TRANSACTIONS
ON ELECTRICAL ENGINEERING
CONTENTS
Arnold, P., Kment, A., Pípa, M., Janíček, F.: On-site Partial
Discharges Measurement of XLPE Cables . . . . . . . . . .
107 – 110
Čapek, J., Sýkora, P., Lenoch, V.: Control of Traction Rail
Vehicle with Free-wheels . . . . . . . . . . . . . . . . . . .
111 – 115
Fikke, S. M.: Climate Change – What Do We Know, What Do
We Not Know, and What May Be the Consequences For
Electric Overhead Line Systems? . . . . . . . . . . . . . . .
116 – 122
Akchurina, S., Tuzikova, V., Tlusty, J.: Optimal Parameters of
Load-Center Supply System for Peripheral Districts of Big
Cities. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
123 – 126
Janiš, R., Kuba, J.: The Possibility of Electronic Equipment
Cooling. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
127 – 129
Vonkomer, J., Žalman, M.: Model Based Design of Electric drives
130 – 133
Vol. 1 (2012)
No.
4
ERGO NOMEN
pp. 107 - 133
TRANSACTIONS ON ELECTRICAL ENGINEERING
Publisher:
ERGO NOMEN, o.p.s., K13114 FEE CTU in Prague,
Technicka 1902/2, 166 27 Praha 6, Czech Republic
E-mail: info@transoneleng.org
Editorial Office:
PIVONKA Pavel
HAVLICEK Radek
MERICKA Jiri
NOVA Ivana
VONDRICH Jiri
ZDENEK Jiri
Periodicity:
Language:
Scope:
On-line version:
Quarterly
English
International scientific journal of electrical engineering
www.transoneleng.org
ISSN 1805-3386
Each paper in the journal is evaluated by two reviewers under the supervision of the International
Editorial Board.
International Editorial Board
Editor in Chief:
Prof. LETTL Jiri, Czech Technical University in Prague, Czech Republic
Members:
Prof. BAUER Palo, Delft University of Technology, Netherlands
Prof. BRANDSTETTER Pavel, VSB-Technical University of Ostrava, Czech Republic
Prof. DOLEZEL Ivo, The Academy of Sciences of the Czech Republic, Czech Republic
Prof. DUDRIK Jaroslav, Technical University of Kosice, Slovakia
Prof. NAGY Istvan, Budapest University of Technology, Hungary
Prof. NOVAK Jaroslav, University of Pardubice, Czech Republic
Prof. ORLOWSKA-KOWALSKA Teresa, Wroclaw University of Technology, Poland
Prof. PEROUTKA Zdenek, University of West Bohemia, Czech Republic
Prof. PONICK Bernd, Leibniz University of Hannover, Germany
Prof. RICHTER Ales, Technical University of Liberec, Czech Republic
Prof. RYVKIN Sergey, Russian Academy of Sciences, Russia
Prof. SKALICKY Jiri, Brno University of Technology, Czech Republic
Prof. VITTEK Jan, University of Zilina, Slovakia
Prof. WEISS Helmut, University of Leoben, Austria
Responsibility for the contents of all the published papers and technical notes is upon the authors.
Template in MS WORD and basic typographic rules to be followed see www.transoneleng.org.
Copyright:
©2012 ERGO NOMEN, o.p.s. All right reserved.
Transactions on Electrical Engineering, Vol. 1 (2012), No. 4
107
On-site Partial Discharges Measurement of XLPE
Cables
Peter Arnold 1), Attila Kment 2),Marek Pípa 3) , František Janíček 4)
Slovak University of Technology in Bratislava, Faculty of Electrical Engineering and Information Technology, Institute
of Power and Applied Electrical Engineering, Ilkovičova 3, 812 19 Bratislava, Slovak Republic, www.fei.stuba.sk
1)
peter.arnold@stuba.sk, 2) attila.kment@stuba.sk, 3) marek.pipa@stuba.sk, 4) frantisek.janicek@stuba.sk
Abstract — The contribution deals with the methodology of
on-line measurement of partial discharges in XLPE cables.
After the specification of measuring methodology and used
apparatus the evaluation of the measured data follows.
Contribution gives an example of a real on-site
measurement on 110kV XLPE cable power line in
operation.
Keywords — Partial discharges, on-line measurement, XLPE
high-voltage cables.
I.
INTRODUCTION
Presence of partial discharges is a key indicator of the
real state of a cable insulation system. By measurement
of partial discharges it is possible to identify them and
take actions to prevent unexpected failure in the power
system. To achieve this goal several methods exist.
When the presence of partial discharges in an insulation
system of medium-voltage cables is investigated,
according to European standard STN EN 60270, using of
a galvanic method is usual. Mentioned methods are
mostly off-line, which means that the line needs to be
switched off. This methodology requires a dischargesfree power source and a suitable analog or digital
measuring system. As soon as the length of investigated
cable exceeds several hundreds of meters, the demand for
the installed capacity of the power source grows. The
measuring apparatus in most cases is supposed to be
mobile, so the needed power supply has larger
dimensions, and weight of course. These facts justify the
search for other than off-line partial discharge monitoring
options.
As said for medium-voltage cables of greater lengths, the
problem is with the discharge-clean power supply
capacity. For high-voltage cables such a power supply
cannot be described as mobile because of weight and
dimension as well. To eliminate the need of huge power
supply for off-line measuring and to enable to investigate
the power line in operation a new methodology is a
necessity.
Recently most interesting methods are that can identify
discharge activity on-line, that means without the need of
switching the devices or parts of power line off. On-line
partial discharges can be investigated by several
procedures:
Spot testing (test time from 15 minutes to few hours per
circuit) is widely used on transmission lines,
Continuous partial discharges monitoring (test time from
several hours to few days) is mainly used on distribution
networks,
Permanent partial discharges monitoring is not widely
used as yet. [1]
II. SPOT TESTING
The advantage of this methodology is that there is no
need to isolate the investigated circuit. Cables can be
tested in operation conditions and it allows an efficient
planning of further investigation and localization of
failure in case, when significant discharge activity is
present.
The measuring apparatus consists of an oscilloscope
based system with four-channel capturing (shown in
Fig.1), filter to eliminate radio frequency interferences,
high frequency current transformers (shown in Fig. 2) and
transient earth voltage sensors (shown in Fig. 3).
The discharges diagnostic unit is based on a high-speed
data acquisition of partial discharges signals at samplerates between 100 and 500 MS/s. Detecting and storing of
the signals at high-resolutions allow to analyze and
classify the pulses based on their true wave-shapes.
Transactions on Electrical Engineering, Vol. 1 (2012), No. 4
Fig. 1: Measuring apparatus
108
Fig. 2: HFTC sensor [2]
The data acquired synchronously with the power cycle of
the applied test voltage enables to save phase distribution
of apparent discharges. High-frequency current
transformers measure partial discharges in the
investigated cable when right attached at the termination
shield. [1]
The transient earth voltage sensor is a small radio
frequency aerial which can detect the high frequency
discharge pulses coming for example from the inside of
the switchgear, bushings, etc. These pulses tend to have
pulse widths of a few tens of nanoseconds, and act as a
good medium for the non-intrusive partial-discharge
detection. [2]
Evaluation of the measured waveforms can be performed
using the PDGold or ScopeControl software. Mentioned
tools utilize the partial-discharge “event recognition” to
find short duration, high-frequency pulses and classify
them as discharges with the origin in the cable, local
equipment or noise.
Cable pulses are characterized as monopole shape
current-impulses detected by high-frequency transientcurrent sensors. They are integrated and their magnitude
given in (pC). They could originate from the cable, cable
termination or transformer.
Local pulses have a large amount of high-frequency
content (>5MHz). They are detected from sources near to
the measurement point by both high-frequency current
transformers and transient earth voltage sensors. The
magnitude of these signals is calculated in dB.
Fig. 3: TEV sensor [2]
Limit values for partial-discharge apparent-charge
observed in XLPE low- to medium-voltage cables and
their terminations in general are summarized in Tab. 1.
For high-voltage cables, of course, there is no tolerable
level of the discharge activity. That means, any discharge
activity detected in these cables require further
localization and repair.
Tab. 1: Limit values of partial-discharge apparent-charge
0 - 500 pC
Acceptable level
500 - 1000 pC
Recommended to monitor
discharges level
1000-2500 pC
Potential risk, recommended the
periodic monitoring
> 2500 pC
Serious risk, location of the source
and subsequent repair needed
III. PROVIDED MEASUREMENTS
Online discharge activity measurements were carried out
as part of a preventive maintenance in order to determine
the status of the high-voltage cable insulation system. The
measurements were performed on two cable lines with
lengths of 430 and 635 meters, each of which consisted of
two parallel cables per phase. Cables have the XLPE
insulation system and composite terminations on both
ends.
A cable shielding on substation side is directly grounded
as shown in Fig.4 on the left; the transformer- side is
Transactions on Electrical Engineering, Vol. 1 (2012), No. 4
109
Fig. 4: Connections of sensors during measurements
connected to ground over a varistor as shown in Fig.4 on
the right.
The cables were tested from both ends. The discharge
signals were measured using high-frequency current
transformers placed around the cable shielding followed
by the transient earth voltage sensors located at the
bottom terminals. The sensors were fitted separately for
each pair of cables for each phase.
Figures in Tab.2 show the observed discharge activity
consisting of interferences at frequencies of 270, 540 kHz
and 1,1 MHz together with a noise during measurements.
After filtering the noise out from the acquired signals and
making corrections to discharge activity resulting from
substation electronic device interference captured by the
capacitive transient earth voltage sensors, the resulting
phase-distribution of partial-discharge apparent charge
had the amplitude below 0,5 pC. That means this time no
harmful partial-discharges were detected in tested objects.
need to use an external low-discharge power supply; the
whole apparatus is mobile, in comparison with galvanic
off-line methods less complex, less susceptible to
interferences and noise. Nowadays, for periodic testing of
supported high-voltage XLPE cables this method seems
to be most effective. However, it does not enable exact
localization of a failure in case of a higher observed
apparent charge in phase-distribution. To reach this goal,
still usual off-line methods are required providing more
diagnostic parameters, as these in insulation system
capacity or loss factor.
ACKNOWLEDGEMENT
These publications are the result of implementation of the
project: “Increase of Power Safety of the Slovak
Republic” (ITMS: 26220220077) supported by the
IV. CONCLUSION
As contribution briefly describes, on-line measurement of
the partial-discharges, more precisely so called “Spot
testing” is a method, which almost does not affect the
operation of investigated power system elements. It is no
Research & Development Operational Programme funded
by the ERDF Supporting the research in Slovakia. The
project is co-financed by EU funds.
Transactions on Electrical Engineering, Vol. 1 (2012), No. 4
110
Tab. 2: Limit values of partial-discharges’ apparent-charge
0.03
0.04
0.02
0.02
0.01
0
-0.02
Local Equipment PD
P D M a g n itu d e ( m V )
0.06
C han 1
C han 2
Available Waveform Display
0
-0.01
-0.04
-0.02
-0.06
-0.03
0
2
4
6
8
10
12
Time (mSec)
Chan 2
14
16
Curs 1
18
20
45
40
35
30
25
20
15
10
5
0
0
Curs 2
Partial-discharges in time-domain (HFCT)
50
100 150 200 250 300
Phase of Pow er Cycle (deg)
350
Partial- discharges in phase-domain (TEV)
Noise Events
N ois e M agnitude (m V)
25
20
15
10
5
0
0
50
100 150 200 250 300
Phase of Pow er Cycle (deg)
Noise in phase-domain (HFCT+TEV)
350
Partial-discharges in phase-domain (HFCT-TEV)
REFERENCES
[1]
Seltzer-Grant, M., Renfort, L., Mackinlay, R., Denisov, D.,
Madarasz, R., Schlapp, H., Petzold, F.; “Portable Online PD
spot Testing and Monitoring Technology for Distribution Class
Cables” 2011. ELEKTROENERGETIKA.
[2]
http://www.hvpd.co.uk/products/sensors. [Online] 20. 4 2012.
The contribution was presented on the conference ELEN 2012, PRAGUE, CZECH REPUBLIC
Transactions on Electrical Engineering, Vol. 1 (2012), No. 4
111
Control of Traction Rail Vehicle with Freewheels
Jan Čapek 1), Petr Sýkora 2) Václav Lenoch 3)
1)
VÚKV a.s., e-mail: capek@vukv.cz
University of Pardubice - Jan Perner Transport Faculty - Department of Electrical and Electronic Engineering and
Signalling in Transport, Pardubice, Czech Republic, e-mail: petr.sykora@upce.cz
3)
University of Pardubice - Jan Perner Transport Faculty - Department of Electrical and Electronic Engineering and
Signalling in Transport, Pardubice, Czech Republic, e-mail: vaclav.lenoch@upce.cz
2)
Abstract—This paper is focused on the low-floor trams
with free-wheels driven by PMSM motors, especially on
their traction control system. The first part describes
modern low-floor trams, particularly the type 15T produced
by Škoda Transportation. The second part of the paper
presents the simulations of characteristics operating such
tram in the rail and the results of these simulations. The
last part of the paper describes the mechanical construction
and electrical equipment of a light experimental rail car,
that will be used as an experimental base for a improvement
of the control system for the tram 15T.
Fig. 1 Partially low-floor tram
Keywords—low-floor tram, free-wheel drive, gearless drive,
PMSM, control system, experimental vehicle.
I. MODERN LOW-FLOOR TRAMS
Tram vehicles have
some specific elements in
comparison with standard rail vehicles. Streetcar tracks
are led through the streets of urban development, often not
separated from car traffic, characterized by the presence of
curves of very small radius (up to 20 m) and steep
inclines. Challenging tracing of tramlines makes the
construction of tram vehicles quite difficult. In addition to
further tram improving it is now required a low-floor
vehicle, which further increases the difficulty of the
structural design of these vehicles and forces some other
specifics. The use of low-floor vehicles in the public
transport system allows not only convenient travel for
citizens with limited mobility or passengers with wheel
chairs but also significantly speeds up the exchange of
passengers at tram stops.
The first low-floor tram vehicles began to be produced
in the second half of the 80’ of the 20th century. Since
then, there was designed a large number of low-floor
trams of different solutions to achieve the highest
proportion of low-floor space. A
high ground is
considered to be a floor mounted at a height of about 600
to 900 mm above the rail, a low floor is then at a height of
about 350-450 mm above the rail. The low-floor vehicles
are divided into two categories - semi low floor and fully
low floor. Typical arrangement of low-floor trams are
illustrated in Fig. 1 and 2. The proportion of the low floor
with semi low-floor trams ranging from about 15% (Fig. 1
top) to 75% (Fig. 1 below), in the case of a fully low-floor
vehicles, the low floor is located along the passenger
lounge.
Fig. 2 Fully low-floor tram
Location of low-floor trams bogies requires the use of
special bogies. For the partially low-floor tram there are
often used normal (non-driven) bogies and as a drive
standard chassis located above the floor in a standard
height. For full low-floor trams all the bogies, including
drive, are then low-rise.
A common feature of almost all types of low-floor
bogies is the use of independently rotating wheels instead
of conventional wheel sets. In case of drive bogie, these
‘axles’ are driven by a pair of longitudinally stored motors
with double-sided output or each wheel is independently
driven by one motor.
Behavior of independently rotating wheels moving in a
dorm is different from that of conventional wheel sets.
While the wheel set is used from the very beginning of
rolling stock and therefore the behavior in rail is well
known, independently rotating wheels on rail vehicles is
relatively new element that needs to be continually
examined in detail.
Using independently rotating wheels in bogies of trams
was elicited from the low-platform. However, it can also
be used for other important improvements of vehicles - in
the case of independently rotating wheels when each of
them is driven by its own motor it is possible to control
each motor to improve road holding, in particular to
reduce wear on wheels and rails and increase vehicle
safety against derailment.
Transactions on Electrical Engineering, Vol. 1 (2012), No. 4
II. TRAM ŠKODA 15T
Škoda Transportation in cooperation with Škoda
Electric has designed and currently manufactures factorytype 15T trams known under the trade mark ForCity (see
Fig 3). These trams - designed specifically for the
Transport Company of the Capital City of Prague - had
already at the design stage to meet very demanding
requirements of the contracting authority. From our
perspective, this is a multistage design with rotating
bogies and drive on all wheel sets and at the same time the
100% low-floor vehicle without any stairs balancing of
level floors inside the vehicle.
Fig. 3: Tram 15T ForCity
These requirements have led manufacturers to apply up
to now seldom used traction drive concept: the wheels of
the trams are no longer linked to a common wheel set in a
rigid axle driven by one traction motor, tram-type 15T
wheel drive was solved individually using low-speed
synchronous motors excited by permanent magnets
attached to wheels without the use of a gearbox, only by a
flexible coupling.
The chosen drive concept enabled to meet the
requirements of the vehicle when entering the contract,
but later on it also brought unsolved problems. The
concept of the drive using sixteen liquid-cooled
synchronous motors excited by permanent magnets
supplied from sixteen voltage source inverters, causes
increasing complexity of the vehicles and places
significant demands on the control system and data
communication between its various components.
An innovation is mainly given by the absence of
classical rigid wheel set in terms of roadholding. We
could simply say that it is advantageous for solid axle ride
in a straight line, which stabilizes the vehicle profitably,
but while driving arc, however, a sweeping wheels on the
rail are causing undesirable tread wear and noise. Using
individual wheel drive provides in this respect a new
potential and risk.
When driving it is possible to prevent an arc sweeping
wheel on the rail, which can dramatically reduce both the
wear of wheel and rail and noise too. In contrast, in a
straight line it does not reach optimal vehicle driving, in
extreme cases it can cause lower safety against derailment.
All these mentioned features, however, can be influenced
by appropriate management of traction drives.
112
Since the Department of Electrical Engineering,
electronics and security systems in transport of Jan Perner
Transport Faculty of the University of Pardubice
participated in the development of tram traction drive 15T
[1], [2], collaboration resulted in another research project,
this time focusing on optimizing control algorithms
mentioned on vehicles with free driven wheels.
As it would be the experimental verification of the
changes effects upon the control software running on a
real tram considerably complicated, we proceeded in
cooperation with Škoda Electric and VÚKV to build an
experimental vehicle. But the first part of the research is
the simulation of the changes influence upon the control
software.
III. SIMULATIONS
The MBS simulation software is used for an analysis of
a rail vehicle dynamic behaviour during running on a
track. In this software a virtual 3D vehicle model is
created, which consists of mass elements (car body, bogie
frames, wheel sets, etc.) that are connected to each other
by kinematic joints (e.g. a rotational joint connecting a
wheel set with an axle box) or force elements (e.g.
suspension elements). After the vehicle model is created,
the software builds up itself equations of motion of the
mass elements and equations of the joints. The resulting
system of differential-algebraic equations is solved
numerically. Adams by the MSC.Software Company and
Simpack by the company of the same name are the most
popular MBS simulation softwares. Fig. 4 shows an
example of the calculation model of a low-floor tram
created in the Adams software.
Fig. 4 Calculation model of a tram in Adams software
Nowadays, the simulations are an inseparable part of
the development and research in the field of railway
vehicles.
The wheel-rail joint plays an important role in the
vehicle running behaviour. Thus, a mathematical model of
wheel-rail contact is one of the most important and also
the most complex elements in the MBS simulation
software. The software enabled using of only a very
simplified linear model with one point contact only in the
past. Today, the simulation software contains a general
nonlinear multipoint contact model. The multipoint
contact between wheel and rail occurs when the vehicle is
running in a curve of a small radius. Thus, such a complex
model of a wheel-rail contact enables very detailed
analysis of a running behaviour of tramcars even in
extreme conditions, such as a vehicle run in a curve of a
very small radius. Real wheel and rail profiles are
considered in the calculation models – Fig. 5 shows an
example of a wheel-rail contact of a tram running in a
Transactions on Electrical Engineering, Vol. 1 (2012), No. 4
straight track (up) and in a curve of a very small radius
(down), with a contact point visualisation. In Fig. 6 the
two-point wheel-rail contact in a curve is clearly to be
seen (a grooved rail is considered).
113
Behaviour of the vehicle running in a track with
irregularities show results of performed simulations
shown in Fig. 7. The figure shows a dependency of lateral
displacement of a wheel set (blue line) and independently
rotating wheels (red line) on a travelled distance. The
irregularities are on a track section from 100 to 800 m. It
is obvious that independently rotating wheels move
laterally with a higher amplitudes while a wheel set is
forced to run in a central position in a track. The
significant lateral movements are undesirable because
they are associated with a higher wheel and rail profile
wear. The graph also shows that after exiting the
irregularities the wheel set moves to the centre of a track
while the independently rotating wheels remain running
laterally shifted.
Fig. 5 Wheel-rail contact in straight track (up) and in curve (down)
With the calculation model of a tram (see Fig. 5)
simulations were performed in order to compare a running
behaviour of the vehicle with rigid wheel sets and
independently rotating wheels. Two basic calculation
cases were considered: run in a straight track and
negotiation of curves of several radii.
The wheels of a rigid wheel set are forced to rotate with
the same angular velocity. When the wheel set is laterally
shifted in a straight track, the radii of rolling circles of the
left and right wheel are different (a conical wheel profile
is considered). This results in the occurrence of
longitudinal slips in the wheel-rail contacts, and therefore
also a pair of longitudinal slip forces. The slip forces
create torque acting on the wheel set that returns the
laterally shifted wheel set back to the centre of the track.
Hence, the wheel set is naturally centred in a track (see
Fig. 6).
Fig. 7 Lateral wheels displacement during vehicle run in a straight track
with irregularities
When a vehicle with wheel sets is running in a curve,
generally three different situations may occur: 1) the
wheels roll without longitudinal slips, 2) longitudinal slip
force acts on the outer wheel in the curve in the direction
of longitudinal movement of the wheel set, on the inner
wheel in the opposite direction, 3) longitudinal slip force
acts on the outer wheel in the curve against the direction
of longitudinal movement of the wheel set, on the inner
wheel in the opposite direction. Which one of these three
situations occurs depends on the curve radius, lateral shift
of the wheel set in the track, wheel rolling radii and their
difference. Which one of these three situations occurs has
got an influence on the size of the forces acting in contacts
between wheels and rails and also on the wheel and rail
profile wear. The wheel set in a curve is shown in Fig. 8.
Fig. 6 Wheel set laterally shifted in a track
Independently rotating wheels are not connected in a
torsion way to each other. Thus, they can rotate
independently with different angular velocity. When the
wheels are laterally shifted in a track, no longitudinal slip
forces are generated. Hence, no forces pushing the wheels
back to the central position are acting on the wheels.
Fig. 9 Wheel set in a curve
In a case of independently rotating wheels in a curve,
there are no longitudinal slip forces acting in the wheel
Transactions on Electrical Engineering, Vol. 1 (2012), No. 4
rail contacts that would influence the wheel and rail
profile wear.
Three bogie configurations were considered in the
calculations: 1) a bogie with two wheel sets, 2) a bogie
with two pairs of independently rotating wheels, 3) a
bogie with one wheel set (in front – the first in the
direction of travel) and a pair of independently rotating
wheels. The results of calculations show that the lowest
wheel and rail profile wear occurs for these
configurations: a bogie with two pairs of independently
rotating wheels in a curve of a very small radius; a bogie
with one wheel set and a pair of independently rotating
wheels in a curve of a large radius. Fig. 9 shows an
example of calculation results – profile wear index for the
three configurations of a bogie passing a curve of a very
small radius.
114
below the main frame near the third axle. . It is located on
the main frame longitudinal table and after the
longitudinal sides of the bench for operators. The entire
vehicle is then covered with lightweight metal roof - see
Fig. 10.
Fig. 10 Mechanical concept of experimental vehicle
The vehicle is equipped with two independent braking
systems. Service brake is electrodynamic regenerative
brake. Parking brake and emergency brake are hand
screw, which is used for third common axle.
V. ELECTRICAL EQUIPEMENET THE OF EXPERIMENTAL
VEHICLE
Fig. 9 Profile wear index for the three configurations of a bogie passing
a curve of a very small radius
The results of the calculations can be used for designing
algorithms for controlling an individual drive of
independently rotating wheels. The wheel drive is
currently being implemented into the calculation model
and will be further developed and optimized. The results
will be verified by experiments with the narrow gauge
experimental vehicle.
IV. MECHANICAL CONSTRUCTION OF THE EXPERIMENTAL
VEHICLE
The vehicle is designed as a narrow gauge of 600mm.
Use of this gauge allows to reduce significantly
procurement costs, while offering the possibility to use as
a test track Mladějov industrial railway, which is not in a
regular week-long operation, moreover, this track has also
a very complicated directional and vertical alignments,
that are desirable for the experiments.
Further construction of the experimental vehicle has
been subordinated closer to the tram-type 15T, or rather
to one of its bogies. Therefore, the selected three-axle
design, where the main element is a rotating bogie, where
four wheels are connected by flexible couplings with four
traction motors was assumed. Traction motors with the
wheels are suspended from the bogie frame by conrods
equipped with strain gauge sensors of operating forces this solution will enable to measure the impact of control
algorithms, the forces that act on the wheel when driving
in a straight line and arc. Said wheel chassis is bound
under the main frame, which is partly carried by non
powered third axle, realized by standard solid wheel set.
On the main frame there is positioned first electrical
switchboard, where all electrical equipment of the vehicle
is located except traction motors and batteries hanging
Because there is no trolley line in Mladějov industrial
railway, the vehicle is built as a battery-equipped. With
regards to traction drive and the estimated consumption of
energy the voltage traction battery 96 V = was chosen.
This battery is composed of 8 traction lead-acid batteries
of the expected capacity of 150Ah@C5.
The wiring diagram is shown in Fig. 11. TBAT traction battery is connected to the main breaker Q. It
provides both safe disconnection of all electrical
equipment from the vehicle battery and also serves as a
emergency switch in case of an accident. As the main
circuit breaker it is mounted fuse disconnector F main,
staffed with 2x125A/aM fuses, followed by a DC bus to
which other circuits are connected. First of all, it is
through the circuit breaker FA CH (16A/C) connected
traction battery charger the used type is AXIstand 96-25
from the firm AXIMA, which is a CPU-controlled fully
automatic charger. Power supply of the charger is realized
from the normal network 3x400 V ~ / 50 Hz.
Fig. 11 Simplified diagram of the power circuits of the traction
electrical equipment
Furthermore, on the DC bus the circuit of the self
consumption of the vehicle is connected through a circuit
Transactions on Electrical Engineering, Vol. 1 (2012), No. 4
breaker FA CTRL (8A/C) . In the said circuit breaker LC
filter is connected together with DC-DC converters
providing power supply of non-traction circuits. Through
a special circuit breaker it is also connected the DC/AC
inverter 96 V = / 230 V ~ 50 Hz as a power supply for
computers and other devices necessary to ensure test runs.
Finally, the DC bus is connected through a circuit
breakers FA TP1 ÷ 4 (25A/C) to four traction drives.
These are implemented as voltage IGBT inverters. The
input capacitors are therefore loaded by auxiliary relays,
which are bridged after charging by contactors K 1 ÷ 4.
Above the contactors there are then placed current sensors
ITP 1 ÷ 4. The inverters are build by common modules
from the firm. Semikron, type SK 75 GD 066 T. These
modules are again over current sensors connected to
traction motors type AKM 74P with integrated position
sensors.
115
requirements, but it is also robust enough for use on
rolling stock. Its other advantage is its programming in
LabView graphical language, resulting in very efficient
and intuitive creation of control algorithms including their
easy modification. This system will be also used to
collect data during the test runs. The structure of the
vehicle control system is shown in Fig. 12.
VI. CONCLUSION
Currently it is done the mechanical part of the
experimental vehicle and traction inverters are developed .
During the year 2012, it should be the completed the
experimental vehicle , so in the following year the driving
tests and self-optimizing control algorithms could be
done. In 2014 the results of the research should be applied
to real trams. We believe that this research project will
help to improve the operational characteristics of the
mentioned tram as well as to fully exploit the new
opportunities that this unconventional method of
propulsion offers.
For more information about the simulations and
experimental vehicle see [3].
VII. ACKNOWLEDGMENT
The paper was created within supplement of grant
project of the Technology Agency of the Czech Republic
TA01030391 "Research driveability and traction drive
control of vehicles with independently rotating wheels."
REFERENCES
Fig. 12 Simplified diagram of the vehicle control system
Management at the level of traction drives is realized
through four two-desk controllers Škoda, that evaluate the
voltage and current waveforms together with the positions
of the traction motor rotors and calculate individual
power-switching of transistors. Connection between
regulators and transistors is realized by a compact exciter
Semikron SKHI 61st.
As a master controller it is used a modular control
system Compact RIO from the firm National Instruments.
This system provides due to its modularity imposed
[1] J. Novák, O.Černý, Simanek, J. „Control of synchronous motor for
light rail traction“, ELECTRIC. 2008, Vol. 18, No. 6, p 4 to 10
ISSN 1210-0889.
[2] O. Černý, R. Doleček, J. Novák, J., Šimánek „Energy and
performance characteristics of traction drive with synchronous
motor with permanent magnets“, XXXI. national conference on
electric drives in Pilsen, [CD-ROM]. Czech Electrotechnical
Society, 2009, ISBN 978-80-02-02151-3.
[3] J. Novák, M. Lata, O. Černý, P. Sýkora, J. Čapek, J. Malinský, J.
Mašek, J. Bauer, L. Sobotka: „Design and construction of the
experimental vehicle with independently rotating wheels and the
results of the relevant theoretical and simulation work“, annual
interim report for 2011 of the project TA01030391 "Research
driveability and traction drive control rail vehicles with
independently rotating wheels", University of Pardubice 2011
Transactions on Electrical Engineering, Vol. 1 (2012), No. 4
116
Climate Change – What Do We Know, What Do
We Not Know, and What May Be the
Consequences For Electric Overhead Line
Systems?
Svein M. Fikke
Meteorological Consultant – Overhead Lines Lindeveien 1, NO-1470 Lørenskog, Norway.
e-mail: fikke@metconsult.no
Abstract — The validity of global climate being
influenced by emissions of CO2 and other greenhouse gases
from human activities is now globally recognized. Although
the global temperature is expected to rise, and consequently
also precipitation amounts, many secondary effects are still
uncertain, concerning for example flooding, storm
frequencies, atmospheric icing and so on. However, many
electrical utilities around the world are already considering
preventive measures for their network, based on the
philosophy that proactive measures in the long run are
cheaper than taking extra costs for maintenance and repair
after expected increases in damage and outage frequencies.
Keywords — Global warming, flooding, wind storms,
atmospheric icing, forecasting of critical events.
Figure 1. CO2 concentrations measured since 1958 on Mauna Loa
Observatory, Hawaii (NOAA Earth System Research Laboratory).
Red curve is winter-summer fluctuations.
I.
INTRODUCTION
Since the industrial revolution started a century and half
ago, and the required energy for the rapid developments
of factories and new machines were produced by burning
of coal, oil and gas, which consequently resulted in
emissions of CO2 into the atmosphere. Figure 1 shows
the time series of CO2 measurements from the most cited
observatory on Mauna Loa on Hawaii from the start of
these measurements in 1958. Due to the greenhouse
effect of this and other gases the atmospheric temperature
has increased in parallel with the CO2 curve during the
same period as demonstrated in Figure 2.
Despite series of lively attacks from deniers of human
influence on global climate in mass media within many
countries, there has been a consensus within scientific
communities since 1995 on this relation [1].
However, on the other hand many questions on secondary
effects arising from this temperature increase, with
respect to other weather elements like precipitation
Figure 2. Annual global temperature anomalies (red columns, from
1901-2000 average. Grey columns: error bars.) 1880 – 2011
(NOAA National Data Climate Data Center).
intensities and distribution, flooding, droughts, wind
storms, hurricanes, tornadoes, avalanches, permafrost,
etc., still remain more or less unsolved. The only
consequences of temperature rise which are quite likely
are that the atmosphere will take more water vapor with
increasing temperature and also that the sea level will
Transactions on Electrical Engineering, Vol. 1 (2012), No. 4
117
Figure 3. Historical and projected seasonal temperature scenarios for Praha – Klementinum over the period 1900 – 2100. Dots are
measured values and thick read line model averages. Pink area represents the error bands of models. Note different scale in DecemberFebruary diagram (upper left).
rise. Consequently, there are also very high probabilities
for increased precipitation amounts and also precipitation
intensity, summer and winter, in addition to higher
variability in extremes, both with respect to flooding,
drought and fire risk.
Although it is not yet possible to detail the developments
in extreme weather, an increasing number of scientist
reports indicate that severe weather events may increase
in parallel with the atmospheric temperature rise. In order
to summarize the state-of-art of the scientific knowledge
in this field, the UN Intergovernmental Panel on Climate
Change (IPCC) recently published a Special Report on
Managing the Risks of Extreme Events and Disasters to
Advance Climate Change Adaption (SREX) [3]. The
main purpose of this report is to raise the attention level
among policymakers and stakeholders to potentially
exposed areas of the world and to encourage adaption
methods and mitigation procedures at an early stage, in
order to minimize economic losses and human strains.
Based on such concerns numerous electrical utilities
around the world are watching this situation closely for
the safe operation of their electric grids, as will be
exemplified in this paper.
Transactions on Electrical Engineering, Vol. 1 (2012), No. 4
II. CLIMATE PROJECTIONS FOR CZECH REPUBLIC
The Ministry of the Environment of the Czech Republic
has issued a Brochure on effects forming climate change
on the Czech Republic [4] where also the power industry
is discussed. However, this discussion relates mainly to
the production of energy from renewable sources. Many
temperature projections are certainly available for the
Czech Republic. However, it appeared easier to go
through the web site of the Norwegian Meteorological
Institute (www.met.no) where temperature projections
were presented through Google Earth for hundreds of
locations around the world [5].
An example of historical and projected seasonal
temperature is shown for Praha – Klementinum over a
200 year period in Figure 3. According to this figure the
average temperature in Prague is expected to rise roughly
about 3 °C for all seasons during our century.
It is unfortunately not yet possible to downscale
projections of other weather phenomena to such a degree
that they are of practical value for an objective discussion
of potential effects on the electric grids. Due to these
limitations the further discussion must therefore be
generalized and based on experiences and knowledge
about how harsh weather do influence the impacts and
stability of our electric power overhead line grids.
118
III. SOME CONSIDERATIONS CONCERNING ELECTRIC
POWER GRIDS IN THE CZECH REPUBLIC
A. General
Electric power overhead lines are on a global scale
generally subject to impacts from weather related
phenomena like:
• high wind speeds,
• ice loadings (wet snow, rime ice, freezing rain),
• lightning,
• pollution,
• avalanches,
• landslides,
• flooding
• forest and grass fires,
• tornadoes and tropical cyclones,
• sea level rise
• high temperature (thermal rating)
• snow depth (lattice towers),
• cloudiness (for helicopter based maintenance
operations), and
• ground water level (foundations).
However, it is frequently stated in climate projections that
the variability is likely to increase too [3]. This means
that, for instance, if the temperature generally increases,
cold spells and seasons may very well be as we have
always seen, although they may occur less frequently.
This applies to any other weather event as well. The
IPCC SREX [3] states that:
All these weather related elements may be influenced by
changes in atmospheric temperature, as briefly outlined in
the previous section. Weather elements like lightning,
pollution, avalanches, landslides, snow depths on ground,
flooding, etc., are subject to national and regional
analyses on a broader scale. In this article only
atmospheric icing (wet snow and rime ice) can then be
discussed in general terms based on experiences and the
knowledge about physical processes in the atmosphere.
Further details about icing conditions in the Czech
Republic are published in the COST Action 727
“Atmospheric icing on Structures”, State of the Art
Report from 2007 [6]
A changing climate leads to changes in the frequency,
intensity, spatial extent, duration, and timing of extreme
weather and climate events, and can result in
unprecedented extreme weather and climate events.
Additionally, some information from other countries is
given, based on the author’s current knowledge. For
further reading about atmospheric icing the Cigré TB 291
is recommended [7].
Since the winter temperature will remain above the
freezing point for a longer time further along the century,
it means that there will be on the average less snow fall
and hence shorter periods with snow on the ground as
well.
This is however a global statement, but it emphasizes the
fact that it is strongly recommended to look at future
climate with an open mind for potential extreme events of
a different kind that could not be foreseen from historic
experiences forming the area.
B. Wet snow
As already stated, the average air temperatures during
winter are likely to increase by around 3 °C, and
accordingly there will be less snowfall on the average,
and accordingly fewer events with wet snow in the
lowlands of Czech Republic. This may indicate that there
Transactions on Electrical Engineering, Vol. 1 (2012), No. 4
In higher altitude areas where dry snow frequently occurs
in past decades, the rising temperature may probably lead
to more wet snow events, when the snow occurs at
temperatures closer to 0 °C, and therefore wet snow
accretions may become more likely in the future. If this
assumption will be correct, then it is also likely that the
design loads (with a certain recurrence period, e.g. 50
years) may increase as well for altitudes above, say, 500
m above sea-level
indicator of climatic changes for the combination of
temperature and cloud humidity.
25
S tu d n ic e 8 00 m , n = 5 9
-1
Q [kg.m ]
will be fewer events with electric disturbances in the grid
during winter. However, when wet snow may occur, the
disturbances may be just as large and widespread as
before. For the design ice loadings it is not possible to
conclude on any trend, neither in magnitude nor in
frequency for such events.
119
20
15
10
5
0
4 0/41
50/51
60 /6 1
7 0/71
8 0/81
90/91
t (ye ars)
Figure 4. The longest continuous time series of ice load
measurements in the world. Mt. Studnice, 800 m above sea-level,
Czech Republic.
Measurements of wet snow are very scarce throughout
Europe, except for Iceland [7]. However, it is nowadays
getting possible to calculate wet snow from regular
weather observations by the use of wet snow accretion
models, as has recently been done for the UK [8].
C. Rime ice (in-cloud ice)
Following the increasing air temperatures, it is quite
certain that clouds will contain more water per unit
volume cloud air on a general basis. Following this, it is
then very likely that rime icing will increase accordingly
for electric overhead lines in mountains which are
exposed to such icing. Hence, it is also then very likely
that the design values (return period 50 years) will
increase as well for electric overhead lines exposed to
such icing, unless the temperature increases so much that
the zero-isotherm will remain above the altitude of such
lines for longer periods.
In the Czech Republic we find however the longest time
series of ice load measurements in the world. Such
homogeneous measurements were taken since 1940/41on
Mt. Studnice, 800 m above sea-level. Figure 4 is taken
from [6] and shows the large variability of ice loads
which has to be expected for such meteorological
phenomena. These measurements were initiated by Mr
František Popolanský of EGU, Brno, and comprise a
unique time series which has been widely shown and
discussed on the international arena where atmospheric
icing ever was an issue.
It is strongly recommended that these measurements are
to be continued, since they may also be a very good
Figure 5. 39 years of ice loadings measured on 10 test spans in
Iceland.
In Iceland they have measured wet snow and rime icing
on automatic test spans since 1972, and have therefore a
time series of nearly 40 years of data from numerous test
spans spread over the country. Annual maxima for the 10
spans operated continuously since 1972 until 2010 are
shown in Figure 5.
It can be seen from Figure 5 that there were some
extreme events around the turn of the century, but also a
slight increasing trend over the last decade or so. Whether
this trend will continue is impossible to say, but the figure
shows, similarly to the Czech measurements in Figure 4
that such data will be very important for the future.
To know more about rime ice loads it is necessary to
monitor such icing on test sites. Atmospheric icing is
generally not a topic for regular weather observations,
therefore it must be monitored specifically by those for
whom this is an economic issue. This applies even more
for rime ice than for wet snow, since it is not possible to
model rime icing as easily as for wet snow from other
regular weather observations. To model rime ice it is
basically necessary to scan through long time series of the
physical atmosphere states in terms of gridded values of
Transactions on Electrical Engineering, Vol. 1 (2012), No. 4
120
all meteorological parameter, as done by regular weather
forecasting models.
However, modified versions of such weather forecasting
models are now available for rime ice studies as shown in
[8]. Such models can also be used for studies of historical
events as well as for detailed forecasting of adverse
weather in the operation phase of electric overhead lines.
d.
2.
IV. INTERNATIONAL CONCERNS AND ACTIONS
The information in this chapter was collected from Cigré
colleagues in preparation of the Tutorial at the Cigré
Study Committee B2 meeting in Reykjavik 2011 [9].
Individual informants are mentioned for each country
below. The lists of items are not complete for most
countries, but some highlighted topics of particular
attention are listed for each country.
3.
4.
A. Australia (Henry Hawes, in collaboration with
Energy Networks Association (ENA))
ENA – Key issues are mentioned as follows:
1. Climate change is emerging as a major issue for
operation of networks in relation to temperature,
heat wave, flooding, wind, and fire weather.
2. Changes in single events (cyclones and flooding) as
well as shift in frequency and intensity of weather
regimes.
3. Intensity increase of East coast cyclones.
4. Hail risks associated with severe thunderstorm
activity in Eastern Australia.
5.
Precipitation
a.
River erosion, flooding
b.
Mudslides
c.
Increased corrosion
d.
Higher frequency and severity of wind and
ice storm, and hail storms
e.
Reduced opportunity for live-line work (due
to more rainy days)
Wind
a.
Changes in wind speeds and prevailing wind
direction will affect failure rates, recovery
time and reliability
b.
Need to adjust vegetation control practices
c.
Wind withstand levels of hardware will
need to be increased
Other effects
a.
Rising sea level
b.
Melting permafrost
c.
Maintenance and structural integrity of
transmission lines could be affected
d.
Increased lightning activities
e.
Fog increase means more in-cloud icing and
reduced line access
f.
Transmission line ratings could be affected
Adaption measures
a.
Driver for technology developments
(robotics, remote sensing)
b.
Specialized weather prediction
c.
Modify maintenance and design
d.
Dynamic thermal rating
e.
Corrosion resistant material
f.
Review emergency response
g.
Probabilistic techniques to assess reliability
5. More frequent and intense droughts and heat waves.
6. Bush fire risk (“Extreme fire weather” may increase
with 100-300 days).
B. Canada (Dr. Janos Toth, BC Hydro, R&D)
Studies are performed in collaboration with Environment
Canada, University of British Columbia, Pacific Climate
Change Consortium, University of Alberta, and others.
Areas for particular attention are identified as follows:
1.
Temperature increase
a.
Beetle infestation
b.
Drier summers (fire risks, slope stability)
c.
Woodpeckers
Changes to energy and peak load
consumption patterns
C. Norway (Svein M. Fikke, with input from the
Norwegian Meteorological Institute (met.no)
and the Norwegian Geotechnical Institute
(NGI))
Norway is a coastal country, stretching from 58° to more
than 71° N, with the warm North Atlantic Current along
its coast. The climate varies from temperate in the south
to arctic in the north. The climate predictions for Norway
Transactions on Electrical Engineering, Vol. 1 (2012), No. 4
depend also on the routes of the extra tropical cyclones,
how often they will go south of Norway (into the
Skagerrak Sea) or northwards along the coast. However,
the following scenarios are likely:
1.
Atmospheric icing depends on wet snow and rime
ice
3.
D. United Kingdom (Dr Brian Wareing, Brian
Wareing.Tech, in collaboration with UK Met
Office)
Some examples of diagnostics are:
1.
Evaluating the change in risk of high wind and
wet snow accretion on overhead conductors
a.
North Norway – More wet snow inland,
little rime ice
2.
Estimating the change in rating associated with
low wind and high temperature
b.
Along the coast (southwards from N.
Norway) – Less frequent wet snow little
rime ice
3.
Calculating the change in the seasonality of
demand due to, for example, increased use of air
conditioning
c.
Central mountain range – More frequent wet
snow and higher extremes, less rime ice
below 900 m above sea level , more rime ice
above 900 m above sea level
d.
2.
121
SE Norway – More frequent wet snow, rime
ice only above 1 000 m above sea level
E. CIGRÉ
In the Technical Brochure 291 [3] Cigré WG B2.16 states
the following concerning atmospheric icing in general
terms:
Avalanches
a.
Transition from dry to wet snow avalanches
below 800 – 1 000 m above sea level
b.
Higher areas – More snow and dry
avalanches give longer discharge ranges
c.
Known avalanches will increase in size
d.
More often wet snow avalanches and mudslides due to more and intense rainfall
e.
“Safe” areas may become unsafe
1.
Coastal areas: More seldom wet snow in
lowlands, may be more in the mountains
2.
Inland areas with cold climate: Wet snow may
increase in frequency and intensity at all
elevations
3.
Mountain areas: Risk of rime ice may decrease in
lower levels and increase in higher levels
4.
Freezing rain: Not possible to evaluate yet.
Operation and maintenance
a.
Large parts of the transmission system pass
through exposed mountain areas
b.
Helicopter is often the only tool to access
such lines, especially in winter
c.
Higher frequency of low clouds and extreme
weather will reduce weather windows for
maintenance
Russia (Sergey Chereshnyuk. VNIEE)
Russian topics of concern are
1.
Temperature increase of about 5 °C in Russian
arctic regions already recorded
2.
Serious damage to buildings due to melting
permafrost
3.
Permafrost area reduction of 15-30 % (by 2050)
4.
Flooding and landslide
5.
Wind loads will decrease in some regions and
increase in others
V. CONCLUDING REMARKS
As mentioned in the Introduction, there are many
questions and research issues which still remain
unsolved. Also, among the more settled issues there are
still a lot of smaller or bigger uncertainties which are
subject to lively discussions in the scientific
communities. In particular, major uncertainties remain as
to the development of greenhouse gas emissions around
the world.
It may seem frustrating that there is so little development
in international agreements and binding actions to
stabilize and reduce emissions on a global scale.
However, on the other hand, there are more encouraging
actions in the other end, with respect to topic like
cleaner electricity production (especially solar and wind),
use of electric cars, energy conservation, reduced energy
demands in buildings (private, public and industrial),
public transportation, recycling of materials and resources
etc.
Transactions on Electrical Engineering, Vol. 1 (2012), No. 4
Similar movements are going on around the globe, and it
may be so well that the “down-top” actions will in the
end work better than “top-down” actions and regulations.
In these matters neither USA nor China should be
disregarded. Such activities are driven by public
demands. People do not like to live any more in houses
or flats with high energy demands for heating, for
instance. And they don’t want to drive cars in densely
populated cities. Therefore it is always also important to
be ahead of the development in order to ensure future
markets and demands.
As a conclusion of this paper some key points may be
noted:
122
“Climate Change Adaption Planning – an update for the
Power Industry. Special Focus: Lessons learned from
Extreme Weather & Natural Disasters” and ”New
Planning Practices Considering Renewable Resources
Integration and Distributed Energy Resources”. Papers
from these sessions may be available from IEEE – PES.
ACKNOWLEDGEMENTS
In writing this paper the author has relied heavily on
scientific and technical input from friends and colleagues
from recent and earlier times of collaboration. Here the
author wants to thank in particular (alphabetic order) R.
Benestad, The Norwegian Meteorological Institute
(met.no), S. Chereshnyuk, VNNIE, Russia, Á.J. Eliasson,
Landsnet, Iceland, H. Hawes, Australia, F. Popolanský,
EGÚ Brno, Czech Republic, professor H.M. Seip, Cicero,
Norway, E. Thorsteins, EFLA, Iceland, J. Toth,
Enginomix Consulting Inc., Canada and J.B. Wareing,
Brian Wareing.Tech. Inc, UK.
•
Evolution in the global climate must be
accounted for.
•
There are no significant indications in storm and
tornado frequencies yet. Probably this applies to
lightning as well.
•
Consider cheap actions before you are forced to
take on the expensive ones.
•
Consider life time of ohls in relation to time
[1]
scales for climate change.
•
Notify and file events in your grid.
As electric grids may become more vulnerable to adverse
weather, it should also be emphasized that the options of
detailed and on-purpose weather forecasts are developing
rapidly. The Cigré Session paper 2012 [8] demonstrates
some applications for such models. However, it is more
up to the utilities to ask for developments in such
technologies to improve the reliability in the operation of
their networks.
REFERENCES
Oreskes, N., Conway, E.M.: Merchants of Doubt. How a Handful
of Scientists Obscured the Truth on Issues from Tobacco Smoking
to Global Warming. New York, Berlin, London: Bloomsbury Press,
2010.
[2]
University of Colorado, http://sealevel.colorado.edu
[3]
Intergovernmental Panel on Climate Change (IPCC): Special
Report on Managing the Risks of Extreme Events and Disasters to
Advance Climate Change Adaption (SREX), Summary for
Policymakers. Cambridge University Press, Cambridge, UK and
New York, USA, 2012.
[4]
http://www.mzp.cz/en/climate_change_brochure
[5]
http://met.no/Klima/Fremtidsklima/Klima_om_100_ar/Hele_verden
/Ny+og+bedre+tilgang+til+klimascenarier.b7C_wlrSWW.ips (In
Norwegian)
Finally, it is important to note that potential challenges [6]
for electric production and transportation systems are
continuously dealt with by organizations and groups like
Cigré and IEEE, in addition to the countries mentioned
above. Cigré has established a new WG B2.54
“Management of Risk Associated with Severe Climatic [7]
Events and Climate Change on Overhead Lines”. This
was established in 2011 and will end in 2014. This WG is [8]
convened by Henry Hawes, Australia.
IEEE Power and Energy Society (IEEE – PES) has [9]
established the “IEEE Climate Change Technology SubCommittee” (CCTSC) to deal with such issues. More
information can be found on their web site:
https://collaborase.com/ieee-cctsc
Fikke, S.M., Ronsten, G., Heimo, A., Kunz, S., Ostrozlik, M.,
Persson, P.-E., Sabata, J., Wareing, B., Wichura, B., Chum., J.,
Laakso, T., Säntti, K., Makkonen, L.: COST 727: Atmospheric
Icing on Structures. Measurements and data collection on icing:
State of the Art. Veröffentlichung MeteoSchweiz Nr 75, Zürich,
2007.
CIGRÉ WG B2.16 TF03: Guidelines for meteorological icing
models, statistical methods and topographical effects. CIGRÉ
Technical Brochure 291, Paris, 2006.
Fikke, S.M, Nygaard, B.E.K., Horsman, D., Wareing, J.B., Tucker,
K.: Extreme weather studies by using modern meteorology. Session
paper B2-202, CIGRÉ 2012.
Fikke, S.M.: Tutorial on Climate Change. Cigré Study Committee
B2 meeting. Reykjavik, July 2011.
During the IEEE – PES General Meeting in San Diego
The contribution was presented on the conference ELEN 2012, PRAGUE, CZECH REPUBLIC
(July 2012) the CCTSC managed two sessions entitled:
Transactions on Electrical Engineering, Vol. 1 (2012), No. 4
123
Optimal Parameters of Load-Center Supply
System for Peripheral Districts of Big Cities
Svetlana Akchurina 1), Valeriya Tuzikova 2) and Josef Tlusty 3)
Czech Technical University in Prague, Faculty of Electrical Engineering, Department of Electrical Power Engineering
Technická 2, 166 27 Prague 6, Czech Republic, k315.feld.cvut.cz
1)
ak4ur@mail.ru, 2) tuzikval@fel.cvut.cz
Abstract — This paper is devoted to the analysis of
parameters of a city power supply system, which is
performed using a high voltage load-center supply system,
taking into account modern tendencies of development of
big cities. A distinguishing feature of modern society
development is a constant growth of a number of big cities
and their population, as well as development of power
supply for both utilities and industrial area. Such a
development of cities and utilities goes along with significant
growth of power consumption which requires new power
sources. It’s essential to foresee the construction of power
sources such as high voltage load-center supply systems
(LCSS). It was created a unique topology and technoeconomical model of the load-center supply system for
peripheral districts of big cities. Based on the formed
techno-economical model it was carried out a technoeconomical analysis of the load-center supply systems
feasibility and optimization of its parameters. Results of the
research and their analysis are shown in the presented
article. The load-center supply system is a compulsory
element of the power supply systems of big cities. In
addressing the optimal implementation of the power supply
system in peripheral areas of large cities it is required
examination of the topological characteristics of the districts
and the schemas of their power supply. The obtained results
allow us to make decisions for the construction of a loadcenter supply system for peripheral districts of the city, and
also to choose the optimum site for the construction of the
load-center substation regarding to the power source.
Keywords — Load-centre supply system, technoeconomical model, power supply systems, topological
characteristics.
I. INTRODUCTION
A distinguishing feature of modern society development
is the constant growth in the number of big cities and
their populations, as well the development of power
supply to both utilities and industrial area. Such
development of cities and utilities sees an accompanying
significant growth of power consumption. For example,
in Moscow from 2000 to 2010 the power consumption
increased from 38 to 53 bn. kWh, and according to
forecasts by the year of 2025 it can reach 96 bn. kWh. [1]
Such high consumption growth rates require new power
sources. It’s obvious that it’s a significant problem to
build new power plants within the city boundaries. It is
also not quite possible to transmit the required amount of
electric power from the remote sources to consumers
through 10-20 kV grid. That’s why it’s essential to
foresee the construction of power sources such as high
voltage load-center supply systems (LCSS). This fact
leads to the necessity of researching economically the
efficient construction of load-center supply systems as
well as defining and calculating their parameters.
In the precedent research examples [2-6 etc.] mainly
LCSSs involved into city centers were examined. But
under the modern conditions of expansion of city
territories the construction of LCSSs in new peripheral
districts located not far from the external power sources
becomes a reality.
II. TOPOLOGICAL CHARACTERISTICS.
In this case it is more reasonable to supply a section of
consumers directly from the external power source
through the medium voltage grid. That’s why a new
topology model of the power supply system was designed
for big city districts receiving power supply from both an
external power source and a radial load-center line
(Fig.1). The distribution of electricity from LCSS is
produced by magistral cable medium-voltage lines, as
each line feeds the same number of transformer
substations, which in turn have the same power and is
also equally spread throughout the district territory [2]. A
low voltage network is also equally spread.
One of the main issues of LCSS planning is the optimal
location of the load center substation (LCS). Therefore, in
addition to power of LCSSLCS the following parameters
as optimum parameters were considered:
Transactions on Electrical Engineering, Vol. 1 (2012), No. 4
124
b
LLCS
LPS
Power
source
LCL
a
LCS
DLMV
Fig. 1: Topology model of power supply system of peripheral district of a big city from an external power source and a radial
load-center system
LPS% – distance from the power source to the boundary of
a part supplied from LCS, expressed as % of district side
band, LLCS% – distance between LCS and the boundary of
a part supplied from LCS, expressed as % of the part
length.
LCS – load-center substation, LCL – load center line,
DLMV – medium voltage dual-circuit transmission line,
a,b – dimensions of a district, LPS – distance from power
source to the boundary of a part supplied from LCS,
LLCS– distance between LCS and the boundary of a part
supplied from LCS.
III. DEFINING THE OPTIMAL PARAMETERS OF LOADCENTER SUPPLY SYSTEMS.
While defining the optimal parameters of load-center
supply systems in this research, as an optimum criteria
was taken a minimum of discounted costs CdΣ for the
accounting period Ta:
Ta
С d ∑ = ∑ (I t + C t − Vt ) ⋅ (1 + E ) −t
(1)
t =1
It, Ct – investments for the construction of the object and
the total cost of its operation in year t,
Vt – residual value of the object at the end of the
accounting period (t = Ta),
E – discounting norm.
The following techno-economical models were
considered:
•
Discounted costs of LCS - CPS;
•
Discounted costs on the line LCL - CLCL and
substation LCS- CLCS;
•
Discounted costs of an external source - CmvPC and
discounted costs on medium voltage lines - CmvLCS:
CdΣ= CPS + CLCL + CLCS + CmvPC + CmvLCS
(2)
As a result we obtained the expression for the discounted
costs depending on the optimized parameters. This
expression was the criterion function during LCSS
parameters optimization. Alpha
Implementation of the power supply system of the city
with the application of LCSS fundamentally changes the
structure of the urban power grid and increases the cost of
its construction and operation. Therefore it was carried
out a techno-economical research of feasibility of LCSS
construction in the city, using the resulting model of
discounted costs. It assessed the optimization of the
parameters of LCSS for districts with a surface density
load σ = 5 MVA/km2 and more, ranging from 3 to 16 km2
(district square area in Moscow [8]) with different aspect
ratio of the district, while defining the boundary values of
σ and geometrical parameters for which the real solution
is possible.
By the results of calculations we constructed nomograms
of feasibility of LCSS for peripheral districts of cities,
which are presented in (Fig.2.) According to the obtained
nomograms we can define a particular peripheral district
with the corresponding geometric parameters, and
Transactions on Electrical Engineering, Vol. 1 (2012), No. 4
125
a) rated voltage of LCSS 110 kV
and
b) 220 kV
Fig. 2: Nomograms of economic feasibility of LCSS
surface density load, for which it is necessary to make a
construction of LCSS.
Analyzing the diagram it was concluded:
1) Construction feasibility of LCSS in the peripheral
area of the city depends on its geometrical
parameters: the values and the ratio of its length b
and width a;
2) LCSS is inappropriate to construct at depth (length)
of district b (deep into the city) less than 1.8 km;
3) With the increase of the surface density load, the
district area is reduced, for which is advisable to
make power supply system with application of
LCSS.
LPS % ,%
Based on the formed techno-economical model the
optimization of the LCSS parameters has resulted.
According to the results of the optimization there was
constructed dependence of optimized parameters SLCS,
LPS%, LLCS% on the surface density load σ, the geometric
parameters of the district a, b, and their ratio a/b for a
district square area ranging from 3 to 16 km2.
Fig. 3 presents as an example the results of optimal
parameters of 110-220 kV LCSS for peripheral district
having an area of 10 km2.
By analyzing these relationships the following results
were obtained:
1)
Optimal power values of LCS for peripheral
district of the city is 80 MVA and more, for rated voltage
S LCS ,MVA
LLCS % ,%
σ,
MVA
km 2
σ,
MVA
km 2
a) rated voltage of LCSS 110 kV
LLCS % ,%
LPS % ,%
σ,
−
a : b = 3 :1;
MVA
km 2
− a : b = 2 :1;
σ,
σ,
MVA
km 2
S LCS ,MVA
MVA
km 2
b) rated voltage of LCSS 220 kV
− a : b = 1 : 2;
− a : b = 1 :1;
σ,
− a : b = 1: 3
Fig.3: Results of optimal parameters of 110-220 kV LCSS for the peripheral district in an area of 10 km2
MVA
km 2
Transactions on Electrical Engineering, Vol. 1 (2012), No. 4
of LCSS 110 kV and 126 MVA and more, for voltage
220 kV depending on the value of the surface density
load. The maximum power value of LCSs that are
constructed in large cities of Russia is 400 MVA. Due to
the results that we obtained, SLCS reaches values of 600700 MW. Consequently, for the peripheral districts of big
cities it is appropriate to include the magistral LCSS. This
statement makes an interest for the further research.
2)
For the districts of 3 to 16 km2 and load density
20 to 50 MVA/km2 it is reasonable to locate LCS at a
distance of 20-45 % of district side b;
This fact is explained by the interaction of two
competitive effects. When the distance between PS and
LCS grows the part of the district supplied directly from
PS increases as well, and the part supplied from LCS
decreases. As a consequence, on the one hand, LCS
construction, operation and losses coverage costs
decrease and also the costs of medium voltage grid of
LCS decrease as well. On the other hand construction
costs of medium voltage grid supplied directly from the
power source grow. Length of the load-center line grows,
and, correspondingly, its costs grow as well.
3)
The optimal location of LCS from the supplied
part is 5 to 40% of part length;
The increasing importance LLCS% with increasing of the
surface density load and the size of the district area is
because of the increasing costs of construction, operation
and compensation for energy losses in medium-voltage
network. The total length of medium voltage lines has a
minimum value in the location of LCS in the center of the
loads feeding by the part of this district. But it must be
kept in mind that the closer the location of LCS to the
center of the loads, the greater the length of the load
center line.
126
In addressing the optimal implementation of the power
supply system in peripheral areas of large cities it is
required examination of the topological characteristics of
the districts and schemas of their power supply. The
obtained results allow us to make decisions for the
construction of a load-center supply system for peripheral
districts of the city, and also to choose the optimum site
for the construction of load center substation regarding to
the power source.
ACKNOWLEDGEMENT
Financial support of the Ministry of Education, Youth
and Sports, through grant number MSM 6840770017, is
highly acknowledged.
REFERENCES
[1]
The Moscow Government Decree №1075-ПП, «The energy
strategy of Moscow up to 2025».
[2]
Glazunov A.A, Kuznetsova T.A, Fedoseev A.A.; Cost-effective
voltage and the power of the deep-lead in the cities / /
Electricity.1983. №2.
[3]
Glazunov A.A., Utkina E.G.; Analysis of the optimal capacity
of 110-220 kV load-center supply system of big cities / /
Vestnik MEI. 1998. №5.
[4]
Glazunov A.A., Leschinskaya T.B., Shvedov G.V. Multicriteria
optimization of load-center supply system of cities, taking into
account the uncertainty of the electrical loads. Agrokonsalt,
2005
[5]
Development of proposals for the optimal structure of the
nominal voltage and power substation networks of 110 kV and
above of power supply system of Moscow and Moscow region
up to 2010 / / Report on the research. MEI.1987.
[6]
Kozlov V.A. Power supply systems of
cities. Energoatomizdat.1988.
[7]
Reference for design of electrical networks / under. Ed. D.L
Faibisovich. ENAS, 2012.
[8]
The territorial office of the Federal State Statistics Service in
Moscow // Indicators of municipalities (2010) / / Section
6. Territory. The total land area of the municipality.
IV. CONCLUSION
The load-center supply system is a compulsory element
of the power supply systems of big cities. Consequently,
we need to research their optimal parameters.
The contribution was presented on the conference ELEN 2012, PRAGUE, CZECH REPUBLIC
Transactions on Electrical Engineering, Vol. 1 (2012), No. 4
127
The Possibility of Electronic Equipment Cooling
Roman Janiš 1), Jan Kuba 2)
Czech Technical University in Prague, Faculty of Electrical Engineering, Department of Electrotechnology
Technická 2, 166 27 Prague 6, Czech Republic, technology.feld.cvut.cz
1)
janisrom@fel.cvut.cz
2)
kuba@fel.cvut.cz
Abstract — Cooling has a large and irreplaceable
importance in electronics and its solution is a part of the
technology design and construction of any real device. Due
to the progressive miniaturization and increasing
performance we can meet cooling more often in consumer
electronic products. In addition to many well-known
conventional methods of cooling there is a possibility to use
a high-capacity heat transport device - heat pipe. A
disadvantage in some applications is the restriction of the
possibility to control the thermal conductivity. The authors
have dealt with so far little explored possibility of control
using an external magnetic field with magnetic fluid as the
heat pipe workload. The system is reversible in terms of
control and uses the unique features of magnetic fluids.
other. The heat pipe is in principle a closed two-phase
system in which heat transfer is realized by continuous
and rapid circulation of steam and liquid phases and
phase transformations of each other – evaporation and
condensation of working fluid. One end of the tube is
usually thermally and mechanically connected to the heat
source and the other end ensures removing of the heat
transferred by tube. Liquid evaporates in the evaporation
section and in form of steam flows into the condenser
section, where in contact with the inner surface of the
tube cooler condenses to a liquid phase. The liquid
condensate then runs down or flows on the walls tube
back into the evaporation section.
Keywords — Cooling, heat pipe, magnetic fluids, thermal
conductivity.
I.
II. ONE OPTION IS HEAT PIPE
Heat pipe elements are intended for highly efficient
transport of heat, a intense development was stimulated
in the middle of last century, particularly in connection
with the space programs in the U.S. where they had to be
dealt with problems of heat transport in the specific space
conditions (high vacuum, weightlessness). Currently,
INTRODUCTION
Heat transport (in terms of physical nature) is done
by three basic mechanisms – conduction, convection and
radiation, which are in most cases combined with each
0-15 min.: B=0T
16-18 min.: B=0,5T
18.5-22 min.: B=0T
22.5-25.5 min.: B=0.5T
te m p e ra tu re
d iffe re n c e
b e tw e e n th e e n d s
o f th e tu b e (o C )
25
20
19,9
18,2
15
14,3
13,7
10
9,8
5,1 4,7
5
4,7
2,8 2,5
1,7
0
0
1
8,5
7,8
6,8
2
3
5
7
10
1,6
4,3
1,6 1,5
12
15
4,2
1,5 0,7 0,6
16 16,5 17 17,5 18 18,5 19 19,5 20
time (min)
Fig. 1: Temperature difference between the ends of the tube
0,4 0,2
21
1,8
22 22,5 23 23,5 24 24,5
Transactions on Electrical Engineering, Vol. 1 (2012), No. 4
heat pipes have a common alternative to conventional
methods of the heat transfer in many devices used in
various areas of life. Outwardly, the heat pipe behaves as
a solid body made of a material of extremely high
thermal conductivity λ (of the order to 104 Wm-1K-1),
significantly higher than for example copper (380 Wm1 -1
K ) and is able to transfer heat so intensive even at
relatively low temperature gradient. Due to its simple
construction, reliable operation, the availability of a
suitable type and price, heat pipes are used now in a wide
area of applications, such as elements of power
electronics cooling, heat recovery, ensuring of the
thermal dynamics of selected technological and
biological processes, cooling, solar panels and many
others.
III. THE DESIGN AND TYPES OF HEAT PIPES
In heat pipe called as gravitational the condensate is
transported by earth gravity. The working substance in
the liquid phase flows down the walls of the tube in form
of a thin film to evaporator section. Precondition of
functionality for this type of the tube is a configuration
with a warm end placed below the cool end of the tube.
The tubes can be oriented vertically or at an angle, the
performance of the heat pipe in this case is influenced by
the slope and specific weight of the condensate.
In case of capillary tubes their function is not
significantly depending on the orientation in a
gravitational field, since the condensate is transported
mainly by capillary pressure, resulting in contact of the
liquid phase and a suitably chosen capillary system. In
this case, along the inner wall of the capillary tubes the
system is located, usually consisting of a groove system,
sieves and fibers.
To increase the heat output of tubes it is sometimes
used a combination of several types of capillaries, such as
grooves in the tube wall and several layers of sieves.
Although the capillary tubes compared with
the
gravitational ones are more complicated, there are today
the most widespread in many applications mainly due to
ensuring functionality in all positions.
IV. EXPERIMENTAL HEAT PIPE WITH CONTROLLED
HEAT TRANSPORT BY MAGNETIC FIELD
A. Description and characteristic
This is an experimental gravity type heat pipe which,
can operate in a uncontrolled mode - that is the maximum
heat
transport
(maximum
equivalent
thermal
conductivity) or under a control with external static
magnetic field - that is the limited transport of heat (lower
thermal conductivity equivalent). The working substance
in this case is the magnetic fluid in form of dispersions
composed of pure water and specially prepared Fe2O3
nanoparticles.
The present sample uses unique properties and
behavior of magnetic fluid in a magnetic field for
128
effective influence
of the continuous opposite flow
intensity of steam and condensate between the ends of the
tube where the heat is conveyed (evaporator) and where
heat is removed outside.
Application of an external magnetic field of
sufficient level (induction B) in properly selected position
between the evaporator and condenser of vertically
placed tube caused an imbalance between the amount of
evaporated and condensed returning substances. It causes
a disruption of the continuity of the internal flow, which
is externally indicated by a temperature difference
between evaporator and condenser - temperature will
increase (decrease the equivalent thermal conductivity of
the tube).
Based on this principle it is possible to control, using
an external magnetic field, temperature between the
evaporator and condenser or almost interrupt thermal
binding between one part and the second part of the tube
(evaporator - condenser) - it looks as the "thermal key".
The situation is well illustrated in Fig. 1, where the
difference between evaporator and condenser heat pipes
on time is drawn in the dependence of the measured
temperature.
It is illustrated gradually for 4 situations where the
external magnetic field was not applied along the tube (B
= 0 T) and when it was applied (B = 0.5 T).
Measurements were performed at ambient temperature +
22.8 °C, with attached external thermal insulation and
with natural cooling by surrounding air.
B. The arrangement and main technical parameters
•
Case is made of clear glass tube. On the bottom side
is the evaporator with an external resistive heating,
on the upper side is the condenser with an external
cooler. Temperature is measured by a thermocouple
type K.
•
Tube length is 350mm, tube diameter is 9mm, glass
thickness is 1mm.
•
Induction of a permanent magnet is B = 0.5 T (two
magnets NeFeB against each other, dimensions
40x20x10 mm, gap width 12 mm).
•
Pipe heating power is 10W.
•
Operating temperature range is about +20 ° C to
+70 º C.
•
Working substance is water + magnetic fluid
Ferrotec with the proportion 1:1, the total quantity of
the working fluid is 2.5 ml.
•
Operating pressure in the tube is about 2.5 kPa.
The actual appearance of the equipment is shown in
Fig. 2. A detail of the tube inserted in the magnetic field
during heating is shown in Fig. 3. Circulation of the
workload is interrupted at this location.
Transactions on Electrical Engineering, Vol. 1 (2012), No. 4
129
V. CONCLUSION
In this paper the authors describe realized functional
sample of the thermal heat pipe which is an original
contribution to the development of new types of heat
pipes with improved properties and to other magnetic
fluid usability in practical applications. There were
verified the functionality of the sample in experiments
but there were not solved purely operational matters such
as reliability and durability. Using magnetic fluid as the
working fluid in heat pipes with controlled heat transport
is minimally explored area, but due to the possibilities of
a widespread use in practical applications, a very
promising area.
ACKNOWLEDGEMENT
This paper is based on the research program for
students No. SGS12/ 065/ OHK3/ 1T/13 “Magnetic fluids
and possibilities of using” of CTU in Prague.
REFERENCES
[1]
[2]
Fig. 2: Actual appearance of the equipment
[3]
[4]
[5]
Cingroš, F., Hron, T., Kuba, J.: Magnetic Field Control of
Cryogenic Heat Pipes. In Proceeding of the International
Conference ISSE 2009. Brno (Czech Republic), 2009.
Ueno, S., Iwaki, S., Tazume, K.: Control of Heat Transport in
Heat Pipes by Magnetic Field. J. Appl. Phys., Vol. 69 (1991), No.
8, p. 4925 – 4927.
Cingroš, F., Hron, T.: Working Fluid Quantity Effect on Magnetic
Field Control of Heat Pipes. In Acta Polytechnica, Vol. 49, No. 23/2009, ISSN 1210-2709, p. 44-47.
Cingroš, F., Hron, T., Kuba, J.: Investigation and Diagnostic of
Magnetic Control of Cryogenic Heat Pipes. In Electroscope, No.
5/2011, ISSN 1802-4564, ZCU Plzen.
Kuba, J., Cingroš, F: Experimental Heat Pipe Controlled Heat
Transport by Magnetic Field, Registered Functional Sample No.
13113/5/2011, CTU in Prague, Faculty of Electrical Engineering,
Department of Electrotechnology.
Fig. 3: Detail of the tube in the magnetic field
The contribution was presented on the conference ELEN 2012, PRAGUE, CZECH REPUBLIC
Transactions on Electrical Engineering, Vol. 1 (2012), No. 4
130
Model Based Design of Electric drives
Jakub Vonkomer 1), Milan Žalman 2)
1)
Institute of Control and Industrial Informatics, FEI STU, Bratislava, Slovakia, e-mail: jakub.vonkomer@stuba.sk
Institute of Control and Industrial Informatics, FEI STU, Bratislava, Slovakia, e-mail: milan.zalman@stuba.sk
2)
Abstract— Electric drives are probably the most widely
used electronic devices in the industry. The development
process of the drives is very time-intensive, but there are
many methods to reduce the development time. In this
paper, the development of vector control of induction
machine using Simulink is presented. Firstly, the creation of
the simulation model is described; then the same model is
used for code generation. We used the dSpace platform;
however, any platform supported by Matlab/Simulink can
be used as well. Finally, the comparison of results between
simulation and real motor is presented.
Keywords— AC drive, vector control,
simulation, Simulink, Code generation, dSpace.
iˆS
φ
Ψ̂r
is2
Fig. 1. Vector diagram of stator current and rotor magnetic flux
2
β
iˆS
development
1
Ψ̂r
υs
I. INTRODUCTION
is2
The development process of electric drives is itself
usually very time-intensive. Simulations can help, but it
also takes a lot of time to transform the model to the final
code. Manual code writing of the control algorithms
introduces a lot of bugs and errors. Some of the bugs even
require many hours to be fixed. However, there is no need
to do it in this way nowadays, because of the MATLAB
code generation possibilities. In our case we used the Real
Time Workshop for generating code for dSpace. The
generated C code is then compiled and downloaded to the
dSpace processor and can be immediately executed.
However, there are many platforms supported,
manufacturers like Analog Devices®, Atmel®, Freescale®,
Intel® or Texas Instruments® are no exception.
This process of the design and development is called
“Model Based Design” and is widely propagated by
Mathworks Inc., the creators of Matlab.
II. VECTOR CONTROL
Vector control is a well-known induction motor control
method. It allows high-performance control of
electromechanical variables like torque, speed, position in
both steady and transient states. Main principle of the
vector control is the decoupled, independent control of
magnetic flux and torque. There are plenty of methods of
vector control, but we will focus on the rotor-flux-oriented
vector control. The control law can be derived from:
ˆ × iˆ ,where k = 3 p′ Lm
M m = km Ψ
r
s
m
2 Lr
is1
(1)
is1
α
Fig. 2. Vector diagram of the IM state variables
Fig. 3. Simplified flow chart of the vector control
All the controllers and estimators are derived from the
mathematical model of the induction machine [1]:
dψˆ s
(2)
+ jωkψˆ s
uˆ s = Rs iˆs +
dt
d ψˆ r
(3)
0 = Rr iˆr +
+ j (ωk − ω )ψˆ r
dt
3 L
(4)
M m = p′ m ℑ{ iˆs ⋅ψ r*}
2 Lr
(5)
ω = p ' ωm
d ωm
(6)
dt
The structure of all controllers is PI and they are designed
by the pole-placement method. All systems are derived
from the linearized induction machine model and are
described by first-order dynamic equations.
Mm − Mz = J
1) Current controllers: Design of current controllers is
derived from the following transfer function:
i (s ) i S 2 (s ) 1 / R1
(7)
G S (s ) = S 1
=
=
u S 1 (s ) u S 2 (s ) T1 s + 1
Final current controller parameters are achieved using the
following equations ( ωo is desired dynamics, b desired
damping):
Kp

1
(8)
Ti =
K p =  2b1ω01 −  R1 ⋅ T1
2
ω
R
T
T
1 1 01

1 
Transactions on Electrical Engineering, Vol. 1 (2012), No. 4
2) Flux controller:
Design of flux controller is derived from the following
transfer function:
Lm
Ψ (s )
(9)
G S (s ) = r
=
i S 1 (s ) Tr s + 1
Final flux controller parameters:

1
K p =  2b2ω02 −
T

r
 Tr

 Lm
Ti =
K p Lm
131
information and error flags. This block will be later copied
to the dSpace code generation model only. Figure 4 shows
the main window of the simulation SimPowerSystems
model.
(10)
Trω02 2
3) Speed controller:
Speed controller is designed from the knowledge of the
mechanical parameters. Transfer function:
L
ω (s ) 3
1
(11)
G S (s ) =
= ⋅ p'⋅ m ψ r ⋅
i S 1 (s ) 2
Lr
Js + B
Final equations:
L
3
K p ⋅ ⋅ p '⋅ m ψ r
Lr
2
( 2b3ω03 J − B ) Ti =
(12)
Kp =
2
J ⋅ ω03
Lm
3
⋅ p '⋅ ψ r
2
Lr
3) Flux weakening controller:
Flux weakening is necessary for achieving speeds
higher than rated motor speed . However, the design of the
flux weakening controller is out of scope of this paper. In
general it is a voltage controller which is activated when
the voltage is close to voltage saturation. This controller
decreases the flux to reduce the back-emf, allowing highspeed motor operation. Instead of the voltage controller,
feed-forward methods can be used as well, but they
usually provide lower robustness.
4) Flux estimator:
Flux estimator is the heart of the vector control, because
of the control algorithm oriented into rotor magnetic flux
reference frame. That is one of the reasons the vector
control is often called “Field Oriented Control”.
Following equations describe the presented flux estimator
(indirect vector control[2][3]).
1
L
(13)
sψˆ r = − ψˆ r + jωψˆ r + m iˆs
Tr
Tr
ψ% r =
Lm
is1
Tr s + 1
(14)
ω% sl =
Lm
is 2
Trψ% r
(15)
III. BUILDING THE SIMULATION MODEL
The aim of the project was to develop the speed vector
control of induction machines with the speed feedback
from the speed sensor (incremental encoder in our case).
Our primary platform for developing and testing
purposes was Simulink with SimPowerSystems toolbox[5].
We used the model of induction machine, three phase
power supply and many other power electronic blocks for
close-to-reality simulation of the drive [6].
The basic block is the “VECTOR CONTROL” block,
which contains all the controllers, observers, state
Fig. 4. Screen capture of the simulation model
Figures 5, 6, and 7 show the inner connections of
selected blocks. The model is purely discrete with sample
time of 200 µs. All the blocks are part of Embedded
MATLAB subset[7], so there should be no problem with
the code generation. Sample time for simulating the
electronics has been chosen as 10 µs, but for a high
quality simulation of the hardware as well, lower values
are better.
Constant blocks for parameters instead of gain blocks
are preferred. Variables inside the constant block may be
configured as exported variables as part of global structure
or as global variables as well. These variables may be read
or written by the user code. In this way, it is possible to
change the behavior of the model on the fly.
The entire model is based on IEEE 754 single precision
floating point numbers (32 bit), to make this model faster
in execution and to allow easy porting to any target with
floating point numbers support. Usually, for an application
like control of power electronics; the precision of 32 bit
floating numbers is pretty enough.
For connecting the signals executed in different
subsystems or sampled in different time, Rate transition
blocks may be used. However, this often involves
additional operations for the Simulink to ensure proper
sampling of the variables; therefore global variables (Data
Store Memory blocks) are recommended [7].
Fig. 5. Screen capture of the VECTOR CONTROL Block
Fig. 6. Screen capture of the inverter block
Transactions on Electrical Engineering, Vol. 1 (2012), No. 4
132
However, the behavior of data integrity can be detected;
it depends on the configuration of the model.
Fig. 7. Screen capture of the current controller block
IV. USING THE MODEL FOR CODE GENERATION
The already tested simulation model can be easily used
for code generation purposes. The main block from
simulation model is copied to the code generation model.
Nevertheless, the main task for the code generation model
is communication with the target platform hardware. This
task requires special blocks, unique for every platform;
therefore it is required to look to the specific
documentation. As result, instead of complicated timeintensive rewriting model to the C code, we developed
the program in Simulink, added the platform interface,
then built and executed the finished model.
We use the laboratory dSpace 1104[4] for testing the
model. DSpace board allows real-time control of different
systems. Its main advantage is integration into the
MATLAB/Simulink environment. Model is executed and
controlled via the ControlDesk application. It is also very
easy to export any graphs as MATLAB´s MAT files.
As mentioned earlier, constant blocks can be easily
tunable, they might be configured as inputs in the
ControlDesk applications and its value can be changed on
the fly. The same is applied for the code generation for the
other platforms. For other platforms, designer of the
model can decide whether the variable will be tunable
from the user code by selecting the variable from
Simulation->Configuration Parameters, Optimization,
Configure.
Fig. 9. Screen capture of the proposed dSpace model
Models for both simulation and code generation are
pretty similar as can be seen in figure 9. Instead of
induction machine and inverter blocks, the block of
communication with an inverter is used. Inputs represent
reference voltages. The communication block with
inverter contains the compensation of DC voltage
fluctuations and the computed duty cycles are inputs for
the dSpace PWM block.
Figure 10 shows the basic schema of the dSpace
controller board to the frequency converter connection.
Fig. 10. Schema of connecting dSpace equipped PC with frequency
inverter and induction motor
V. COMPARISON
Figures 11-14 show the comparison of simulation and
real behavior of the modeled system. Wide range of
speed is demonstrated. Small differences and noise in
dSpace figures are caused by non-calibrated current
sensors, but indeed we can say that we reached
comparative results within the toleration.
Speed: dSpace - w* = 0-20-40-60-80-100 rad/s
120
measured
reference
100
Fig. 8. Screen shot of variable configuration in Simulink
angular speed [rad/s]
80
By using global variables defined by Data Store
Memory blocks, several advantages are gained. These
variables can be easily read (and written) in the custom
code; they always have the same name as defined in
Simulink. However, in contrast to signals transferred over
Rate transition blocks, no data integrity is guaranteed so
data might be for example written three times and read
just once, so some information can be easily lost.
60
40
20
0
-20
0
0.5
1
1.5
2
time [s]
2.5
3
3.5
4
Fig. 11. dSpace experiment: steps of the reference speed up to
100 rad/s
Transactions on Electrical Engineering, Vol. 1 (2012), No. 4
133
VII. APPENDIX 1
DESCRIPTION OF SYMBOLS AND ACRONYMS USED IN THIS
Speed: Simulation - w* = 0-20-40-60-80-100 rad/s
120
measured
reference
100
PAPER
angular speed [rad/s]
80
Symbol
60
40
20
0
-20
0
0.5
1
1.5
2
time [s]
2.5
3
3.5
4
Fig. 12. Simulation: steps of the reference speed up to 100 rad/s
Speed: dSpace - w* = 10 rad/s to -10 rad/s
15
measured
reference
angular s peed [rad/s ]
10
5
ωm
ωsl
p’
σLs
Ls
Lr
Lm
Rs
Rr
R1
T1
Tr
J
Mm
Mz
iˆs
û s
0
ψ̂ r
Description
motor angular speed
slip speed
number of pole pairs
stator leakage inductance
stator winding inductance
rotor winding inductance
mutual inductance
stator resistance
rotor resistance
combined resistance (R1=Rs+Lm2/Lr2*Rr)
current model time constant (T1=σLs/R1)
rotor time constant (Tr=Lr/Rr)
moment of inertia
motor torque
load torque
stator current vector
stator voltage vector
rotor flux vector
-5
-10
-15
0
0.5
1
1.5
2
time [s]
2.5
3
3.5
4
Fig. 13. dSpace experiment: performance of low speed and speed
reversal
ACKNOWLEDGMENT
This work was supported by the Slovak Research and
Development Agency under the contract No. VMSP-II0015-09.
REFERENCES
Speed: Simulation - w* = 10 rad/s to -10 rad/s
15
measured
reference
angular speed [rad/s]
10
5
0
-5
-10
-15
0
0.5
1
1.5
2
time [s]
2.5
3
3.5
4
Fig. 14. Simulation: performance of low speed and speed reversal
VI. CONCLUSION
In this paper we described several aspects regarding the
induction machine vector control development. Since the
beginning we tried to keep the model simple, robust and
fully discrete, because we planned to share the basic
model with other platform for code generation – dSpace in
our case. The basic testing during the creation and
development of the model was done by simulation only
on a dSpace platform where real capabilities were
demonstrated.
This development process called “Model based design”
can be used for any control structure with many
advantages. Creating the model and testing in Simulink
greatly saves time and other resources. This leads to
reduced development time, reduced number of bugs and
increased user-friendliness of the whole development
process. This paper is an updated version of the paper
published in 2009[8].
[1] Žalman M.: “Akčné členy (Slovak language)”, STU Bratislava
2003
[2] Doki, S.; Kinpara, Y.; Okuma, S.; Sangwongwanich, S., "Unified
interpretation of indirect and direct vector control [of electric
machines]," Power Conversion Conference, 1993. Yokohama
1993., Conference Record of the , vol., no., pp.297-302, 19-21 Apr
1993,URL:
http://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=264167&isnu
mber=6639
[3] Marwali, M.N.; Keyhani, A.; Tjanaka, W., "Implementation of
indirect vector control on an integrated digital signal processorbased system," Energy conversion, ieee transactions on , vol.14,
no.2, pp.139-146, Jun 1999 URL:
http://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=766961&isnu
mber=16622
[4] DS1104 R&D Controller Board, dSpace Inc. 2009 Catalogue,
URL:http://www.dspaceinc.com/ww/en/inc/home/medien/papers/d
ownload_page.cfm?FileID=778
[5] The MathWorks, SimPowerSystems Documentation, 2011
URL:
http://www.mathworks.com/access/helpdesk/help/toolbox/physmod
/powersys/index.html?/access/helpdesk/help/toolbox/physmod/pow
ersys/
[6] The MathWorks, SimPowerSystems Demos
[7] The MathWorks, Code Generation from MATLAB®, User’s
Guide, R2012b URL:
http://www.mathworks.com/help/pdf_doc/eml/eml_ug.pdf
[8] Vonkomer, J.,Radičová T., Žalman M., Suchánek M. “Fast AC
Electric Drive Development Process Using Simulink Code
Generation Possibilities”, Technical Computing Prague 2009 : 17th
Annual Conference Proceedings. Prague, Czech Republic,
19.11.2009. - Prague: Humusoft, Ltd., 2009. – ISBN:978-80-7080733-0. - CD-Rom
_____________________________________________________________________________________________
TRANSACTIONS ON ELECTRICAL ENGINEERING VOL. 1, NO. 4 HAS BEEN PUBLISHED ON 21ST OF DECEMBER 2012