSIENT PRESSURE VARIATION
YINGXUE WANG
BO GAO
CHAO ZHANG
XUZHOU HE
School of Civil engineering,
Southwest Jiaotong University,
Chengdu 610031, China
Summary: Setting airshaft is one of the efficient methods to control the aerodynamic
effects during the train traveling through a tunnel. Using numeral simulation method, this
paper analyzed the difference between setting airshaft and without setting airshaft in transient
pressure in cabin. After setting airshaft, the strength of pressure fluctuation in cabin can be
reduced more than 40%. By analyzing the rule of airshaft alleviating pressure fluctuation, the
expression to determine the optimal airshaft position was deduced.
Key words: Tunnel; airshaft; high-speed train; transient pressure.
I. INTRODUCTION
When a high-speed train passes through a tunnel, compression wave and expansion wave
will be generated and transmitted to and fro, resulting in micro-compression wave radiation at
tunnel exit and pressure fluctuation in cabin [1]-[3]. To deal with this, considerable efforts [4][8] have been made to alleviate the aerodynamic effects in the process of railway passing
through tunnel, such as improving cabin’s airproof parameter, enlarging the cross section of
tunnel, building hood at tunnel entrance. A desirable structure for a long tunnel is in the form of
two shorter tunnels connected by pressure relief ducts [9]. Besides, airshaft is generally installed
to reduce pressure intensity [10]-[11].
To controlling aerodynamic effects, most studies [5], [6], [8] and [12] focused on microcompression wave at tunnel exit. In fact, the pressure fluctuation in cabin will reduce the
comfort and even jeopardize passengers’ health; thus ameliorating transient pressure in cabin is
an important project.
When a high speed train enters into a tunnel, the compression wave is generated ahead of
the train due to the piston-like action of tunnel entry motion. The compression wave is
discharged to the atmosphere as the form of a pulse-like wave, which is called the micropressure wave. The compression wave also arouses pressure fluctuation in cabin, which is called
transient pressure. To relieve the intensity of the booming noise, many methods have been put
forward, such as the installation of tunnel entrance hood, side tunnel branches, and shelters with
slits linking adjacent tunnels [13]. Sealing carriage can realize the goal of reducing transient
pressure and promoting passenger comfort. In practice, however, complete sealing of a train
body is not possible because a train should have a lot of air conditioning devices, ventilation
systems, etc. Thus, the large amplitude pressure variations on the train body during traveling
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
127
inside tunnel can penetrate into passenger’s room, often leading to ear discomfort in passengers
[14[–[16]. This phenomenon will be more serious with the increase in train speed. The
passenger’s ear discomfort inside train is, in general, associated with the magnitude of pressure
variation, the rate of pressure variation, the sign of pressure variation, etc. [17]–[19]. In real
high-speed railway trains, a special ventilation system is usually adopted to alleviate the
pressure variations. This system controls the flow rate supplied into and exhausted from train,
using a fan blower, according to the pressure variations occurring inside train. These pressure
variations are proportional to the square of train speed [18]-[19]. The ventilation system may not
be enough to reduce the pressure variations as the train speed increases. In order to alleviate the
pressure variation inside train, several control methods have been investigated using a damper
system, a continuous ventilation system, and a continuous ventilation control system [18-19].
All of these methods are to control the magnitude and the rate of the pressure variation
occurring inside train.
Apart from solving pressure variation from the train point of view, building subsidiary
construction in tunnel can also realize the goal. In this paper, using numeral simulation method,
the compression wave transportation process in tunnel will be analyzed, and the transient
pressure variation in cabin in the conditions of tunnel with airshaft and without airshaft will be
compared.
II. GOVERNING EQUATIONS
During the course of a high-speed train passing through tunnel, the boundary condition is
changed with time. The dynamic mesh model is applied to simulate this process.
The integral form of the conservation equation for a general scalar, φ, on an arbitrary
control volume, V, with moving boundaries is written as [20]
d
dt
∫
∫
V
∂V
ρφ dV +
∫
ρ φ (u − u g ) ⋅ d A =
∂V
Γ∇ φ ⋅ dA +
∫
V
Sφ dV ,
(1)
u
Where ρ is the fluid density; u is the flow velocity vector; g is the grid velocity of the
moving mesh; Γ is the diffusion coefficient;
the boundary of the control volume V.
Sφ
is the source term of φ; ∂V is used to represent
Using a first-order backward difference formula, the time derivative term in (1) is written as:
d
( ρφV ) n +1 − ( ρφV ) n
ρφ dV =
,
∫
dt V
Δt
(2)
Where n and n+1 denote the respective quantity at the current and next time level. The
(n+1)th time level volume Vn+1 is computed from
V n +1 = V n +
dV
Δt ,
dt
Where dV/dt is the volume time derivative of the control volume. In order to satisfy the
grid conservation law, the volume time derivative of the control volume is computed from
128
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
n
f
dV
= ∫ ug ⋅ dA = ∑ ug , j ⋅ A j ,
∂V
dt
j
Where nf is the number of faces on the control volume and
dot product
ug , j ⋅ A j
δVj
is the j face area vector. The
on each control volume face is calculated by
ug , j ⋅ A j =
Where
Aj
(3)
δVj
Δt
,
(4)
is the volume swept out by the control volume face j over the time step Δt .
Here, the large eddy simulation model (LES) is chosen. To model a free-stream condition
at sites far from tunnel entrance and exit, pressure-far-field boundary conditions are adopted.
The transmission medium is assumed as ideal gas.
III. AIRSHAFT EFFECTS
3.1. Calculation parameters
The parameters of the railway, tunnel and airshaft are shown in table 1. To analyze the
effect of the shaft area on peak pressure at train body, the calculation results under different
shaft open ratios will be given. The grid mesh at the tunnel entrance and train head is shown in
fig 1.
3.2. Calculation results
Without airshaft, the transportations of compression wave and expansion wave and
pressure fluctuation at train body are shown as fig. 2 and fig 3.
After setting shaft, the compression wave, expansion transportation process and pressure
fluctuation on railway body are shown as fig. 4 and fig 5.
Table 1. Calculation parameter
Condition
Tunnel
length LT /m
Tunnel
area AT /m2
Carriage parameter
Speed
V (km/h)
Area
(m2)
Shaft parameter
Length (m)
Shaft position
Open ratio
—
—
200 m to tunnel
entrance
14%
Ⅰ
ⅠⅠ
ⅠⅠⅠ
700
65
300
12.7
50
25%
Note: Shaft open ratio=shaft open area/tunnel free cross-sectional are
(a)
Fig 1. The grid meshes at the tunnel entrance and train head
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
(b)
129
Train
Distance to tunnel entrance
0m
4
150 m
P/kPa
0
4 Expansion
wave
0
350 m
-4
4
Tail
00
2
4
6
8
10
12
500 m
-4
Head
Compression
wave
700 m
t (s)
Fig 2. Compression wave transportation process in tunnel (without shaft)
2
P (kPa)
0
-2
-4
-6
0
2
4
6
t (s)
8
10
12
Fig 3. Pressure fluctuation at train body (without shaft)
Train
Distance to tunnel entrance
-4
4
0m
P (kPa)
2
0
-2
Shaft
150 m
-4
200 m
P (kPa)
2
Expansion
wave
0
-2
350 m
-4
tail
P (kPa)
2
0
-2
-4
0
Compression
2
wave
4
6
t (s)
head
8
10
12
700 m
Fig 4. Compression wave transportation process in tunnel (shaft open ratio=14%)
130
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
1
Shaft open ratio=14%
Shaft open ratio=25%
0
P (kPa)
-1
-2
-3
-4
-2
0
2
4
6
8
10
12
t (s)
Fig 5. Effect of shaft open ration on pressure fluctuation at train body (with shaft)
Tab 2. Calculation Result Comparison
Maximum pressure
on train body (kPa)
Maximum pressure at measurement point (kPa)
Working
condition
P150
P350
P500
positive
Maximum
pressure transient
negative kPa /s
kPa /3s
positive
negative
positive negative positive negative
Without
shaft
3.3
-2.9
3.2
-2.9
3.2
-2.7
0.3
-4.6
3.8
5.0
With shaft
3.2
-2.1
2.1
-3.2
2.2
-3.1
0.3
-3.4
3.0
3.1
The difference of aerodynamic effects with and without airshaft is shown in tab 2.
At the train body, the first peak pressure occurred the moment the train entered the tunnel.
And the first negative peak pressure appeared when the train encountered with expansion wave
converted from compression wave at tunnel exit. When compression wave passed through
airshaft, energy was released, the peak value of the compression wave fell. From P150 to P350,
the peak value of the compression changed from 3.2 kPa to 2.1 kPa. The reduction rate nearly
amounted to 40%.
Figs 3 and 5 show that the compression wave at the tunnel exit induced the negative
fluctuation at train body. By optimizing airshaft position to let the expansion wave pass through
airshaft before encountering carriage, the peak value of expansion wave would be reduced.
From this point of view, we proposed an expression to determine the optimal airshaft position:
lshaft + ltrain ltunnel + ltunnel − lshaft
>
,
V
c
Where, lshaft is shaft position to tunnel entrance, ltunnel is tunnel length, V is the speed of
railway, c is the speed of sound, c=340 m/s. In this case, the optimal airshaft position is: 700
m＞lshaft＞235 m.
When compression wave passed through airshaft expansion wave formed， and then
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
131
encountered train body, causing negative pressure in cabin. Comparing fig 2 with fig 4, it is
found that with airshaft, the moment first negative peak pressure appeared was ahead of the one
without shaft, and the absolute value of peak negative pressure was larger than the pressure at
2.3 s in the case of no shaft. This is because the expansion wave transmitted to tunnel entrance
and encountered with train body. When the train passed through airshaft, secondary
compression wave was also generated. The expansion wave from the tunnel exit was reduced
when meeting the secondary compression wave. In this way, the negative pressure on train body
was lessened. In this case, expansion wave met train before it passed through airshaft. The peak
negative pressure resulted from the expansion wave at the tunnel exit was almost equal the peak
negative pressure resulted from the expansion wave passing through the airshaft. This validates
that the choice of open area of the airshaft.
With the airshaft, the strength of pressure fluctuation was reduced, and the aerodynamic
effects in cabin were moderated. In the calculation example, after setting airshaft, the peak
pressure in three seconds, was reduced from 5 kPa to 3.1 kPa. The reduction rate is more that
40%.
IV. ENDING REMARKS
(1) The first negative peak pressure at carriage resulted from the train encountering with
expansion wave, which was converted from compression wave at tunnel exit.
(2) By optimizing the shaft setting, the efficiency of ameliorating aerodynamic effects can
be promoted.
(3) When a train passes through airshaft, secondary wave will be generated. The expansion
wave at the tunnel exit can be reduced when meeting the secondary compression wave.
(4) With an airshaft, the strength of pressure fluctuation is reduced, and the aerodynamic
effects in cabin are moderated.
ACKNOWLEDGEMENT
This paper is supported by the Fundamental Research Funds for the Central Universities
(SWJTU09CX009).
References
[1]. R.S. Raghunathan, H.D. Kim, T. Setoguchi, Aerodynamics of high-speed railway train, Progress in
Aerospace Sciences, 2002, 38(6-7): 469-514.
[2]. S. Mashimo, K. Iwamoto, T. Aoki, et al., Characteristics of compression wave generated by a high-
132
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
speed train entering tunnel, Engineering Sciences Reports, 1997, 18(4): 297-302.
[3]. S. Ozawa, Y. Moritoh, T. Maeda, et al., Investigation of pressure wave radiated from a tunnel exit,
Tokyo: Railway Technical Research Institute, 1976 (in Japanese).
[4]. S. Ozawa, T. Uchida, T. Maeda, Reduction of tunnel exit boom by hood at tunnel entrance, Tokyo:
Railway Technical Research Institute, 1977 (in Japanese).
[5]. S. Ozawa, T. Uchida, T. Maeda, Reduction of micro-pressure wave radiated from tunnel exit by hood
at tunnel entrance, Quarterly Reports of the RTRI, 1978(19): 77-83 (in Japanese).
[6]. S. Ozawa, Studies of micro-pressure wave radiated from a tunnel exit, Tokyo: Railway Technical
Research Institute, 1979 (in Japanese).
[7]. J.A. Fox, A.E. Vardy, The generation and alleviation of air pressure transients caused by the high
speed passages of vehicles through tunnels, In: Proc. 1st ISAVVT, BHRA, 1973.
[8]. M. Tastsuo, Reduction of micro-pressure wave radiated from tunnel exit by branched in tunnel,
Tokyo: Railway Technical Research Institute, 1977 (in Japanese).
[9]. A. Baron, M. Mossi, S. Sibilla, The alleviation of the aerodynamic drag and wave effects of highspeed trains in very long tunnels, Journal of Wind Engineering and Industrial Aerodynamics, 2001,
89(5): 365-401.
[10]. A.E. Vardy, The use of air shafts for the alleviation of pressure transients in railway tunnels, In:
Proc. 2nd ISAVVT, 1976.
[11]. Y.X. Wang, B. Gao, K. Su, et al., Experiment research on reducing aerodynamics effect
comprehensive measurement, Journal of Experiments in Fluid Mechanics, 2009, 23(1): 31-34.
[12]. Y.X. Wang, B. Gao, C.Q. Zheng, et al., Micro-compression wave model experiment on the highspeed train entering tunnel, Journal of Experiments in Fluid Mechanics, 2006, 20(1): 5-8.
[13]. R.S. Raghunathana, H.D. Kimb, T. Setoguchic, Aerodynamics of high-speed railway train, Progress
in Aerospace Sciences, 2002, 38(6-7): 469-514.
[14]. M. Schultz, H. Sockel, Pressure transients in railway tunnels, W. Schneider, H. Troger, F. Ziegler,
ed., Trends in applications of mathematics to mechanics, Harlow, 1989.
[15]. N. Komatsu, F. Yamada, The reduction of the train draft pressure in passing by each other, In: Proc.
World Congrence Railway Research, Tokyo: Railway Technical Research Institute, 1999.
[15]. R.G. Gawthorpe, Pressure comfort criteria for rail tunnels operations, A. Haerter, ed., Aerodynamics
and ventilation of vehicle tunnels, New York: Elsevier, 1991.
[16]. K. Sato, M. Ikada, M. Nakagawa, Effects of pressure changes on pain sensation of human ears,
RTRI JNR,1989, 3(3) (in Japanese).
[17]. Y. Zenda, Study on the ventilating system of Shinkansen vehicle by simulating the internal pressure,
RTRI JNR,1988, 2(12) (in Japanese).
[18]. Kobayashi, Suzuki Y, Akutsu K. Alleviating ear pains by controlling air pressure in ventilating
system of Shinkansen car. RTRI JNR 1990, 4(7) (in Japanese).
[19]. Fluent user’s guide, Fluent Inc. 2004-09♦
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
133
SERVICE LIFE ESTIMATION OF REINFORCED CONCRETE
STRUCTURES WITH CONSIDERING THE DAMAGE
OF CONCRETE COVER
TRAN THE TRUYEN
Department of Civil Engineering
University of Transport and Communications
Summary: This paper presents an estimation of the long-term durability of reinforced
concrete structures based on the rebar corrosion criteria. Service life is defined as the period
of time required for the chloride concentration on rebar surface to reach the critical chloride
level (Ccr). The corrosion rate of rebar is evaluated with considering the damage of concrete
cover. The results show that when concrete cover is damaged, the service life of concrete
structures is significantly reduced in comparison with the case, in which, concrete cover
remains intact. In various exposure conditions of marine environment including spray zone,
tidal zone, and atmospheric zone; the service life of concrete structures is different due to
different corrosion rates of rebar.
Key words: Concrete, corrosion, rebar, durability, damage, time, chloride, diffusion,
service life, reinforced concrete structures (RCS).
I. INTRODUCTION
Long-term durability of reinforced concrete structures (RCS) is affected by many
physicochemical process. These processes include the chloride attacks, carbonation
phenomenon, mechanical effects, etc. One of the most typical among them is the corrosion of
rebar due to the penetration of chloride into concrete. The latter becomes serious when concrete
cover is damaged in construction and in service due to mechanical effects. The damage of
concrete first makes increase the seepage capacity of concrete cover, the chloride ion is then
easily diffused into concrete causing damage of passive protection layer around rebar [8]. Once,
the rebar is no longer protected, they are exposed to the attack of oxygen and water and finally
becomes rusted.
The service life of RCS in this study is taken as the time from which the corrosion of rebar
occurs due to the diffusion of chloride into concrete; or more accurately, this is the time from
which, the chloride concentration C on the rebar surface reaches the critical value Ccr. The
deterministic model used for the calculation of the service life is based on Fick’s second law [7];
the service life is then represented as a function of the surface chloride concentration Cs,
chloride diffusion coefficient Kc, critical chloride level Ccr, and concrete cover thickness h. The
service life is also effected by environments conditions; in this study, the marine environment
conditions including spray zone, tidal zone, and atmospheric zone are considered. The
134
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
environment parameters in these zones included surface chloride concentration (Cs), chloride
diffusion coefficient (Kc) is introduced based the studies of Mangat & Molloy (1994) [6] and A.
Costa & J. Appeleton (1998) [3].
There is only the diffused damage state of concrete cover is considered in this study.
Because, when concrete damage becomes localized or fully damage (fracture), the corrosion of
rebar occurs immediately. Evaluation of RCS durability based on rebar corrosions criteria is no
longer significant.
Effect of damage time of concrete on the RCS durability is considered in two cases: (i)
Concrete is damaged just during the construction phases; (ii) Concrete is damaged after a period
of putting into service.
II. PERMEABILITY AND CHLORIDE DIFFUSION OF DAMAGED CONCRETE
The relationship between gas permeability and chloride diffusion of damaged concrete is
introduced based on the experimental results of Choinska & al (2008) [2]. After this research,
the evolution of chloride diffusion with permeability of damaged concrete is represented in fig1.
Fig 1. Relationship between chloride diffusion and gas permeability of concrete
(Choinska & al (2008))
In fig 1, Kv(d) = K is the permeability of damaged concrete; Kvo = Ko is the initial value of
permeability (permeability of intact concrete): Ko ≈ 10-17 m2; Demig = Kc is the chloride diffusion
coefficient into damaged concrete; and Demigo = Kco is the chloride diffusion coefficient into
intact concrete. The relationship between chloride diffusion and permeability of ordinary
damaged concrete (BO on fig.1) could be represented as a linear regression line with function as
flow:
Kc/Kco = 0.22(K/Ko) + 1
(1)
By replacing the permeability law of damaged concrete [10] in (1), we have:
Kc/Kco = 0.22[α.exp(βD)] + 1
(2)
With the considered concrete M30 [10], (2) can be rewritten as follow:
Kc/Kco = 0.22 exp(15.529D) + 1
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
(3)
135
11
Kc / Kco
9
7
5
3
1
0
0.05
0.1
0.15
0.2
0.25
D
Fig 2. Increase of chloride diffusion with diffuse damage of concrete
The variation of chloride diffusion with diffuse damaged of concrete is shown on
fig 2. We found that when the damage D < 0.1, the increase of chloride diffusion
coefficient is weak. After this critical value, the chloride diffusion coefficient increases
rapidly at reaches a values greater 10 times than the initial value.
III. SERVICE LIFE PREDICTION: NON-DAMAGED CONCRETE COVER
Concrete could be considered to be intact when the behavior of concrete are
elastic phase, i.e. no damage appears in concrete due to mechanical or temperature effects.
in
The model for evaluating the service life of RCS is developed from the equation
calculating the concentration of chloride on rebar surface based on the second Fick law
(RILEM 14 (2005) - Sara A.& E. Vesikari) [7]:
⎛
⎛ x ⎞⎞
⎟⎟
C x = C s ⎜1 − erf ⎜
⎜ 2 K t ⎟⎟
⎜
c ⎠⎠
⎝
⎝
(4)
Where Cx is the chloride concentration at the depth x of concrete cover; Cs is chloride
concentration on the surface of concrete structures; Kc is the chloride diffusion coefficient into
concrete; t is the considered time; erf is the error function.
The corrosion process of rebar starts when Cx = Ccr; at this moment, x = h (concrete cover
thickness), (4) can be rewritten as follow:
⎛
⎛ h ⎞⎞
⎟⎟
C cr = C s ⎜1 − erf ⎜
⎜ 2 K t ⎟⎟
⎜
c ⎠⎠
⎝
⎝
(5)
The formula (5) could be simplified by using the Parabola function [7]:
⎞
⎛
h
⎟
C cr = C s ⎜1 −
⎜ 2 3K t ⎟
c ⎠
⎝
2
(6)
The chloride diffusion coefficient Kc and the chloride concentration as concrete surface Cs
are time-depending factors.
136
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
To consider the variation of Kc, we can used the proposed model of Mangat & Molloy
(1994) [6]:
Kcot = Kco1.t-m
(7)
Where, Koct is the chloride diffusion coefficient after t year; Koc1 is the chloride diffusion
coefficient after first year; m is an empirical coefficient.
According to Mangat & Molloy (1994), values of Koc1 and m in different conditions of
marine environments are different:
+ The spray zone: Kco1 = 3.12 x 10-12 m2/s; m = 0.51;
+ The tidal zone: Kco1 = 5.32 x 10-12 m2/s; m = 0.60;
+ The atmospheric zone : Kco1 = 1.21 x 10-12 m2/s; m = 0.42.
Variation of surface chloride concentration Cs could be taken as the proposition of A. Costa
& J. Appeleton (1998) [3]:
Cst = Cso. tn
(8)
Where, Cso is the surface chloride concentration after first year, n is an empirical
coefficient.
Depending on the environment conditions, the values of Cso (% of concrete weight) and n
for ordinary concrete are different (A. Costa & J. Appeleton (1998) [3]):
+ The spray zone: Cso = 0.24; n = 0.47.
+ The tidal zone: Cso = 0.38; n = 0.37.
+ The atmospheric zone: Cso = 0.12; n = 0.54.
If we consider the variation with time of chloride diffusion coefficient and surface chloride
concentration, the formula (6) can be rewritten as follow:
⎛
h
C cr = C so t n ⎜1 −
⎜ 2 3K t 1-m
co
⎝
⎞
⎟
⎟
⎠
2
(9)
The minimum value of concrete cover thickness h to protect rebar from chloride attack is
finally calculated by the following formula:
⎛
C cr
h = 2 3K co t 1− m ⎜1 −
⎜
C so .t n
⎝
⎞
⎟
⎟
⎠
(10)
Fig 3 show the relationships between concrete cover thickness and the service life of RCS
(Ccr = 0.06 % concrete weight [7]). We found that when concrete cover thickness increases, the
service life is prolonged too. However, the evolution of the latter becomes slowly after a
threshold of h. The environment conditions has significant effects on the service life. Indeed, at
the atmospheric zone, the structure durability is the better; on the contrary, the tidal zone is the
most dangerous zone for the deterioration of structures. Taking the case, in which, the concrete
cover thickness equal to 6 cm: the service life of RCS is respectively about 80 years, 26 years,
and 18 years at the atmospheric zone, at the spray zone, and at the tidal zone.
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
137
10
h (cm)
8
6
Tid a l zo ne
4
S p ra y zo ne
2
A tm o s p he ric zo ne
0
0
20
40
60
80
100
120
t (y e a r)
Fig 3. Relation between concrete cover thickness and service life of RCS
IV. SERVICE LIFE PREDICTION: DAMAGED CONCRETE COVER
4.1. Concrete is damaged during construction phase
In this case, the chloride diffusion coefficient Kc is taken as the initial value Kco ≈ 5E-12
m /s. The surface chloride concentration Cs is also the initial value Cso ≈ 0.35. The service life
of RCS defined as the rebar corrosion criteria is then calculated by the following formula:
2
⎛
⎜
⎜
1
h
t =
⎜
12K co [ 0 . 22 exp( 15 . 529 D ) + 1] ⎜
C cr
⎜1− C
so
⎝
⎞
⎟
⎟
⎟
⎟
⎟
⎠
2
(11)
8
h (cm)
6
D = 0 .2 5
4
D = 0 .2 0
D = 0 .1 5
2
D = 0 .1 0
D = 0 .0 5
0
0
1
2
3
4
5
6
t (y e a r )
Fig 4. Effect of damage on the service life: concrete is damaged during construction phase
Fig 4 shows the influence of concrete cover damage level on the relationship between the
service life and the concrete cover thickness h. We found that according to a value of the latter,
the durability of structure decreases in consequence of the increase of damage level. The service
138
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
life is no longer significant when damage reaches the maximal value D = 0.25 (Threshold of
diffused damage).
In comparison with the case, concrete is intact, we realized that the service life of RCS is
strongly reduced even while the damage of concrete cover is still weak. Thus, the damage of
concrete cover occurs during construction phase is extremely dangerous for the long-term
durability of reinforced concrete structures.
4.2. Concrete is damaged after a period of putting structures into service
In this case, the chloride diffusion coefficient Kc is taken as the value just before the
beginning of the damage of concrete cover: Kc = Kcot; Kcot is calculated according to (7). The
surface chloride concentration Cs is also the value just before the damage of concrete Cs = Cst;
Cst is calculated by the formula (8).
In this study, we carried out an example of calculation with supposition that concrete cover
is damaged after 5 years of putting structure into service. The service life is then calculated by
the following formula:
2
⎛
⎞
⎜
⎟
⎜
⎟
1
h
t =
⎜
⎟ +5
−m
12K co .5 .[0.22 exp(15.529 D) + 1] ⎜
C cr ⎟
1−
⎜
C so 5 n ⎟⎠
⎝
(12)
Three areas including spray zone, tidal zone, and atmospheric zone are also considered as
above.
Fig 5, 6, 7 show the relationship between the service life (calculated according to (12)) and
the concrete cover thickness in different considered zones.
35
D=0.25
30
D=0.20
D=0.15
25
h (cm)
D=0.10
20
D=0.05
15
10
5
0
0
5
10
15
20
25
30
35
40
45
50
55
t (year)
Fig 5. Relation service life – concrete cover thickness: Concrete is damaged
after 5 years in service (spray zone)
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
139
60
D =0.25
50
D =0.20
D =0.15
h (cm)
40
D =0.10
30
D =0.05
20
10
0
0
5
10
15
20
25
30
35
40
45
50
55
t (ye ar)
Fig 6. Relation service life – concrete cover thickness: Concrete is damaged
after 5 years in service (tidal zone)
25
D=0.25
D=0.20
20
D=0.15
D=0.10
h (cm)
15
D=0.05
10
5
0
0
5
10
15
20
25
30
35
40
45
50
55
t (year)
Fig 7. Relation service life – concrete cover thickness: Concrete is damaged
after 5 years in service (atmospheric zone)
Results in figures 5, 6, 7 show that the damage state of concrete cover has significant
impacts to the service life of RCS. Indeed, with the same concrete cover thickness, when
damage increases, the long-term durability of RCS reduces rapidly. In the same meaning, to be
able to have an expected durability, the concrete cover thickness should be increased when
damage level increases. In comparison to the case that concrete is not damaged yet, the service
life of RCS greatly decreases from the time that the damage process of concrete occurs.
The service life of RCS based in the tidal zone is always lowest in comparison with the
value in atmospheric zone and spray zone.
III. CONCLUSIONS
The service life of reinforced concrete structures has been evaluated based on rebar
corrosion criteria due to the chloride diffusion into concrete. The service life according to this
criteria depends on the chloride diffusion coefficient Kc, the chloride concentration on concrete
140
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
surface Cs, and the concrete cover thickness h. When we suppose that the two first factors are
unchanging, the service life is very low and doesn’t match with the real observation. By
considering the variation of Kc and Cs with time, the calculated values of service life has
reached the experimental values.
The marine environment conditions included atmospheric zone, spray zone, and tidal zone
has significant effects on the service life of reinforced concrete structures. Indeed, the calculated
results show that the latter is lowest at the tidal zone, and highest at the atmospheric zone. These
results fully matche the mesured results of reinforced concrete structure durability in marine
conditions.
The damage state of concrete cover strongly affects on the service life. Once, the damage
level increases, the service life obviously decreases. Corresponding to the maximal value of
diffused damage, the service life is no longer significant. The damage time has also strong effect
on the service life. If the concrete cover is damaged during construction phase, the service life is
very low, even no longer significant.
References
[1]. Angst.U & al, COIN Project Report, Critical Chloride content in reinforced concrete, Coin workshop,
Trondheim, Norway, June 2008.
[2]. Choinska.M, Bonnet.S, Khelidj.A, Couplages endommagement – perméabilité – transferts d’ions
chlorures, GeM, Université de Nantes, 2008.
[3]. COSTA.A, APPLETON. J: Chloride penetration into concrete in marine environment - Part II:
Prediction of long term chloride penetration, Materials and structures, 1998.
[4]. Khatri.R.P, Sirivivatnanon.V, Characteristic service life for concrete exposed to marine
environments, Cement and Concrete Research, 34, p745-752, 2004.
[5]. Makeset.G, Vennesland. O, “Critical chloride content in reinforced concrete”, COIN Workshop, June
2008, Trondheim, Norway.
[6]. Mangat, P.S. & Molloy, B.T. 1994, ‘Influence of PFA, slag and micro-silica on chloride induced
corrosion of reinforcement in concrete’, Cement and Concrete Research, 21, 819–834
[7]. SAJIA.A, VESIKARI.E, “Durability design of concrete structures”, RELEM Report 14, 1996.
[8]. SCHIESSL.P, Corrosion of Steel in concrete, RELEM-Report of the Technical Committee 60-CSC,
1998.
[9]. Stanish. K.D., Hooton .R.D, Thomas.M.D.A, Testing the Chloride Penetration Resistance of Concrete:
A Literature Review, Department of Civil Engineering, University of Toronto, FHWA Contract DTFH6197-R-00022.
[10]. TRAN T .TRUYEN, CHARLIER.R, “A method for evaluating the gas permeability of damaged
concrete”, Science Journal of Transportation, No.03, Moscow-Chengdu-Hanoi, 11/2011.
[11]. TRAN T. TRUYEN, TRAN V. TIEN, “Influence of marine environment conditions on the service life
of reinforced concrete structures”, (In Vietnamese) Journal of Transport and Communications Research,
No 32, Page 78 – 84, 6/2011.
[12]. Zheng.j.j, Zhou. x.z, “Prediction of the chloride diffusion coefficient of concrete, Materials and
structures, 2007♦
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
141
THE BLOCK OF MODELING AND FORECASTING
OF GAS NETWORKS SYSTEMS OPERATION MODES
FOR DISPATCHER DECISION SUPPORT SYSTEM
BERNER, L.I.
KOVALEV A.A.
NIKOLAEV A.B.
State Technical University – MADI,
Moscow, Russia
Summary: The paper describes new methods of supervisory control automation aimed to
help managers to solve the problem of maintaining a balanced mode transmission system.
Forecasting the demand for gas and simulation modes of transmission system is
complemented by a system of automated support for decision making.
This work was supported by the Government of the Russian Federation (Russian Ministry
of Education) as part of the project under the Contract № 13.G25.31.0064 on October 22,
2010.
Keywords: A system of automated support for dispatcher decision making, the task of
balancing gas networks.
Maintaining a balanced mode of operation of gas transmission networks (GTS) is one of
the most important tasks of supervisory control. The gas transportation system must have
sufficient capacity to meet the demand of consumers for natural gas, even in cases of sudden
fluctuations in demand caused by calendar, weather, economic or other reasons. The traditional
solution of this problem is the usage of underground gas storage (UGS) and the gas stock in the
tube as regulators. Gas suppliers can also introduce some regulators, but very often their ability
to vary quickly the production volumes is limited by technology.
The task of balancing the severity of the regime is extremely important in cases of
operation of long gas pipeline with a limited supply of gas in the tube. So far such pipelines
were considered to be typical for foreign gas networks. However, in recent years in connection
with the development of gasification of the Russian Federation and the construction of new
modern pipeline (North - European, Sakhalin - Khabarovsk - Vladivostok pipelines, etc.), the
problem is becoming more and more important for Russian gas industry. The solution of the
problem is becoming even more urgent due to the development of market mechanisms of supply
of gas, gas purchase on the "spot" markets, as well as due to climate variations, etc.
The paper describes some new automated dispatch control methods to help managers to
solve the increasingly complex problem of balancing the GTS.
142
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
These methods have been worked out jointly by LLC "VIP" (the Russian subsidiary of the
Group PSI AG, Germany) and JSC "AtlanticTransgasSystem" for their usage at the facilities of
"Gazprom".
THE TASK OF BALANCING THE CTA AND DECISION-MAKING
Fig 1 shows the pipeline, which is the subject of balancing.
Fig 1. The structure of the system balancing
The gas enters the system from vendors (fisheries) and other companies to transport gas,
and can also be selected from the storage facilities where available. The gas is supplied from the
system to consumers, other companies (gas transit), is pumped into underground storage
facilities, as well as is consumed for their own needs (the work of gas pumping units and other
technological problems). The pipeline itself functions also as storage of gas. As a result, the
supply of gas in the pipeline may both increase and decrease. The task of balancing the
presented system in principle is obvious - the gas must flow into the system in an amount
sufficient to meet consumers' needs, personal needs, as well as for further transit. For various
reasons, the demand for natural gas may fluctuate. First of all, the demand depends on weather
conditions and, often, on the calendar date - the day of the week, holidays/workdays, etc.. A
proportion of demand is associated with spontaneous or predictable seasonal fluctuations in
economic activity. For example, in agricultural regions the plants consume gas is only a few
months of the year for the processing of the crop. Unfortunately, most part of the demand
fluctuation is much less predictable. It should be noted that taking into consideration the length
of the pipelines, the pipeline suppliesnatural gas to consumers in areas with different weather
conditions, which complicates the forecast of demand fluctuations.
Assessing the situation and determining either the excess or deficiency of gas in the
system, the dispatcher must decide on balancing "supply and demand". Possible actions may
include:
• Selection or gas injection from / to storage facilities (if any);
• Increase or decrease the supply of gas in the tube (ie, injection of gas into the pipe or the
selection of the gas pipe - in fact, an analogue of UGS smaller scale);
• Request for additional volumes of gas from suppliers or the opposite - a request for
reduction of gas supplies;
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
143
• In an extreme case - failure / restriction of additional or "spot" applications for the supply
of gas and very emergency - the restriction of some consumers.
Thus, in order to balance the system it is necessary to settle two main questions: (1) what is
the current situation and trends of its changes - is there enough gas in the near future, (2) what
measures from those listed above to take in order to align the imbalance.
Taking into consideration the inertia of the gas transmission system and of storage
facilities, a prerequisite is to predict imbalances for the period of time, which at least allows the
dispatcher to take effective measures to eliminate potential problems.
DISPATCHING ORGANIZATION ALTERNATIVE
The following solution is proposed for organizing and controlling the gas transportation
system to balance the supply of gas, as well as for solving problems of safe operation of the
pipeline and the optimization of the equipment. In addition to the basic system of supervisory
control (SCADA PSIControl), "on-line" model of the pipeline (PSIGanesi), «off-line" model, a
system of forecasting gas consumption, and the newly developed PSIPrognosis software system
analysis and decision support (SPPDR) are realised.
Fig 2. The structure of the system (supply)
A generalized algorithm SPPDR shown in fig 3.
Modeling of the "future", based on the targets is a tool to identify imbalances in the system.
Analysis of the results of modeling is performed by the dispatcher "manually", as well as with
the help of automated mechanisms for expert systems - SPPDR module. Heuristic rules SPPDR
allow the dispatcher to make recommendations - to change the mode of the compressor station
(CS), request a change in the regime of fishing, to change the supply of gas in the tube, away
from the gas storage facilities, restrict consumers. Recommendations SPPDR can be considered
to be preliminary. Optimal solutions for the pipeline mode of operation should be obtained
interactively using the gas-hydrodynamic model of "Astra." Model "Astra" calculates various
options for MG modes, which on the advice of an expert system SPPDR or proceeding from its
own considerations, the manager offers. Option is selected, or the best of the fuel and energy
144
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
costs, or preferred by some subjective reasons.
Fig 3. The control algorithm (proposal)
The final choice of options for changing the operating mode of the COP and the MG as a
whole takes the dispatcher. The selected option mode takes the form of supervisory tasks and
settings for transfer to lower-level automation system.
The proposed approach implements not only remote but also the intelligent management of
the pipeline mode of operation and allows quicker and with less expenditure of energy to
balance the important pipeline system. Domestic consumers and, in the future,
foreignconsumers are provided with a guaranteed supply of gas.
In addition to the problem of balancing work of MG and ensuring guaranteed delivery of
normal modes, the system can be used in emergency situations, helping to identify alternative
ways of organizing supply and install a real need of limiting gas consumption.
CONCLUSIONS
This paper proposes a modern and perspective, according to the authors, method of
organization of supervisory control and solving an important task of balancing modes of gas
transmission system. The authors hope to put into practice the developed (at the level of a
model) idea and apply it to provide efficient and reliable operation of gas transmission
networks.
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
145
INDUSTRIAL INTEGRATION AND ECONOMICAL CRISIS
GALINA BUBNOVA
Moscow State University of Railway Transport
Russia
Summary: Taking into consideration the dominance of the feed component in the
tradable mass produced by the majority of Russian enterprises and destabilizing tendencies on
the world’s markets, the task of organization of tolling business with the participation of major
shipping companies becomes actual. In the given context the differences between available
technology of domestic and foreign companies, their productive capacity, volume and quality
of reprocessing of the primary product, necessary volumes of resources of the primary
products to ensure the steady production in Russia are taken into consideration.
The Business participation of the shipping holding company PLC ‘Russian Railway’ in
organization and support of the tolling business does not only broaden the bounds of its
business, but it is also a long-range tendency of decreasing the painful consequences of the
international economic crisis.
The analysis of the Russian industrial enterprises with tolling schemes allows us to reveal
problems that limit the effectiveness of its application – these are the legitimacy of application
of such operations, cohesion and uncertainty of bounds of tolling and customer-owned schemes.
The absence of the definition of tolling on the legislative level leads the experts in this field
to multiple-valued conclusions regarding its application and the development of these
operations.
According to the definition of the International Trade Centre (UNCTAD/WTO) a tolling
contract is a variant of the take-or-pay contract, which means that the condition of contractual
obligations is unconditional purchase of a product or a service offered to the purchaser with the
condition of buying outright. In the case of refusal of the purchase the buyer must pay the seller
an appointed sum of cash means. In the case of the tolling contract it is the payment for the
usage of the infrastructure objects, and in the case of the take-or-pay contract it is also for the
non-fulfillment of the contractual obligations imposed on one of the participants.
The Russian equivalent of the mechanism of the tolling truck represents an operation of
reprocessing of the customer-owned primary product and materials. Mutual relations, based on
such schemes, correspond to the relations of rendering of services provided by specialized
enterprises such as fulfillment of functions of maintenance supply of one enterprise and
marketing in the bounds of the existing production process. The primary product and
components act as customer-owned primary product, they are passed from the owner of
production to the manufacturer for the purpose of their completion or reprocessing into a
146
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
finished product and, by that, they represent the basis of the tolling truck.
Why such relations are substituted for simple operations of customer-owned primary
product industrial enterprises in a number of cases? To a great degree it depends on the practice
of using the conception of the reprocessing of the customer-owned primary product in tolling
schemes during the times of formation of aluminium production. There is still no definition of
the concept of tolling in the Russian law, but there are characters based on which the operations
of industrial enterprises, that work on customer-owned primary product, are labeled as such.
In general, the roles of the participants of the tolling truck are acted by such subjects as: the
owner of the primary product, the reprocessor of the primary product, the owner of the finished
product (the result of the reprocessing), the consumer and the middlemen.
The indication for labeling the requestor of reprocessing as a middleman during the
realization of the tolling schemes coincides with the indication of transition of the proprietary
right for the results of the reprocessing of the primary product and materials to the client. In this
context it means that the corresponding proprietary right has been passed to him/her according
to the sale agreement of primary products and materials. This indicated that the tolling
middleman acts as a specialized company that combines the functions of maintenance supply
for the reprocessor of the primary product, on the one hand, and, on the other hand, the
marketing function for the primary owner of the primary product and materials for the further
reprocessing in the higher redistribution of the products (fig 1).
3
The supplier
of primary
product
The
reprocessor of
the primary
product
2
The tolling
middleman
4
The
consumer
1
Fig 1. The diagram representing production on the customer-owned material
with participation of the tolling middleman
During the realization of this type of the tolling scheme the primary owner of the primary
product becomes the supplier of the primary product for the tolling middleman, the proprietary
right for the result of the reprocessing belongs to the middleman until the primary product is
given to the reprocessor. In the end the finished product is distributed to the consumers by
means of the tolling middleman. Organizationally, the participants of this scheme are not
connected, and in some cases the tollinger acts as a financial middleman.
In general the scheme consists of the following steps:
1. The supplier passes the proprietary rights (sells) to the primary products or materials that
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
147
are at his/her disposal, because he/she does not possess the necessary and/or free facilities for its
reprocessing, or the passing product is the final step in his/her production process;
2. The tolling middleman, acting as a specialized company, that does not possess any specific
equipment for the reprocessing of the acquired primary product, sends it for reprocessing to an
enterprise that possesses the right technological equipment or available equipment for this purpose;
3. In accordance with the results of the reprocessing of the primary products and materials
the ready product of higher redistribution is passed to the tolling middleman;
4. The tolling middleman sells the finished product to the customers.
It should be noted that in some cases tolling is viewed as a variant of counter trade that can
be placed in-between barter operations, trade compensation and operations of industrial
compensation. In this case the scheme is completely different, the role of the tolling middleman
is passed to the owner of the primary product and materials, and the owner passes them to the
reprocessor of the primary products (1) for the purpose of acquisition of products of higher
redistribution (2) and, later, the owner sells it to the customers (3) (fig 2).
2
The reprocessor of the
primary product
1
Primary product
owner
3
The consumer
Δ1 or Δ2
Fig 2. Diagram representing production on the customer-owned material
The fundamental difference of the tolling scheme from such forms of organization of the
distributed production consists in the fact that tolling can be characterized by saving the
proprietary right for the products that are received in 100% volume from the received primary
products or materials and it does not make a provision for non-monetary compensation of the
reprocessing services (Δ1 or Δ2). However it is stipulated that a part of the primary product or
material or a part of the finished product is given to the reprocessing enterprise as a payment for
the reprocessing services.
We find the explanation of the specificity of the mutual settlements in Russia during the
period of cash shortages characterizing the economy in the beginning of the 1990s as an
argument for calling tolling a barter type operation. Classification of these types of schemes as
an operation with the customer-owned material left traces and has common features with the
tolling schemes in question.
Tolling schemes (fig 1) started their development by putting the requestor of the
reprocessing in the middleman category in realization of such schemes, acting as a collaborating
link in the organization of divided production and was also the owner of the primary product
(materials). Being such a ‘middleman’ the reprocessing client performed in this scheme such
functions that the primary owner of the primary production did not have. Such functions were:
148
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
1. Maintenance supply for the reprocessor of the primary product;
2. Marketing functions for the primary owner of the primary product.
If we proceed from the practice that existed when the owner of the primary product was
also the requestor of the reprocessing (both with monetary and barter (non-monetary) types of
settlement), then this variant is not being used nowadays, but it cannot be excluded during the
times of the economic crisis.
The maintenance supply problem is resolved only in part also because the owner of the
primary product cannot solve it on his own, having a limited quantity of resources for the
reprocessing which cannot provide full utilization of the available productive capacity of the
reprocessing enterprise. Besides, the quality of the supplied primary products and materials
must satisfy the technical requirements of the processing enterprise.
The insufficiency of such scheme reveals itself in the fact that the client solves his/her own
problems, particularly the inability to reprocess given primary products and/or materials. The
economical point of this scheme is justified - the finished higher redistribution product made of
one’s own primary product has much higher value which has a good effect on the financial
result of both the processing enterprise and the requestor of such reprocessing.
The limits of the usage of the scheme are obvious, it solves problems of only one
participant in an industrial relation and does not have the same selection of functions that would
allow, as a result of such cooperation, to create a larger surplus value which permits to solve the
tolling schemes.
Therefore, unlike all the other forms of lean production (including production as a part of
agreements concerning division of production and unreliable transactions) tolling does not
substitute for barter schemes in the settlement between the parties. Also tolling does not
substitute for partial settlement with the finished product for the services of reprocessing of the
customer-owned materials (primary products) to the reprocessor of the primary products, which
is reflected in the current Russian practice.
Since 2006 questions of the tolling production were included in the list of pressing
problems by a group of experts in national income accounting under the UN Economic and
Social Council (ECOSOC). Following the results of the first meeting in April of 2008, which
was held with the assistance of the Economic Commission for Europe, Statistics Unit of the EU,
Organization for Economic Cooperation and Development, Working Group on the Impact of
Globalization on National Accounts (WGGNA) special attention was paid to the tolling
production and reprocessing of goods.
The results of the influence of tolling schemes, that are part of a special-purpose chain of
the delivery of goods, will be reflected in June 2010 during the conference of European
statisticians in the outcome document of the WGGNA.
According to the WGGNA the subject matter of tolling is as follows:
- The proprietary rights on goods are kept by the supplier and are also extended to the
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
149
products of reprocessing;
- The reprocessor is paid for the work;
- The enterprise sends the goods to the other country with the purpose of reprocessing
keeping the proprietary rights on these goods which will later extend to the products of
reprocessing that will later be sold in third countries.
This definition corresponds to the international chain of product deliveries which also
specializes on the tolling production. Consumers buy the proprietary right from the enterprise
which sent the goods to reprocessing. This procedure is characterized by the difference in the
quantity of the surplus value between the reprocessed goods and the goods accepted for
reprocessing and which cannot be determined in advance by the supplier. The beneficial
economic effect attracted the interest of different participants of foreign economic activity,
which was the reason for the European Economic Commission’s attention to this process. The
ultimate aim of organizing tolling production for international chains of goods delivery will be
to supply the entire chain with steady production processes and at the same time observe the
necessary balance of interests among all the international specialization participants and to
promote cooperation during international trade.
Taking into consideration the UN definition of reprocessing of goods and tolling
production, the necessary condition for tolling is direct participation of parties of such an
agreement in the foreign commerce turnover during trade through the customs border of
countries. This calls for observance of routine of the movement of goods through the customs
borders of countries.
The existing legislation of Russia contains the rules according to which manufacturing or
other activities of the subjects of foreign-economic activity (international trade) is classified by
the experts as tolling. In general tolling relations are regulated at the levels of existing norms of
Russian and international law, even though these rules and norms are not directly described as
tolling relations.
The customs code makes provisions for the following customs regime which are applied to
the goods for which reprocessing is provided:
- Reprocessing in the customs territory;
- Reprocessing for domestic consumption;
- Reprocessing outside of the customs territory.
Customs regimes that are described in the latest editions of the Customs Code are in
content similar to procedures applied in accordance with the International Convention for
Simplification and Harmonization of Customs Procedures. Since this international document
was not ratified we will make an analysis of procedures of customs in reference to the criterion
for labeling foreign trade dealing as these regimes with the purpose of specification of tolling
schemes, the results of which are shown in the tab 1.
150
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
Tab 1
Criterion
Reprocessing in the
customs territory
Reprocessing for the domestic
consumption
Reprocessing
outside of the
customs territory
Customs
territory
Russian Federation
Russian Federation
Other countries
The requirements
on the term of
reprocessing
Not more than 2 years
Not more than 1 year
Not more than 2
years
The conditions
of
customs
regime
Reprocessing of goods
with further export of
products of reprocessing
from
the
customs
territory
Reprocessing with further release of
reprocessing products for the free
circulation
Reprocessing
of
goods with further
import
of
reprocessing
products
Economical
cosequences
Full
conditional
exemption from the
payment of customs
duties and taxes
Full conditional exemption from the
payment of customs duties till the release
of the products of reprocessing for the free
circulation, the payment of customs duties
for the reprocessing products afterwards.
Full or fractional
conditional
exemption from the
payment of customs
duties and taxes
The customs regime of reprocessing for domestic consumption is not being applied at
present, because permissions for reprocessing are not issued until the Government of the
Russian Federation (according to p.2 section 188 of the Customs Code of the RF) has ratified
the list of goods as regards to which the reprocessing for domestic consumption is allowed.
Analyzing reprocessing regimes in the customs territory and reprocessing regimes outside
the customs territory we have discovered that the main difference is in the customs territory, in
the country where the reprocessing of goods occurs. Accordingly, for the first regime such
country is the Russian Federation and for the second regime it is another country. In 2008 the
ratio of regimes was 1:13, fig 3.
7%
Reprocessing in the customs territory
Reprocessing outside of the customs territory
93%
Fig 3. The share in the cost estimation of goods of customs regimes of reprocessing
Reprocessing on the customs territory is a regime under which the imported goods are used
in the customs territory of the Russian Federation during the time-frame (the time of the
reprocessing) for the purpose of executing operations of reprocessing goods with full
conditional exemption from the payment of customs duties and taxes with the condition of
taking out the results of reprocessing from the customs territory of the Russian Federation
during the given period (fig 4).
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
151
Russian Federation
1
The
consumer
3
The reprocessor of
the primary product
Tolling
enterprise
2
Fig 4. Diagram representing the reprocessing regime in the customs territory
According to the content of the reprocessing regime in the customs territory, the goods that
were imported for the purpose of reprocessing (1) are liable to obligatory export from the
customs territory of the Russian Federation as reprocessing results (2) and in the end the can be
sold to consumers (3). This regime will be reclassified as a regime of reprocessing for the
domestic consumption with all the economic consequences, if the conditions of placing the
goods under this customs regime (such as identification of exported goods as products of
reprocessing) are violated.
In the category of goods can be placed any moveable property that is moved across the border,
and for the purpose of reprocessing primary products and materials can be classified as such goods.
In accordance with the customs regimes during the organization of tolling schemes the
following operations of reprocessing goods are possible:
1. Proper reprocessing or processing of products;
2. Manufacturing of new products, including assembling, installation and disassembling of
products;
3. Epairs of goods, including their restoration, changing of their components, restoration of
their essential properties.
In 2008 the largest part from the goods placed under the reprocessing customs regimes
were received by equipment and mechanisms in group of reprocessing in the customs territory
regime and transport in group of reprocessing outside of the customs territory (tab 2).
Tab 2
The name of the heading
(the group of goods)
1. Equipment, mechanisms
2. Plastic products, rubber
3. Base metals and products
4. Transport
5. Textile
6. Chemical products
7. Optical devices and instruments
8. Mineral products
9. Wooden, paper and cardboard mass
10. Different industrial goods
Total
152
Reprocessing in the
customs territory
40,68%
16,54%
13,26%
4,59%
7,90%
5,75%
4,95%
2,93%
1,91%
1,48%
100,00%
Reprocessing outside of
the customs territory
10,23%
1,14%
6,20%
75,57%
0,18%
0,37%
2,01%
4,05%
0,01%
0,24%
100,00%
Total
38,55%
15,47%
12,77%
9,54%
7,37%
5,38%
4,75%
3,01%
1,78%
1,39%
100,00%
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
In the joint group of goods without division into customs regimes the following are
prevalent: equipment and mechanisms, plastic and rubber products and, of course, base metals
and products made from them. The placed orders in these groups prove that the reprocessing
enterprises are well equipped with productive assets and a relatively inexpensive labor force.
The materials, primary products and goods received for the reprocessing are characterized by
a moderate level of redistribution. The resulting products of reprocessing will be components or
semi-finished product for another enterprise. And the high level of ‘foreign’ good that are
reprocessed speaks about the inapplicability of its usage in domestic production, and in case of the
impossibility of completing reprocessing independently these products should be ordered in other
countries where they are applicable, as in case of goods of the transport category (75,57%).
The content of this customs regime reflects the essence of the tolling scheme and its
difference from other forms of lean production, which assumes non-monetary compensation of
reprocessing services, and the differences are revealed in the following:
- The conditions of exporting products of reprocessing;
- The relations of leftover goods that were placed under the ‘in the customs territory’
customs regime.
The conditions of exporting products of reprocessing (in particular those that are contained
in p.2 section 178 of Customs Code of the RF) state that the description, quality and quantity of
reprocessed products are finally defined after the reconcilement of the reprocessed products
yield norm, which, in their turn, are reconciled with customs during the expert and/or laboratory
appraisal based on the specific technological process of reprocessing.
In respect to the leftover remains of the goods, that can be used as a payment means for the
services of reprocessing in barter type deals, these goods are characterized by the customs
clearance charge and are not subject to declaration.
The remains of goods placed under the customs regime of reprocessing in the customs
territory can be exported from the customs area of the RF without customs clearance charge or
can be placed under the customs regime of reprocessing in the customs territory.
The dispositions of these norms of the law exclude every type of compensation deals or
barter deals the basis for which can be the reprocessing of the customer-owned primary product
(material), but that do not fit in the definition of the tolling scheme according to the UN version,
and in particular remission of customs duty. This fact makes participation in tolling schemes one
of the arguments for the attractiveness of such a form of business partnership.
Taking into consideration the legislation in force, tolling is excluded from the list of barter
business, therefore tolling enterprise cannot be viewed as a commercial middleman in unreliable
barter transactions.
Total exemption of customs clearance charges will be executed under the condition of
exporting the products of reprocessing from the customs territory or their importing in
accordance with the conditions of customs regimes.
Keeping in mind the results of the analysis of principles of tolling organizing tolling
schemes and the contract relations connected to them, we have a fundamentally new scheme of
cooperation of economic entities, in conformity with the recent requirements of the existing
legislation of the RF in the field of international trade (fig 5).
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
153
1.2
1.1
Primary product owner
The reprocessor of the
primary product
Tolling enterprise
The primary supplier
The owner of the
finished manufactures
The consumer
Fig 5. The diagram of production on the customer-owned primary product (tolling scheme)
On the diagram the home manufacturing is centered in the hands of the reprocessor of the
raw material when supply and sales functions are distributed among other subjects of the tolling
scheme. This tolling scheme unites the following participants:
1. Tolling enterprise:
1.1. The owner of the primary product;
1.2. The owner of the finished product;
2. The reprocessor of the primary product;
3. The original supplier;
4. The consumer.
Acting as a requestor of reprocessing tolling enterprise acts as the owner in respect to the
supplied primary products and materials. During the reprocessing under conditions like tolling,
the proprietary right on the products of reprocessing is kept by the owner of the primary product
and materials.
On this diagram the division of the tolling enterprise is done under the condition that the
customs regime of reprocessing for domestic consumption (the regime is allowed, but does not
yet exist) may be joined into force, because at present there is no direct law that prohibits that.
In this case the products of reprocessing are exempt from exportation, and the subsidiary
production unit (that exists as a part of cooperating tolling enterprise) should act as the owner of
the finished products. In the legal context it means that being a foreign company, that supplies
the primary products and materials for the purposes of reprocessing, this company (under the
regime of reprocessing for domestic consumption) should also act as the owner of the finished
manufactures in the Russian Federation and should have all the taxpayer’s right and
responsibilities connected to the transition of goods through the customs borders of countries.
In this case the subsidiary production unit of the tolling company together with the
registered outside capital represents its interest on the territory of the Russian Federation.
Therefore, the functions of the tolling enterprise are divided according to the territorial mark
connected with the allowed reprocessing regimes with the condition of keeping proprietary
154
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
rights on the conversion product.
The functions of the tolling enterprise are not reduced to the function of a middleman in the
tolling schemes. Middlemen are singled out in an independent category, and the transport
enterprises act in this role. It is defined according to the following:
1. The realization of logistical aims:
- Marketing function for the primary owner of raw products and materials;
- Maintenance supply for the reprocessor of the primary product.
2. The aims of managing ‘just in time’ commodity flow (assuming the fixed term of
reprocessing goods) may be accomplished only by a specialized company, the function of which
is given to a tolling enterprise for outsourcing. Under the conditions of cooperating for tolling
production, the transfer to outsourcing is an economically justified decision within the limits of
production processes of distributed production with the reconcilement of goods supply chains.
The tolling enterprise, accomplishing the aims of logistics on its own without the
involvement of transport enterprises, is not capable of accomplishing them measurably, and in
the opposite case the independent accomplishing of these aims and functions is more
extravagant.
Summarizing the results of the conducted analysis, let’s point out the most important, to
our mind, advantages for the participants of the tolling schemes.
For the suppliers of primary products and materials: tolling relations will allow them to
broaden the range of launched products as a result of a magnified number of reprocessing
enterprises capable of increasing not only the quantity of launches, but are also capable of
improving the quality of the mass of goods.
The basis of tolling enterprises is balancing the interests of participants of such schemes
and relations under the conditions of managing the flow of primary products, materials and the
reprocessing products, so the tolling enterprises will be ensured of the distribution of such
technologies among the regions and also penetrating new markets. It is obvious that the
advantages and profits of such integration will be applied to all the participants of tolling
relations.
To the reprocessor of the primary product tolling schemes will allow ensuring the full load
of its productive capacities without the involvement of additional sources of financing for its
own productive program and without a substantial loss of economical profits in the future. At
the same time the enterprise that is provided with the utilization ratio indirectly solves the social
problem in a town where this enterprise is strategic.
For the transport company (PLC ‘Russian Railway’) shared participation in the tolling
enterprise business is one of the most promising areas of focus of the business development of
PLC ‘Russian railway’. The tolling enterprises in question are those that are profiled for the
nomenclature of primary products and materials that are used in transport production or that are
in demand for production of other goods that are in their turn necessary to ensure the transport
production.
The tolling enterprises as international integrators can serve as a peculiar kind of ‘a
pillow’, the amortisseur of painful strokes of the economic crisis gathering pace in the different
countries of the world.
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
155
INITIAL RESULT OF USING CINDER PARTICLES AT SOME STEEL
FACTORIES IN BA RIA - VUNG TAU PROVINCE
AS A MINERAL ADDITIVE FOR CEMENT CONCRETE
IN BUILDING MOTORAY SUFRACE
DR. LE VAN BACH
University of Transpor
and Communications
ENG. TRAN HUU BANG
Saigon Construction Quality Control
Jont Stock Company
Summary: The report presents initial result of studying characteristics of cement
concrete using slag as a mineral additive at some factories in Ba Ria - Vung Tau province to
manufacture cement concrete for construction of motorway. This application is, in fact, piloted
in some road sections.
I. EVALUATION OF SLAG PARTICLES AT SOME STEEL FACTORIES IN BA RIA VUNG TAU PROVINCE
Steel slag is waste in metallurgy, which is used as a waste product in metal production
process from iron ore or unpure metal refining process. Iron ore normally contains clay and
sand, so people usually add appropriate content of limestone to the furnace during production.
During burning, iron ore and limestone react with each other to form silicate calcium, alumina
silicate and aluminate silicate and magnesia calcium. Steel slag is melted at 1400 - 1600oC. At
this temperature, compounds are completely melted. The specific weight of melted compounds
is smaller than that of cast iron, so it floats on top. People remove and call this product slag.
Steel slag has various forms depending on burning process and cooling mode after melting.
Quick cooling is accompanied with limited quantity of water, then the steam is withdrawn,
leaving pore, honeycombed in slag structure similar to holystone. Light foamed slag is then
crushed and used a mineral additive for concrete.
Ba Ria - Vung Tau where there is concentration of many steel factories is considered as
metallurgy center in the South with output of 3.75 million bloom yard ton/year, forecasted
weight of 412,000 - 562,000 steel slag ton/year. Steel slag arising from steelmaking which isn’t
recycled generates dust, overflown rainwater through slag storage. This will be a burden on
environmental and economic issues for steelmaking plants. At the same time, it will lose land
and waste resources if being dumped (Source - Green Resources Joint Stock Company,
24/06/2011).
156
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
Figure 1. Steel slag storage of steel factory
in the South, Ba Ria - Vung Tau
Figure 2. Steel slag storage at Toc Tien commune,
Tan Thanh district, Ba Ria – Vung Tau province
II. SOME TESTING RESULTS IN CRITERIA OF CEMENT CONCRETE USING
WASTE STEEL SLAG AS A MINERAL ADDITIVE
2.1. Preparation of materials for testing
- Cement: Holcim PCB40 cement.
- Sand: Dong Nai river sand (type of sand used to produce cement concrete).
- Rock: exploited and manufactured in Hoa An stone quarry, Dong Nai province.
- Water: water for conventional concrete.
- Mineral additive: Steel slag particles have form of black grey clots, spongy volume
weight of 900kg/m3, specific weight of 2.5 g/cm3.
Figure 3. Steel slag sample before and after being crushed to produce a mineral additive
- Chemical components of steel slag: SiO2: 28 - 38 %; Al2O3: 8 - 24 %; Fe2O3: 0,64%;
CaO: 30 - 50 %; MgO: 5 %; S: 1 - 2.5%.
- Mineral components of steel slag:
Ghilenit (2CaO. Al2O3.SiO2, CaO.SiO2, 2CaO.SiO2) mineral component. In addition, there
are Monticenit (CaO.MgO.SiO2), Akemanit (2CaO.MgO.2SiO2), Merwinit (3CaO.MgO.2SiO2),
Anorthit (CaO. Al2O3.2SiO2), Spinel (MgO.Al2O3), Fortenit (2MgO.SiO2) và các Aluminate
canxi (CaO.Al2O3, 12CaO.7Al2O3).
Design of conventional cement concrete aggregate components and cement concrete uses
steel slag mineral additive (table 1).
Tested cement concrete grades: 25MPa; 30MPa and 35MPa.
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
157
Table 1. Calculated results for conventional cement concrete grade components
Concrete grade component
Design humidity W=0%. 25Mpa grade
Adjustment of humidity of Wđ = 0.29 % and Wc =
7.35%. 25Mpa grade
Designed humidity W=0%. 30Mpa grade
Adjustment of humidity of Wđ = 0.29 % and Wc =
7.35%. 30Mpa grade
Design humidity W=0%. 35Mpa grade
Adjustment of humidity of Wđ = 0.29 % and Wc =
7.35%. 35Mpa grade
Material component for concrete (1m3)
X (kg)
C (kg)
Đ (kg)
N (kg)
353
754.9
1098
205
353
810
1101
147
404
722
1085
205
404
775
1088
149
468
669
1065
211
468
718
1068
160
Note: X - cement; C - sand; Đ - rock; N - water.
Design of cement concrete uses steel slag mineral particles: based on design steps of
conventional concrete. But replacement of cement content is in proportion to percentage: 10%,
12% and 15% slag in mixed components of the aggregate. Results are shown in table 2 after
adjusting humidity of Wđ = 0.29 % and Wc = 7.35%.
+ Age of tested cement concrete sample:
- Sample for testing compressive resistance Rn applicable to 5 periods of days: 7, 14, 28, 60
and 90;
- Sample for testing tensile resistance Rku applicable to 2 periods of days: 28 and 60.
+ Sample dimensions:
- Compressed sample: 15x15x15 cm;
- Tensile sample: 15x15x60 cm.
Table 2. Design of concrete grade components using steel slag as a mineral additive
Materials components for 1m3 cement concrete
with mineral additive
Steel slag Steel slag Steel slag
X (kg)
C (kg) Đ (kg) N (kg)
10% (kg) 12% (kg) 15% (kg)
317.50
810
1101
147
35.50
25Mpa
310.64
810
1101
147
42.36
300.05
810
1101
147
52.95
363.60
775
1088
149
40.40
30Mpa
355.52
775
1088
149
48.48
343.40
775
1088
149
60.60
421.20
718
1068
160
46.80
35Mpa
411.84
718
1068
160
56.16
397.80
718
1068
160
70.20
+ Number of samples to be manufactured:
- Types of sample for testing compression 15x15x15cm:
Number of samples using steel slag: 3 slag grades x 5 types of age x 3 grades x 3
samples/group = 135 compression samples; Number of check samples (without steel slag): 5
types of age x 3 grades x 3 samples/group = 45 compression samples.
- Types of sample for testing tensile resistance 15x15x60 cm:
Components of
cement
concrete
aggregate
158
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
Number of samples using steel slag: 3 slag grades x 2 types of age x 3 grades x
3samples/group = 54 samples; Number of check samples (without steel slag): 2 types of age x 3
grades x 3 samples/group = 18 samples.
Figure 4. Cubic compression samples 15x15x15cm and tensile samples 15x15x60cm
Figure 5. Testing for compression and tensile testing
for cement concrete containing 15% mineral additive
Some remarks:
- Cement concrete strength containing steel slag mineral additive develops more slowly
than conventional cement concrete.
- Next study results after 28 days show that in age of long-time when minerals in Portland
cement have nearly liquefided, components in steel slag continue to react with Ca(OH)2 to form
C-S-H products, to increase condensation and strength for cement concrete containing steel slag
mineral additive to approximate strength of concentional cement concrete. (For example, after
60 days, compression strength of conventional cement concrete compared to concrete
containing 10%, 12% and 15% mineral additive in term of 30MPa sample is, in turn, 35.9, 35.0,
34.2 and 32.7MPa - table 3.
Table 3. Compression strength of cement concrete with and without steel slag
Sign of sample
Conventional 25MPa
25Mpa; 10% steel slag
12% steel slag
15% steel slag
30MPa, conventional
30MPa; 10% steel slag
12% steel slag
15% steel slag
35MPa, conventional
35 MPa; 10% steel slag
12% steel slag
15% steel slag
Compression strength in different ages of days (MPa)
R7
R14
R28
R60
R90
19.6
23.3
28.1
30.1
30.8
18.9
22.6
27.5
29.0
30.5
17.8
21.7
26.5
28.5
29.5
15.9
19.8
25.2
27.0
27.2
23.8
27.4
33.5
35.9
36.1
22.5
26.5
32.5
35.0
35.8
21.4
25.9
31.4
34.2
34.4
20.2
24.4
30.5
32.7
32.7
26.5
31.7
38.5
40.6
40.7
25.6
30.5
37.2
39.4
40.4
24.8
29.5
36.7
38.9
38.9
23.6
28.3
35.5
36.7
36.7
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
159
Table 4. Tensile strength of cement concrete with and without steel slag when bending
Sign of sample
25MPa, conventional
25 MPa, 10% steel slag
12% steel slag
15% steel slag
30MPa, conventional
30MPa, 10% steel slag
12% steel slag
15% steel slag
35MPa, conventional
35MPa, 10% steel slag
12% steel slag
15% steel slag
Tensile strength (daN/cm2)
R28
R60
33.51
34.31
32.58
33.64
29.38
30.22
28.31
29.02
38.97
39.78
36.28
37.73
34.21
35.07
32.28
33.02
40.60
41.56
38.91
40.40
36.41
37.73
33.56
34.71
Figure 6. Diagrams for comparing tensile strength
of cement concrete f 25, 30 and 35 MPa grades with and without steel slag in different ages of days
III. PILOTING FOR CONSTRUCTION OF A ROAD SECTION BY CEMENT
CONCRETE WITH STEEL SLAG
3.1. Introduction to the project
- Project: Internal road.
- Location: My Xuan commune, Tan Thanh district, Ba Ria - Vung Tau province;
Technical specifications of the road:
160
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
- Length: 430m;
- Road surface width: 9.2m;
- Concrete slab thickness: 15cm;
- Concrete slab width: 4.58 x 3.0m;
- Expansion joint width: 15mm;
- Cement concrete grade: 25MPa.
- Percentage of steel slag mineral additive: 12%.
The employer: My Xuan brick and tile joint stock company;
Concrete supply unit: Development Investment Construction Concrete Joint Stock
Company (DIC);
Table 5. Strength of concrete cement in the laboratory and on the site
Tensile strength after
Compression strength, MPa
Type of concrete
28 days, MPa
7 days
14 days
28 days
Conventional concrete
19.2 (19.6) 22.9 (23.3)
27.7 (28.1)
3.28 (3.35)
cement
Concrete cement with
17.4 (17.8) 21.4 (21.7)
26.1 (26.5)
2.89 (2.94)
12% slag
Note: Result in the parenthesis in on the site.
3.2. Remark
When applying to projects in fact, testing results in the laboratory is applied to reality. For
conventional concrete, compression strength in the laboratory is 1-3% different from in reality,
tensile compression reaches 98% after 28 days when bending. For concrete with 12% slag, the
compression strength in the laboratory is 1-3% different from in reality, tensile compression
reaches 98% after 28 days when bending.
Figure 7. Some pictures of constructing the pilot section
by cement concrete containing mineral additive
3.3. Evaluation of economic efficiency
To compare conventional cement concrete and 12% steel slag cement concrete for the
above works.
When selecting 25Mpa cement concrete, cement content used in 1m3 concrete is compared
as follows:
The price list is taken according to basic building price unit of Finance - Construction
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
161
Departments in October, 2011 of Ba Ria - Vung Tau province (the price doesn’t includes VAT transported to the construction site).
In calculating expenses with depreciation of remaining value of production lines; revenue
for support use of slag from steelmaking plants, the price of producing steel slag as a mineral
additive is defined at approximately: 174,000 VND/ton.
Therefore, 1m3 cement concrete with steel slag is 62,654 VND cheaper than conventional
concrete.
A road section with length of 430m, road surface width of 9.2m, road surface thickness of
15cm will require 593.4 m3 cement concrete. Use of cement with steel slag will be cheaper,
saving 14,861,529 VND (forteen million eight hundred sixty one thousand five hundred twenty
nine Vietnam dong), reducing 11% expense. However, the value doesn’t include benefit due to
consuming large quantity of steel slag from steelmaking plants, harmful to the environment.
IV. CONCLUSION
- When using steel slag as industrial waste as a mineral additive in cement concrete
components. It’s possible to partly replace and to reduce cement content, reduce hydration heat,
to improve aggregate gradation, to enhance properties of cement concrete;
- For cement concrete using steel slag mineral particles, it’s required to replace part of
cement according to percentage of 10%, 12% and 15%. Initial testing results for grades of 25,
30 and 35 MPa show that its mechanical properties meet technical requirements of cement
concrete used for construction of motorway;
- On the other hand, when using steel slag slag particles as a mineral additive, it’s required
to replace part of cement to minimize greenhouse effect and to make use of waste from
steelmaking plants;
The study result shows that steel slag particles have characteristics to produce mineral
additives meeting TCVN 6260-1997 on mineral additives in manufacturing cement and cement
concrete.
In addition, use of steel slag from industrial waste remedies dumping and storage at
concentrated waste treatment zone at Toc Tien commune, Ba Ria - Vung Tau province, which
currently seriously pollutes the environment.
Economic value from cement concrete using steel slag as a mineral additive is also an issue
of concern.
References
[1]. Nguyễn Văn Tránh và Trần Vũ Minh Nhật - Đề tài Nghiên cứu dùng xỉ trong công nghiệp xản xuất xi
măng Portland xỉ (RESEARCHING OF USING LASTFURNATED SLAG INPRODUCING SLAGPORTLAND CEMENT INDUSTRY).
[2]. PGS. TS Phạm Huy Khang - Tro bay và ứng dụng trong xây dựng đường ôtô và sân bay trong điều
kiện Việt Nam.
[3]. Thông tin Khoa học Kỹ thuật xi măng, số 1/2006 - Nghiên cứu hàm lượng xỉ lò cao tới độ bền sun
phát của đá xi măng.
[4]. Quyết định số 798/QĐ-TTg ngày 25-5-2011 của Thủ tướng Chính phủ về phê duyệt chương trình đầu
tư xử lý chất thải rắn giai đoạn 2011-2020♦
162
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
MECHANISMS FOR INTEGRATION FEDERAL
AND REGIONAL STRATEGY
FOR ENSURING ROAD TRAFFIC SAFETY IN RUSSIA
DR. ALEXANDER CHUBUKOV
PROF. VALENTIN SILYANOV
State Technical University - MADI
Summary: In the paper the effect of Federal target program on increment of road traffic
safety in 2006-2012 in Russia is under discussion. The mechanisms for integration federal and
regional strategies for ensuring road traffic safety are recommended.
Keywords: Federal target program; ensuring road traffic safety; road accidents.
Successful course of realization of the Federal target program « Increment of road traffic
safety in 2006-2012», dynamics of change of target indicators of the program, as a whole
positive reaction of a society to changes spent in this sphere, create premature optimistic
estimations of the further development of a situation with road traffic safety in the regions of the
Russian Federation.
Some results of realization of the Program (2004-2009) are the following:
• 8500 human lives are saved;
• Life of 562 children is saved;
• Preservation of indicators of 2004 would lead to death on roads of Russia of 13800
persons;
• The situation on roads of the Russian Federations has essentially improved;
• There is a full confidence of achievement of target indicators of the program.
Considering safety of traffic as one major, but not a unique indicator or the indicator of
efficiency of functioning of an road transport complex of the Russian Federation, we should
agree that its increase directly is connected with perfection of work of the basic components of
transport system - technical, technological, economic, organizational-administrative, social,
ecological, and also qualitative and quantitative interrelations and interdependence between
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
163
them. Proceeding from principles and rules of the system-target approach, we should recognize
that now the main effect of realization of the Federal target program will reach basically at the
expense of the decision of the most obvious, "simple" system problems, and the further
preservation of positive tendencies of increase of safety of traffic will be defined in a greater
degree by the decision of intersystem problems or, speaking more precisely problems of
intersystem interaction and interdependence.
Examples of multisystem problems, which did not receive the decision:
• Organization and carrying out of the state vehicles technical inspection;
• Traffic management;
• Realization of projects within the limits of “Public-Private Partnership (PPP)”;
• Introduction in full the European protocol by the Russian insurers.
At the same time it is necessary to notice that the program realized till 2012 puts serious
basic bases of the decision of similar problems, forms fundamental bases of the decision of
interdepartmental and intersystem problems that under condition of maintenance of the further
efficiency of the realized actions will provide preparation and performance of the second
Federal target program “Increment of safety of traffic on 2013 – 2020”, focusing attention to
internal motivation to observance of traffic regulations by all its participants, creation of a
favorable climate on roads of Russia.
Outlined in first half of 2010 some delay of rates of positive change of some program
indicators, indirectly confirms “exhaustible” influences of "simple" or monosystem actions on a
traffic safety status.
Realization of mechanisms of integration of federal and regional programs, both within the
limits of the new Federal target program, and with use of the resources put in other federal both
regional programs and projects in sphere of public health services, formation, power,
development of the industry, culture, agriculture can become one of directions of the further
increase of safety of traffic.
Regional indicators of safety of traffic as result of interaction of a wide complex of factors
and their essential dependence on level of social and economic development of region.
The detailed analysis of dynamics of change of indicators of safety of traffic on subjects of
the Russian Federation indicates:
164
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
- Change of indicators of quantity of road accident, numbers of death and wounded men,
social and transport safety, various kinds of road accident with participation of children, road
accident because of drunk drivers differs from 2 to 11 times;
- In a number of regions negative dynamics of change of positive indicators of safety of
traffic is observed;
- A number of regions show stably smaller in relation to average across Russia dynamics of
change of indicators.
It is represented that the reason of similar statistical divergences can be caused as action
subjective (organizational-administrative, professional, administrative), and objective (natureclimatic, economic, social) factors.
Subject of more detailed consideration will be the objective factors caused by features of
given region and in many respects predetermining as level of safety of traffic in region,
dynamics of its change, and set of the methods providing the most essential positive changes of
safety of traffic. Such approach is represented justified, in connection with high expenses for
realization of actions in sphere of safety of traffic taking into account a budgeted deficit, first of
all in regions. It is necessary to notice that the formulated purposes of the Federal program are
focused on achievement of Central European level that taking into account relative indicators of
economic development of EU Member States and the Russian Federation it is represented
enough ambitious and thus executable in practice by a problem. About it followed remember
ardent, but to unsubstantiated critics of the program.
Among the general set of the indicators comprehensively characterizing development of
subjects of the Russian Federation, it is possible to allocate two groups - direct and indirect,
rendering, or able to influence indicators of safety of traffic in region.
The first group of indicators concern: motorization level, the indicators characterizing a
condition of vehicle park, extent and high system density, quantity and number of inhabitants of
settlements, relative charges of a high system, relative sales volume of alcoholic production
through a retail network and some other.
To the second - the indicators characterizing cultural and educational qualifications of the
population, employment, social activity, security services and development of system of public
health services, a crime rate.
The total of indicators of both groups taken from the annual report of the State Statistical
INTERNATIONAL COOPERATION ISSUE OF TRANSPORTATION - Especial Issue - No.04
165
Committee of the Russian Federation, makes nearby 30 and work on formation of such base is
in an end stage.
Estimated, predesigns have shown that for various regions weightiness of those or other
indicators of group fluctuates in essential limits and testifies to presence of interrelations, and
with different quantitative interdependence. End of formation of system of significant indicators
of both groups and carrying out of calculations by definition of their weightiness and
interrelation with indicators of safety of traffic in concrete region, will allow to define priority
spheres and directions of realization of a complex of actions in sphere of the road traffic
management, economy, social development of the region, providing, taking into account a limit
of financial assets, the maximum positive changes in safety of traffic.
The separate attention is deserved by a problem of realization of the detailed analysis of
full statistics of regional road accidents taking into account all fixed parameters on accounting
incidents. Experience of the countries of the European Union shows high effectiveness of
operative actions in sphere of the traffic management and work of divisions similar by our
traffic police in concrete directions, conditions and participants of traffic.
Integration and coordination of separate directions of Federal target programs, dot
financing from regional budgets in actions and the projects providing the maximum return,
taking into account regional features and priorities, are capable to provide synergistic effect in
the field of safety of traffic.
At the same time at sessions of the Governmental Road Traffic Safety Commission the
question of principle about necessity of investment by function of the practical organization of
traffic of uniform body, jurisdiction and which powers would extend on a roads and streets
network of all without an exception of subjects was repeatedly discussed: a federal road
network, an road-street system of subjects of federation, municipal and rural formations.
Occurrence of barriers on "borders", and the most important thing of their consequence have
brightly been shown in the summer of 2010 at overpass repair on 24 kilometer of the MoscowLeningrad highway near the International Airport "Sheremetyevo".
In the conclusion it is necessary to notice that the high lath of indicators of safety of the
traffic, set by the Federal target program, focused on the European indicators of safety, should
be confirmed including by corresponding level of the "European" capital investments in this
sphere, including material compensation of employees of all structures providing safety of
traffic on roads of Russia.
166
INTERNATIONAL COOPERATION ISSUSE OF TRANSPORTATION - Especial Issue - No.04
BOARD OF EDITORS - IN - CHIEF
Prof. V. Prikhodko; Prof.V. Silyanov; Prof. Wanming Zhai; Assoc.Prof. Nguyen Van Vinh
EDITORIAL COUNCIL
MADI’s Editorial
SWJTU’s Editorial
UTC’s Editorial
Prof. A. Buslaev
Prof. Dr. Zhao Yong
Assoc.Prof. Dr.Tran Dac Su
Prof. A. Chubukov
Prof. Dr. Zhai Wanming
Assoc.Prof. Dr.Nguyen Van Vinh
Prof. I. Demyanushko
Prof. Dr. Qiu Yanjun
Assoc.Prof. Dr.Nguyen Duy Viet
Prof. I. Fedorov
Prof. Dr. Li Qiao
Assoc.Prof. Dr.Nguyen Ngoc Long
Prof. V. Gerami
Prof. Dr. Gao Shibin
Assoc.Prof. Dr. Tran Tuan Hiep
Prof. A. Ivanov
Prof. Dr. Peng Qiyuan
Assoc.Prof. Dr. Nguyen Van Long
Prof. L. Makovskyi
Prof. Dr. Pan Wei
Dr. Nguyen Quynh Sang
Prof. A. Nikolayev
Prof. Dr. Li Fu
Prof. Dr. Do Duc Tuan
Prof. V. Nosov
Prof. Huang Nan
Assoc.Prof. Dr. Bui Xuan Cay
Prof. P. Pospelov
Prof. Dr. Liu Xueyi
Dr. Le Hai Ha
Prof. A. Rementzov
Prof. Dr. Zheng Kaifeng
Assoc.Prof. Dr. Nguyen Van Bang
Prof.. M. Shatrov
Prof. Dr. Zhang Jin
Assoc.Prof. Dr. Do Viet Dung
Prof. Yu. Trofimenko
Prof. Dr. Liu Dan
Assoc.Prof. Dr. Le Hung Lan
Prof. M. Ulitskyi
Prof. Dr. Yang Yiren
Assoc.Prof. Dr. Tu Sy Sua
Prof. V. Vlasov
Prof. Liu Bin
Prof. Dr. Sc Nguyen Huu Ha
Prof. V. Zhurakovskyi
Prof. Dr. Song Jirong
Assoc.Prof. Dr. Le Hong Lan
Prof. V. Zorin
Dr. Nguyen Tai Quang
SECRETARY SECTION
MSc. V. Vinogradova
MSc. V. Lipskaya
Assoc. Prof. Lan Junsi
Dr. Nguyen Quynh Sang