Appendix 13: PSS/E ANALYSES AND RESULTS

advertisement
Appendix 13:
PSS/E ANALYSES AND RESULTS
Prepared By:
SECI Interconnection Study Task Group (ISTG)
March, 2005
GENERATION INVESTMENT STUDY
Transmission network checking
Draft Report
March, 2005
carried out by
ELECTRICITY COORDINATING CENTER, Ltd
11040 BELGRADE - VOJVODE STEPE 412, SERBIA AND MONTENEGRO
and
ENERGY INSTITUTE ”HRVOJE POŽAR“ - ZAGREB
ZAGREB – SAVSKA CESTA 163, CROATIA
Authors:
Project leaders:
Mr. Miroslav Vuković – EKC, Belgrade
Mr. Davor Bajs – Energy Institute Hrvoje Požar, Zagreb
Project coordinators:
Mr. Trajce Čerepnalkovski – ESM, Skopje
Mr. Kliment Naumoski – ESM, Skopje
Ms. Marija Stefkova - Secretary
Participants on project:
Mr. Predrag Mikša – EKC, Belgrade
Mr. Dobrijević Djordje – EKC, Belgrade
Mr. Nijaz Dizdarevic – Energy Institute Hrvoje Požar, Zagreb
Mr. Goran Majstrovic – Energy Institute Hrvoje Požar, Zagreb
miro@ekc-ltd.com
dbajs@eihp.hr
ii
____________________________________________________________________________
Complete GIS Report
Table of contents
Volume 1 – Executive Summary
1 EXECUTIVE SUMMARY
7
Volume 2 – Main Report - Electricity Demand Forecast
2 ELECTRICITY DEMAND FORECAST
29
29
2.1
Objectives
2.2
What Are We Forecasting?
31
33
2.3
Background
40
2.4
Approach
2.5
Assumptions And Data Sources
44
48
2.6
Forecasting Model
51
2.7
Results And Validation
Volume 3 - Main Report - Generation & Transmission Study
3 GENERATION AND TRANSMISSION STUDY
75
75
3.1
Introduction
3.2
Computer Models
76
84
3.3
WASP And GTMax Runs
87
3.4
Candidate Plant
3.5
Fuel Costs
94
108
3.6
Fuel Costs – Utility Data
111
3.7
Fuel Costs – Reconciliation, Forecast Study Prices
3.8
Base Case Assumptions
116
121
3.9
Scenario A, B and C Results
163
3.10 Conclusions
3.11 Recommendations
172
174
4 REFERENCES
Volume 4 – Electricity Demand Forecast Appendices
Appendix 1: Review of econometric studies into the relationship between GDP per capita and
electricity demand
Appendix 2: Details of the econometric analysis of the relationship between GDP per capita
and net electricity consumption
Appendix 3: Basis for long term GDP per capita growth forecasts
Appendix 4: Load shape adjustments
Appendix 5: Findings from ECA methodology review
Appendix 6: Country electricity demand forecasts
Volume 5 – Generation & Transmission Study Appendices
Appendix 7: Country data profiles
Appendix 8: Specific candidate plant and rehabilitation
Appendix 9: Screening curve analysis and cost of rehabilitation
Appendix 10: Hydro sensitivity analysis
Appendix 11: GTMax Analyses and Results
Appendix 12: Scenario A, B & C results
Volume 6 – PSSE Appendix
Appendix 13: PSSE Appendix
____________________________________________________________________________________
iii
CONTENTS:
ABBREVIATIONS
vii
1. INTRODUCTION
1.1
2. STARTING CONDITIONS
2.1
2.1. Description of PSS/E Model
2.2
2.2. Prerequisites and Assumptions
2.4
2.3. Technical Specification of New Transmission Lines Candidates and Substations
Costs of the New Interconnection Overhead Lines
Costs of New Substations in 2010 and 2015
2.7
2.10
2.12
2.4. Remarks on PSS/E Transmission System Model and GTMax Model Harmonization
Load Demand
Network Topology
Production distribution
Exchange programs-desired interchange
2.14
2.14
2.14
2.19
2.38
3. ANALYSED GENERATION, DEMAND AND EXCHANGE SCENARIOS
3.1
3.1. Year 2010 Scenarios
3.2
3.2. Year 2015 Scenarios
3.6
4. LOAD FLOW AND CONTINGENCY ANALYSIS – REFERENCE CASES
4.1
Introduction
4.2
4.1 Scenario 2010 – average hydrology – topology 2010
4.1.1 Lines Loadings
4.1.2. Voltage Profile in the Region
4.1.3. Security (n-1) Analysis
4.2
4.2
4.7
4.8
4.2 Scenario 2010 – dry hydrology – topology 2010
4.2.1. Lines Loadings
4.2.2. Voltage Profile in the Region
4.2.3. Security (n-1) Analysis
4.11
4.11
4.14
4.14
4.3 Scenario 2010 – wet hydrology – topology 2010
4.3.1. Lines Loadings
4.3.2. Voltage Profile in the Region
4.3.3. Security (n-1) Analysis
4.16
4.16
4.20
4.21
4.4 Scenario 2015 - average hydrology – topology 2010
4.4.1. Lines Loadings
4.4.2. Voltage Profile in the Region
4.4.3. Security (n-1) Analysis
4.24
4.24
4.27
4.28
4.5 Scenario 2015 - average hydrology – topology 2015
4.5.1. Lines Loadings
4.5.2. Voltage Profile in the Region
4.5.3. Security (n-1) Analysis
4.5.4. Summary of Impacts - 2015 topology versus 2010 topology
4.31
4.31
4.34
4.35
4.36
iv
4.6 Scenario 2015 - dry hydrology – topology 2010
4.6.1. Lines Loadings
4.6.2. Voltage Profile in the Region
4.6.3. Security (n-1) Analysis
4.38
4.38
4.41
4.42
4.7 Scenario 2015 - dry hydrology – topology 2015
4.7.1. Lines Loadings
4.7.2. Voltage Profile in the Region
4.7.3. Security (n-1) Analysis
4.7.4. Summary of Impacts - 2015 topology versus 2010 topology
4.44
4.44
4.47
4.48
4.50
4.8 Scenario 2015 - wet hydrology – topology 2010
4.8.1. Lines Loadings
4.8.2. Voltage Profile in the Region
4.8.3. Security (n-1) Analysis
4.51
4.51
4.56
4.57
4.9 Scenario 2015 - wet hydrology – topology 2015
4.9.1. Lines Loadings
4.9.2. Voltage Profile in the Region
4.9.3. Security (n-1) Analysis
4.9.4. Summary of Impacts - 2015 topology versus 2010 topology
4.61
4.61
4.66
4.67
4.69
5. LOAD FLOW AND CONTINGENCY ANALYSIS – SENSITIVITY CASES
5.1
Introduction
5.2
5.1 Scenario 2010 – average hydrology import/export – topology 2010
5.1.1 Lines Loadings
5.1.2. Voltage Profile in the Region
5.1.3. Security (n-1) Analysis
5.2
5.2
5.8
5.9
5.2 Scenario 2010 – average hydrology high load – topology 2010
5.2.1. Lines Loadings
5.2.2. Voltage Profile in the Region
5.2.3. Security (n-1) Analysis
5.11
5.11
5.15
5.15
5.3 Scenario 2015 – average hydrology import/export – topology 2010
5.18
5.4 Scenario 2015 – average hydrology import/export – topology 2015
5.4.1. Lines Loadings
5.4.2. Voltage Profile in the Region
5.4.3. Security (n-1) Analysis
5.4.4. Summary of Impacts - 2015 topology versus 2010 topology
5.20
5.20
5.25
5.26
5.27
5.5 Scenario 2015 – average hydrology high load – topology 2010
5.5.1. Lines Loadings
5.5.2. Voltage Profile in the Region
5.5.3. Security (n-1) Analysis
5.28
5.28
5.32
5.33
5.6 Scenario 2015 – average hydrology high load – topology 2015
5.6.1. Lines Loadings
5.6.2. Voltage Profile in the Region
5.6.3. Security (n-1) Analysis
5.6.4. Summary of Impacts - 2015 topology versus 2010 topology
5.36
5.36
5.40
5.40
5.42
6. ANALYSES SUMMARY AND RECOMMENDATIONS
6.1 Reference cases
6.1.1. Analyses comparison
6.1.2. Overview on possible solutions for system relief
6.1
6.2
6.2
6.5
v
6.2 Sensitivity cases
6.2.1. Analyses comparison
Import/export sensitivity scenarios
High load growth rate
6.2.2. Overview on possible solutions for system relief
6.7
6.7
6.7
6.8
6.11
6.3 List of priorities
6.12
LITERATURE
7.1
vi
ABBREVIATIONS
Country codes
Country
AUT
ALB
BUL
BiH
SUI
GER
GRE
HUN
CRO
ITA
MKD
ROM
SLO
TUR
UKR
SCG
--
Name
Österreich
Shqiperia
Bulgarija
Bosna i Hercegovina
Schweiz
Deutschland
Hellas
Magyarorszag
Hrvatska
Italia
FYR Makedonija
Romania
Slovenija
Türkiye
Ukraina
Srbija i Crna Gora
Fictitious border node
country
code
nodes
O
A
V
W
S
D
G
M
H
I
Y
R
L
T
U
J
X
country
code
ISO
AT
AL
BG
BA
CH
DE
GR
HU
HR
IT
MK
RO
SI
TR
UA
CS
--
Other abbreviations
AC
Alternating current
DC
Direct current
CCGT
Combined Cycle Gas Turbine
CHP
Combined Heat and Power
HPP
Hydro-power plant
NPP
Nuclear power plant
TPP
Thermal power plant
FACTS
Flexible AC transmission systems
GDP
Gross domestic product
MW/Mvar
Megawatt/Megavar
HV
High voltage
NTC
Net Transfer Capacity
SEE
South Eastern European Countries
SECI
South East European Cooperation Initiative
RTSM
Regional transmission system model
TRM
Transmission Reliability Margin
TSO
Transmission System Operator
TR
Transformer
HL
High voltage line
OHL
Overhead high voltage line
PSS/E
Power System Simulator for Engineering
Abbreviations of Electric power utilities and Transmission system operators:
EPCG
Electric power utility of Montenegro
EPS
Electric power utility of Serbia
HTSO
Hellenic transmission system operator
CENTREL
Association of transmission system operators of Czech, Hungary, Poland, and Slovakia
UCTE
Union for the Coordination of Transmission of Electricity
vii
1 INTRODUCTION
The main objective of the Generation Investment Study is to assist the EC, IFIs and donors in
identifying an indicative priority list of investments in power generation and related electricity
infrastructure from the regional perspective and in line with the objectives of SEE REM. The study
would determine what the optimal timing, size and location would be of future generating capacity
in the region over the next 15 years (2005 – 2020). It will also identify priority investments in main
transmission interconnections between the countries and sub-regions to help optimize investment
requirements in power generation over the study horizon. The expansion of the generation system
will be optimized over a 15 year horizon (2005 – 2020) for three scenarios (isolated operation of
each power system, regional operation of power systems, market conditions) using the WASP and
GTMax models.
The third scenario includes power system constraints, such as the capacity of the interconnections,
and other constraints such as the maximum amount of import capacity and energy each country will
be willing to depend upon. For the analysis of the third scenario the study team used, in addition to
the WASP model, the Generation and Transmission Maximization Model (GTMax). GTMax was
used to perform modeling of the electricity grid operated under the SEE REM conditions. It took
into account the transfer capabilities of the interconnection lines among the utility systems.
For more detailed view on regional transmission network operation under market conditions SECI
Regional Transmission System Model in PSS/E was used. Input data concerning demand and
production for several scenarios analyzed in GTMax was used in PSS/E Regional Transmission
System Model (PSS/E RTSM) in order to check feasibility of such demand/production scenarios
from transmission network prospective.
PSS/E RTSM was created by SECI (South East Europe Cooperative Initiative) Project Group on the
Regional Transmission System Planning, sponsored by USAID. With a participation of all power
system utilities in South-East Europe, the Project Group finalized PSS/E RTSM for year 2005 and
2010, suitable for load flow, short-circuit and dynamic analysis. The following countries/companies
were involved in PSS/E RTSM creation: Albania – KESH; Bosnia and Herzegovina – ZEKC,
EPBiH, EPRS, EPHZHB; Bulgaria – NEK; Croatia – HEP, EIHP; Macedonia – ESM; Greece –
PPC/HSTO; Hungary – MVM; Romania – Transelectrica; Serbia and Montenegro – EPS, EKC,
EPCG; Slovenia – ELES; Turkey – TEAS. Two models were created for each time horizon: winter
maximum demand and summer minimum demand models for 2005 and 2010.
Analysis on PSS/E RTSM should provide insight to transmission network adequacy and determine
what transmission reinforcements or additions priorities are eventually required to meet GTMax
2010 and 2015 generation dispatch under normal and (n-1) operating conditions.
Total of 14 GTMax scenarios were analyzed on PSS/E RTSM:
Scenario 2010 - topology 2010
- average hydrology
- dry hydrology
- wet hydrology
Scenario 2015
- average hydrology - topology 2010
- topology 2015
- dry hydrology
- topology 2010
- topology 2015
1.1
- wet hydrology
- topology 2010
- topology 2015
Scenario 2010
– average hydrology import/export – topology 2010
high load – topology 2010
Scenario 2015
– average hydrology import/export – topology 2010
– topology 2015
high load
– topology 2010
– topology 2015
PSS/E RTSM was adjusted according to GTMax model concerning network topology, demand,
production and exchange data. SEE Region was observed as self sufficient without any additional
exchanges with UCTE.
For each GTMax scenario steady-state load flows were calculated and contingency analyses (n-1)
were performed. Security criterion based on voltage profile and lines congestions (thermal loadings)
were checked for each analyzed scenario. Special attention was directed on existing and planned
interconnection lines between different SEE power systems (countries).
Possible network bottlenecks were identified and some solutions for transmission system relief
were described. The role of new interconnection lines candidates in bottlenecks removal was
evaluated.
Following chapters describe PSS/E RTSM, analyzed GTMax production, demand and exchange
scenarios, results of load flow and contingency analyses. List of important regional transmission
system infrastructure projects from a viewpoint of support given to analyzed GTMax dispatch
scenarios is presented at the end of this report.
1.2
2.
STARTING CONDITIONS
2.1
2.1.
Description of PSS/E Model
For all calculations performed in this part of the study, professional software package PSS/E™
(Power System Simulator for Engineering) is used. The Power System Simulator for Engineering
(PSS/E) model is a system of computer programs and structured data files designed to handle the
basic functions of power system performance simulation work, namely:
• Data handling, updating, and manipulation
• Power Flow
• Optimal Power Flow
• Fault Analysis
• Dynamic Simulations + Extended Term dynamic Simulations
• Open network Access and Price calculation
• Equivalent Construction
Since its introduction in 1976, the PSS/E tool has become the most comprehensive, technically
advanced, and widely used commercial program of its type. It is widely recognized as the most
fully featured, time-tested and best performing commercial product available in the market. The
program employs the latest technology and numerical algorithms to efficiently solve networks large
and small.
PSS/E is comprised of the following modules:
PSS/E Power Flow: This module is basic PSS/E program module and it is powerful and easy-touse for basic power flow network analysis (Figure 2.1.1). Besides analysis tool this module is also
used for Data handling, updating, and manipulation.
Figure 2.1.1 – PSS/E model Graphical interface
PSS/E Optimal Power Flow (PSS/E OPF): PSS/E Optimal Power Flow (PSS/E OPF) is a
powerful and easy-to-use network analysis tool that goes beyond traditional load flow analysis to
fully optimize and refine a transmission system. This task is achieved with the integration of PSS/E
OPF into the PSS/E load flow program. PSS/E OPF improves the efficiency and throughput of
power system performance studies by adding intelligence to the load flow solution process.
2.2
Whereas the conventional load flow relies on an engineer to systematically investigate a variety of
solutions before arriving at a satisfactory solution, PSS/E OPF directly changes controls to quickly
determine the best solution. From virtually any reasonable starting point, you are assured that a
unique and globally optimal solution will be attained; one that simultaneously satisfies system
limits and minimizes costs or maximizes performance.
PSS/E Balanced or Unbalanced Fault Analysis: The PSS/E Fault Analysis (short circuit)
program is fully integrated with the power flow program. The system model includes exact
treatment of transformer phase shift, and the actual voltage profile from the solved power flow case.
PSS/E Dynamic Simulation: PSS/E offers users uncompromising dynamic simulation capabilities.
It models system disturbances such as faults, generator tripping, motor starting and loss of field.
The program contains an extensive library of generator, exciter, governor, and stabilizer models as
well as relay model including underfrequency, distance and overcurrent relays to accurately
simulate disturbances.
The organization of PSS/E is illustrated in the following figure Figure 2.1.2:
Figure 2.1.2 – Organization of PSS/E model modules
Current version of this software package and version used for calculations is version 30.2.
2.3
2.2.
Prerequisites and Assumptions
For more detailed view on regional transmission network operation under market realistic
conditions SECI Regional Transmission System Model (RTSM) in PSS/E is used. Figure 2.1.1
shows which countries and their transmission systems were modeled.
Figure 2.2.1 - SECI Regional Transmission System Model – countries modeled
High voltage transmission network of 750 kV, 400kV, 220kV, 150kV (Greece and Turkey), and
110 kV voltage levels is implemented in the model. Also, all new substations and lines that are
expected to be operational till 2010 (according to the long term development plans) are modeled
too. All generation units that are connected to the transmission voltage level are modeled as they are
in reality (with step-up transformers). Model is designed for load-flow calculations and analysis, but
with adequate data input (already developed and tested) it can be used for other type of analysis too:
• Short-circuit calculations
• Dynamics (transient stability assessment)
Figure 2.2.2 shows the main characteristics of the SECI Regional Transmission System Model.
With adequate changes and appropriate data input this model is used for all calculations and
analyses in this project.
ALBANIA
10
ELEMENT TYPE
#
W S
BUSES
97 97 97
PLANTS
14 12 14
MACHINES
30 23 13
72 72 72
LOADS
LINES
107 107 107
400 4
220 21
150
110 82
66
30
50 44 39
TRANSFORMERS
STEP UP 23
NETWORK 27
AREA
AREA
ELEMENT TYPE
BUSES
PLANTS
MACHINES
LOADS
LINES
SLOVENIA
75
W S
190 190
48 44
54 23
111 111
231 231
#
190
57
63
111
232
400 14
220 9
150
110 209
66
30
TRANSFORMERS
75
STEP UP 57
NETWORK 18
BULGARIA
BIH
CROATIA
20
30
40
#
W
S
#
W
S
#
W
S
635 602 586 236 230 231 247 247 245
112 81 65 38 36 36
59
59
59
144 84 49 42 27 10
62
52
32
644 606 590 137 131 131 162 147 147
747 711 709 278 260 261 325 308 308
40
13
21
56
39
28
651
226
170 129 110
102
68
72
38
34
#
0
TURKEY
80
W
S
75
0
GREECE
FYRM
ROMANIA
50
60
70
#
W
S
#
W
S
#
W
S
993 889 749 117 117 117 1067 1006 963
100 84 56 22 22 22 132 183 139
107 86 54 22 20
3
132 182
87
415 379 267 78 76 73 831 711 679
1051 901 789 145 138 138 1295 1135 1135
96
10
47
3
114
12
0
12
933
132
8
14
260 236 208 32
94
22
166*
10
276
66
65
95
60
35
93
93
12
12
SERBIA
MONTENEGRO
SCG
90
91
90
#
W
S
#
W
S
#
W
S
417 408 368 43
39
35 460 447 403
69 63 25
11
80 70 28
7
3
77 71 25
11
7
3
88 78 28
251 249 240 19
19
19 270 268 259
502 471 468 50
48
48 552 519 516
41
5
46
64
11
75
397
75
HUNGARY
45
#
W
S
47 45 47
8
8
8
23 22 12
32 32 33
249 202 202
114
135
139 129
72
67
34
92
21
11
10
#
3
3
3
0
5
UKRAIN
65
W
3
3
3
0
5
S
3
3
3
0
5
0
0
0
1134
32
32
UCTE-WEST
55
#
W
S
88 88 88
34 32 30
34 31 25
72 72 71
156 149 151
46
110
293
144
149
203
158
#
2
2
2
0
2
CENTREL
95
W
2
2
2
0
2
S
2
2
2
0
2
0
0
0
431
17
13
160 146 105
83
77
19
0
19
19
19
#
4182
661
752
2824
5144
451
590
0
4074
8
14
1238
623
615
REGIONAL
MODEL
W
3963
640
664
2605
4668
0
0
0
0
0
0
1055
0
0
S
3721
506
341
2433
4554
0
0
0
0
0
0
916
0
0
# total number of elements
W number of elements with status in in Winter maximum model
S number of elements with status in in Summer minimum model
* - equivalent lines included
Figure 2.2.2 – SECI Regional Transmission System Model – model characteristics
2.4
Input data concerning demand and production for several scenarios analyzed in GTmax are used ase
input data for PSS/E RTSM in order to check feasibility of such demand/production scenarios from
transmission network prospective.
Electric power systems presented on Figure 2.2.3 are more detail modeled according to GTmax
results.
Hungary
Slovenia
Romania
Croatia
B&H
Serbia
Bulgaria
Monte
negro Unmik
Alba
Mace
donia
Turkey
nia
Greece
Figure 2.2.3: Analyzed electric power systems
For the purpose of network analyses, following network models in PSS/E software tool are
developed according to GTmax:
Reference cases:
Year
Hydrology Topology
average
2010 dry
2010
wet
2010
average
2015
2010
2015 dry
2015
2010
wet
2015
Sensitivity cases (average hydrology):
Special
Year Topology
conditions
2010
2010
Import/Export
2010
2015
2015
2010
2010
High load
2010
2015
2015
2.5
For all these models, expected topology and load distribution in corresponding system substations
are modeled. In Regional Transmission Network Model for the years 2010 and 2015 the most
recent information including new planed lines generator units with step-up transformers,
transformers, compensators, phase shift transformers, shunts, etc are included. Generator units are
connected at generator voltage level.
In the models, whole 110 kV and above network is included. Each interconnection line has assigned
an X node which is placed on border of each country (not at middle of tie line). The model for year
2015 is obtained by increasing production and consumption in each electric power system
according to results from GTmax calculation and respect to new planed lines and substations.
Voltage levels, with the upper and lower limits used in the study, are presented in the Table 2.2.1
These limits are used in load-flow calculations as in contingency analysis.
Table 2.2.1: Defined limits for voltage levels
Defined voltage levels
750 kV
400 kV
220 kV
150 kV
110 kV
min
max
min
max
min
max
min
max
min
max
kV
712
787
380
420
198
242
135
165
99
121
p.u.
0,95
1,05
0,95
1,05
0,90
1,10
0,90
1,10
0,90
1,10
Generator
min
max
0,95
1,05
These limits are according to the operational and planning standards used in the monitored region,
and they are used for full topology and "n-1" analyses. Although, in emergency conditions for some
voltage levels wider voltage limits are allowed, these are not taken into consideration.
The system adequacy is checked for operating conditions using “n-1” contingency criterion.
List of contingencies includes:
• all interconnection lines;
• all 400 and 220 kV lines, except lines which outage cause “island” operation (in case of
parallel lines and double circuit lines, outage of one line-circuit is considered);
• all transformers 400/x kV (in case of parallel transformers, outage of one transformer is
considered).
Current thermal limits are used as rated limits for lines and transformers. These limits are
established based on a temperature to which conductor is heated by current above which either the
conductor material would start being softened or the clearance from conductor to ground would
drop beyond permitted limits. For these analyses is used that conductor current must not reach
limits imposed by thermal limit defined for conductors material and cross-section according to
standard the IEC (50) 466: 1995 – International Electro technical Vocabulary - Chapter 466:
Overhead Lines. For transformers, rated installed MVA power is used as thermal limit. Every
branch with current above its thermal limit is treated as overloaded.
All system states in which voltage level is outside permitted limits or branches are loaded beyond
thermal limit (overloaded), by full topology or "n-1" contingency analyses, are treated as "insecure
states" and referenced as such in this study.
2.6
2.3.
Technical Specification of New Transmission Lines Candidates and Substations
Interconnection lines between analyzed electric power systems are presented on the Figure 2.3.1
and Table 2.3.1.
CENTREL
Slovakia
PIVDENNOUKRAINSKA
AES
WestUkraina
Moldavia
UCTE
Hungary
Austria
HEVIZ
VULKANESTI
SANDORFALVA
Romania
ARAD
ISACCEA
Slovenia
ZERJAVINEC
MRACLIN
ERNESTINOVO
MELINE TUMBRI MEDJURICDJAKOVO
KRSKO
SUBOTICA
S.MITROVICA
TUZLA
PRIJEDOR GRADACAC UGLJEVIK
Croatia (Jajce) Bosnia VISEGRAD
Hercegovina
RAMA
TANTARENI
PORTILE de
FIER
DJERDAP
ISALNITA
VARDISTE
SARAJEVO
ZAKUCAC
MOSTAR 220
KONJSKO TREBINJE
Serbia
PIVA
PLAT
PODGORICA
PRIZREN
MONTENEGRO
KOSOVO B
PERUCICA
VARNA
NIS
KOZLODUY
SOFIA
Bulgaria
DOBRUJA
MARITSA 3
FIERZE
Macedonia
VAU
DEJES
SKOPJE
BLAGOEVGRAD
BABAESKI
DUBROVO
Albania
Turkey
LAGAD
ELBASAN
ZEMLAK
HAMITABAT
KARDIA
THESSALONIKI
Greece
750 kV line (3)
400 kV line (33)
220 kV line (22)
Italy
Interconnected
network of SE
countries
only border lines - status 2005 year
Figure 2.3.1: Interconnection lines in Southeast Europe in 2005 year
Planned interconnection lines in Southeastern Europe for years 2010 and 2015 are given in Figure
2.3.2 and Table 2.3.2 and Table 2.3.3. All of these planned interconnection lines are included in
PSS/E models for target years respectively, depending of years of commissioning.
Basic technical data of new interconnection transmission lines candidates which are analyzed in
2015 year as variants only, are shown in Table 2.3.4. It should be noted that new interconnection
transmission lines candidates, which are shown as doted lines, are analyzed in 2015 year as variants
only.
2.7
Table 2.3.1: List of interconnection lines in Southeast Europe in 2005 year
Interconnection line
Interconnected Voltage
countries
level
(kV)
BG - RO
750
HU - UA
750
RO - UA
750
HU - SK
400
HU - SK
400
AL - GR
400
BA - HR
400
BA - HR
400
BG - GR
400
BG - RO
400
BG - TR
400
RO - MOLD
400
BG - RO
400
BG - SER
400
BG - TR
400
HR - HU
400
MK - GR
400
MK - SER
400
GR - IT
400
HU - AT
400
MN - BA
400
RO - HU
400
RO - SER
400
RO - UA
400
SER - HR
400
SER - HU
400
SI - AT
400
SI - HR
400
SI - HR
400
SI - IT
400
UA - HU
400
AL - MN
220
AL - SER
220
BA - HR
220
BA - HR
220
BA - HR
220
BA - HR
220
BA - HR
220
BA - HR
220
BA - HR
220
BA - MN
220
BA - MN
220
BA - SER
220
HR - SI
220
MK - SER
220
MK - SER
220
HU - AT
220
HU - AT
220
SI - AT
220
SI - HR
220
SI - IT
220
UA - HU
220
UA - HU
220
Type
Conductors
Size
Number
(mm2)
per phase
300
5
400
5
400
5
500/350
2/3
500/450
2/3
500
2
490
2
490
2
500
2
400
3
400
3
400
3
500/300
2/3
500
2
500
2
490
2
490
2
490
2
Transfer
Capacity
(MVA)
2390
5360
5360
1440
1440
1309
1318
1318
1309
1715
1715
1715
2490
1330
1309
1318
1330
1330
500
2563
1330
1212
1330
1212
1330
1330
1330
1318
1318
1330
1386
301
301
300
300
300
300
300
491
491
301
366
301
300
301
301
305
305
366
350
366
308
308
Varna - Isaccea*
ACSR
Albertirsa - Zapadoukrainska
ACSR
Isaccea - Pivdenoukrainska*
ACSR
God - Levice
ACSR
Gyor - Gabcikovo
ACSR
Zemlak - Kardia
ACSR
Mostar4 - Konjsko
ACSR
Ugljevik - Ernestinovo
ACSR
Blagoevgrad - Lagad (Thesaloniki)
ACSR
Dobruja - Isaccea
ACSR
Maritsa Istok - Hamitabat
ACSR
Isaccea - Vulcanesti*
ACSR
Kozlodui - Tintareni (double)
ACSR
Sofia - Nis
ACSR
Maritsa Istok - Babeski
ACSR
Zerjavinec - Heviz (double)
ACSR
Dubrovo - Thessaloniki
ACSR
Skopje - Kosovo B
ACSR
Arachtos - Galatina HVDC
HVDC
Gyor - Wien Sud (double)
ACSR
500
2
Podgorica - Trebinje
ACSR
490
2
Arad - Sandorfalva
ACSR
450/500
2
Portile de Fier - Djerdap
ACSR
967
2
Rosiori - Mukachevo
ACSR
450
2
S. Mitrovica - Ernestinovo
ACSR
490
2
Subotica - Sandorfalva
ACSR
490
2
Maribor - Keinchtal (double)
ACSR
490
2
Divaca - Meline
ACSR
490
2
Krsko - Tumbri (double)
ACSR
490
2
Divaca - Redipuglia
ACSR
490
2
Mukachevo - Sajoszeged
ACSR
400
2
V.Dejes - Podgorica
ACSR
360
1
Fierze - Prizren
ACSR
360
1
Gradacac - Djakovo
ACSR
360
1
Prijedor - Mraclin
ACSR
360
1
Mostar4 - Zakucac
ACSR
360
1
Prijedor2 - Medjuric
ACSR
360
1
TE Tuzla - Djakovo
ACSR
360
1
Trebinje - HE Dubrovnik
ACSR
240
2
Trebinje - HE Dubrovnik
ACSR
240
2
Trebinje - HE Perucica
ACSR
360
1
Sarajevo20 - HE Piva**
ACSR
490
2/1
Visegrad - Pozega
ACSR
360
1
Zerjavinec - Cirkovce
ACSR
360
1
Skopje - Kosovo A
ACSR
360
1
Skopje - Kosovo A
ACSR
360
1
Gyor - Wien Sud
ACSR
360
1
Gyor - Neusiedl
ACSR
360
1
Podlog - Obersielach
ACSR
490
1
Divaca - Pehlin
ACSR
490
1
Divaca - Padriciano
ACSR
490
1
Mukachevo - Kisvarda
ACSR
400
1
Mukachevo - Tiszalok
ACSR
400
1
*not in model
**built as 400kV line Sarajevo20 - B.Bijela and 220kV B.Bijela - Piva, but operated as 220kV Sarajevo20 - Piva
length
km
I to border border to II
150
85
268
254
5
395
88
36
29
15
21
80
41
69
39
53
72
102
81
150
59
90
5
54
14
102
37
86
50
77
99
69
55
60
36
68
/
/
59
63
60
21
5
52
1
2
39
36
41
52
27
21
26
37
41
26
16
32
39
10
8
142
47
21
26
45
19
27
66
49
50
34
32
65
27
7
5
7
5
20
42
61
23
18
51
19
51
18
65
18
65
59
63
55
27
46
20
47
6
10
2
54
10
97
35
total
235
522
400
124
44
101
110
92
174
231
149
59
116
123
127
168
115
104
0
122
81
57
3
75
93
48
63
66
48
49
150
68
71
46
66
99
66
92
12
12
63
84
69
70
82
82
122
82
65
53
11
64
132
2.8
CENTREL
UCTE
Hungary
NADAB
Romania
PECS
Slovenia
TUMBRI
SOMBOR
ERNESTINOVO
S.MITROVICA
BANJA LUKA
Croatia
UGLJEVIK
Bosnia
Hercegovina VISEGRAD
Serbia
PLJEVLJA
MONTENEGRO
PODGORICA
KOSOVO B
Bulgaria
VRANJE
C.MOGILA
Macedonia
VAU DEJES
SKOPJE
STIP
BITOLA
MARITSA 3
BABAESKI
FILIPI
KERHOS
AlbaniaKASHAR
Turkey
FLORINA
ZEMLAK
2010 2015 candidates
750 kV line
400 kV line
220 kV line
Interconnected
network of SE
countries
Greece
planned border lines only
(status 2010 and 2015 year)
Figure 2.3.2: Planned interconnection lines in Southeast Europe in years 2010 and 2015
Table 2.3.2: List of interconnection lines in Southeast Europe, planned to come into operation in year 2010
Interconnection line
Ugljevik - S. Mitrovica
Kashar - Podgorica
Maritsa Istok - Filipi
C. Mogila - Stip
Ernestinovo - Pecs (double)
Bekescaba - Nadab (Oradea)
Florina - Bitola
(Filipi) - Kehros - Babaeski
Interconnected Voltage
countries
level
(kV)
BA - SER
400
AL - MN
400
BG - GR
400
BG - MK
400
HR - HU
400
HU - RO
400
GR - MK
400
GR - TR
400
Type
ACSR
ACSR
ACSR
ACSR
ACSR
ACSR
ACSR
ACSR
Conductors
Size
Number
(mm2)
per phase
490
2
490
2
400
3
490
2
490/500
2
490
2
490
2
400
3
Transfer
Capacity
(MVA)
1330
1330
1715
1330
1330
1178
1330
1715
length
km
I to border border to II
49
29
115
30
133
110
80
70
44
41
31
87
20
18
50
50
total
79
145
243
150
85
118
38
100
Table 2.3.3: List of interconnection lines in Southeast Europe, planned to come into operation in year 2015
Interconnection line
Zemlak - Bitola
Kashar (V. Dejes) - Kosovo B
Skopje - Vranje - (Leskovac) - (Nis)
Interconnected
countries
AL - MK
AL - SER
MK - SER
Voltage
level
(kV)
400
400
400
Type
ACSR
ACSR
ACSR
Conductors
Size
Number
(mm2)
per phase
490
2
490
2
490
2
Transfer
Capacity
(MVA)
1330
1330
1330
length
km
I to border border to II
25
60
135
80
55
137
total
85
215
192
Table 2.3.4: New transmission lines candidates
Interconnection line
Visegrad - Pljevlja
Tumbri - Banja Luka
Pecs - Sombor
Interconnected Voltage
countries
level
(kV)
BA - MN
400
HR - BA
400
HU - SER
400
Type
ACSR
ACSR
ACSR
Conductors
Size
Number
(mm2)
per phase
490
2
490
2
490
2
Transfer
Capacity
(MVA)
1330
1330
1330
length
km
I to border border to II
30
30
50
150
20
60
total
60
200
80
2.9
All interconnection lines candidates that are investigated in the analyses are shown as dotted lines in
Figure 2.3.2. Table 2.3.5 shows typical electrical parameters of different types of lines. These
parameters are used for both the planned lines and interconnection lines candidates in the Study.
Table 2.3.5: Electrical parameters for OHL per phase and kilometer
*Type of conductor
A
-
B*
C
ACSR
ACSR
CARDINAL
ACSR
1x490 mm2 3x400 mm2
360 mm2
Positive sequence
2x490 mm2
Series resistance
r [Ω/km per phase]
0.0294
0.058
0.0207
0.08
Series reactance
x [Ω/km per phase]
0.341
0.427
0.2824
0.436
Charging susceptance
b [µS/km per phase]
3.371
2.67
4.056
2.6
1920
960
2292
790
Rated current [A]
* In Turkey and Greece, Canadian-American standards are used
Costs of the New Interconnection Overhead Lines
Electricity towers, and the wires and conductors that they support, are the major way of transmitting
electricity. They are generally a lattice steel structure with a number of cross arms. The type, size,
height and spacing of towers are determined by geographical, operational, safety and environmental
considerations. A typical overhead line route will involve three types of tower:
−
−
−
suspension (used for straight lines),
deviation (where the route changes direction),
terminal (where the lines connect with substations or underground cables).
A suspension tower is typically between 40 to 60 meters in height with a phase to phase spacing of
between 7 and 25 meters, depending on the type of tower. The two principal types are the “pine”
which narrows at the top and the Y shaped “delta”. The width of the tower right of way will depend
on the level of power to be transmitted but typically range between 30 and 50 meters for 400 kV.
For 400 kV, towers are usually spaced around 350 to 450 meters apart and provide ground
clearance of at least seven meters in all weather conditions. Higher clearances usually apply if the
route crosses motorways or high-pressure water hoses and minimum clearance for trees and public
street lighting also apply. Towers for 400 kV are typically made from steel.
In the absence of a defined methodology to calculate capital charges and costs, and in order to
evaluate investments, unit price method is implemented in following review. Given unit price
related to construction costs take into consideration configuration of terrain (flat land, medium
mountain, high mountain). In Table 2.3.6 and Table 2.3.7 average estimated prices for each new
element that is expected to be in operation in 2010 and 2015 year are presented.
Total investment values for lines construction and corresponding elements till 2010 year is
presented in Table 2.3.6 in EUROS. The investments cover total length of transmission lines and
construction of 400 kV transmission line bays in appropriate substations. Total investment costs of
transmission line bays include costs of following elements:
−
−
construction of 400 kV transmission line itself
transmission line bays (400 kV)
o Breakers
o Disconnectors with blades ground
o Disconnectors without blades
2.10
o Current Measuring Transformers
o Voltage Measuring Transformers
o Lightning Arrester
These total price values of lines do not take into consideration lease of land.
Table 2.3.6: Total investment sum of interconnection lines in Southeast Europe, planned in year 2010
Lines
Interconnection line
Ugljevik - S. Mitrovica
Kashar (V. Dejes) - Podgorica
Maritsa Istok - Filipi
C. Mogila - Stip
Ernestinovo - Pecs (double)
Bekescaba - Nadab (Oradea)
Florina - Bitola
(Filipi) - Kehros - Babaeski
Interconnected
countries
Length
BA - SER
AL - MN
BG - GR
BG - MK
HR - HU
HU - RO
GR - MK
GR - TR
km
79
145
243
150
85
118
38
100
Unit
price
Euro/km
200,000
235,000
235,000
220,000
240,000
250,000
235,000
240,000
TL Bays
Total
price
Euro
15,720,000
34,075,000
57,105,000
33,000,000
20,400,000
29,500,000
8,930,000
24,000,000
Unit
Number
Total
price
of bays
price
Euro
Euro
650,000
2
1,300,000
650,000
2
1,300,000
650,000
2
1,300,000
650,000
2
1,300,000
650,000
2
1,300,000
650,000
2
1,300,000
650,000
2
1,300,000
650,000
2
1,300,000
Lines & Bays
Total price
Year of
commisionning
Euro
17,020,000
35,375,000
58,405,000
34,300,000
21,700,000
30,800,000
10,230,000
25,300,000
2005/06
2006/07
2007/08
2006/07
2007/08
2008
2006/07
2007/08
Comments:
ƒ Interconnection line 400 kV Ugljevik (BA) – S.Mitrovica (CS).
This tie line should increase system stability, security and transmission capacity
between west and east region and between Bosnia and Serbia and Montenegro. It is
under construction (part in Bosnia is finished) and should be completed by the end of
this year.
ƒ
Interconnection line 400 kV Podgorica (CS) – Kashar (V.Dejes) (AL)
This new tie line should increase system stability, security and transmission capacity
between west and east (CS–AL–GR). Feasibility Study was made two years ago.
Participation for construction of this line will be proportional to the length of the line
from the border (27 km in Montenegro and 115 km in Albania). It should be completed
by the year 2006/7. Estimated investment cost is about 35,375 M€.
ƒ
Interconnection line 400 kV Maritsa Istok (BG) – Filipi (GR).
Feasibility Study was made 3 two years ago. Participation for construction of this line
will be proportional to the length of the line from the border (133 km in Bulgaria and
110 km in Greece). To be completed by the year 2007/8. Estimated investment cost is
about 58,5 M€ and investor is unknown yet.
ƒ
Interconnection line 400 kV C. Mogila (BG) – Stip (MK)
Feasibility Study was made 2-3 two years ago. This new tie line should increase system
stability, reliability, security and transmission capacity between north and south as well
as between Bulgaria and Macedonia. Participation for construction of this line will be
proportional to the length of the line from the border (80 km in Bulgaria and 70 km in
Greece). It should completed by the year 2006/7. Estimated investment cost is about 35
M€. Investor is unknown yet.
ƒ
Interconnection line 400 kV Ernestinovo (HR) – Pecs (HU) (double line)
This tie line should increase system stability, security and transmission capacity
between north and south regions and between Croatia and Hungary. It is included
development plans of HEP (Electric Company of Croatia) and MVM (Electric
Company of Hungary). It should be completed by the year 2007/8. Estimated
investment cost is about 21.7 M€. Investor is unknown yet.
ƒ
Interconnection line 400 kV Bekescsaba (HU) – Nadab (Oradea) (RO)
Construction Agreement between MVM and Transelectrica is under signing procedure.
It should be completed by the year 2008.
2.11
ƒ
Interconnection line 400 kV Florina (GR) – Bitola (MK)
Feasibility Study was made 3 two years ago. This line should increase system security
and transmission capacity between north and south (MK–GR). Participation for
construction of this line will be proportional to the length of the line from the border (20
km in Greece and 18 km in Macedonia). It should be completed by the year 2006/7.
Estimated investment cost is about 10 M€. Investor is unknown yet.
ƒ
Interconnection line 400 kV Filipi (Kehros) (GR) – Babaeski (TR)
Feasibility Study was made 2-3 two years ago. Participation for construction of this line
will be proportional to the length of the line from the border (190 km in Greece and 70
km in Turkey). It should be completed by the year 2007/8. Estimated investment cost is
about 25 M€. Investor is unknown yet.
Total investment values for lines construction and corresponding elements till 2015 is presented in
Table 2.3.7 in EUROS. The investments cover total length of transmission line and construction of
400 kV bays in both substations.
Table 2.3.7: Total investment sum of interconnection lines in Southeast Europe, planned in year 2015
Lines
Interconnection line
Zemlak - Bitola
Kashar (V. Dejes) - Kosovo B
Skopje - Vranje - (Leskovac) - (Nis)
Interconnected
countries
Length
AL - MK
AL - SER
MK - SER
km
85
215
192
Unit
price
Euro/km
200,000
220,000
240,000
TL Bays
Total
price
Euro
17,000,000
47,300,000
46,080,000
Unit
Number
Total
price
of bays
price
Euro
Euro
650,000
2
1,300,000
650,000
2
1,300,000
650,000
2
1,300,000
Lines & Bays
Total price
Year of
commissioning
Euro
18,300,000
48,600,000
47,380,000
2010/15
2010/15
2010/15
Comments:
ƒ
Interconnection line 400 kV Zemlak (AL) – Bitola (MK)
It is included in development program of KESH-Albania. Estimated investment cost is
about 18 M€. Investor is unknown yet.
ƒ
Interconnection line 400 kV Kashar (V.Dejes) – Kosovo B (CS)
It is included in transmission development program of KEK-Kosovo. Estimated
investment cost is about 48 M€. Investor is unknown yet.
ƒ
Interconnection line 400 kV Skopje (MK) – Vranje – (Leskovac – Nis) (CS)
This tie line should increase transmission capacity, system stability and security
between north and south. Feasibility Study was made one year ago. Estimated
investment cost is about 21 M€ (for Skopje – Vranje) plus about 26 M€ (for Vranje –
Leskovac – Nis). Donation of Greece government is expected. First phase of the project
is building of the Skopje–Leskovac–Nis, and in second phase is installing of the
substation Vranje on this line.
Costs of New Substations in 2010 and 2015
The purpose of the substation (or switchyard) is to transform the voltage of electricity or switch
electricity circuits. Substations are usually contained within secure sites to ensure public safety.
Most substations today are unmanned sites although road access is necessary for staff and for the
2.12
transport of equipment, maintenance or repair. Table 2.3.8 and Table 2.3.9 present total investment
costs that include costs of following elements:
−
−
transformer unit
transformer bays (400 kV, 220 kV, 110 kV)
o Breakers
o Disconnectors without blades
o Current Measuring Transformers
o Voltage Measuring Transformers
o Lightning Arrester
Also these tables show investment cost for all linked lines between existed transmission lines and
new substations.
Table 2.3.8: new SS which will be in operation till 2010 year
2010
1
2
3
4
Kashar
Sofia South
Maritsa East 1
Pleven
Voltage
levels
kV/kV
400/220
220/110
400/220
220/110
5
6
7
8
Ernestinovo
Stip
Nadab
Oradea
400/110
400/110
400
400/110
1x300
1x300
1x300
no transformer
new transformer only
Total
9
10
11
S. Mitrovica
Beograd 20
Kolubara B
400/220
400/110
400/110
1x400
1x300
2x300
new transformer only
Pec
400/110
UNMIK
12
*position in geographical map (Figure 2.4.2)
1x300
Country
Albania
Bulgaria
Croatia
Macedonia
Romania
Serbia
Position*
Name of
substation
New
transformers
MVA
2x400
1x250
1x630
1x200
Total cost
in Euros
Remarks
new transformer only
phase-shifter
new transformer only
Total
new transformer only
Total
4,100,000
1,830,000
3,600,000
1,730,000
7,160,000
2,730,000
3,430,000
1,400,000
2,730,000
4,130,000
3,150,000
2,730,000
5,930,000
11,810,000
4,130,000
Table 2.3.9: new SS which will be in operation till 2015 year
2015
Romania
13
14
15
V. Dejes
Brasov
Suceava
Voltage
levels
kV/kV
400/200
400/110
220/110
Serbia
16
17
Nis
Vranje
400/110
400/110
Country
Albania
Position*
Name of
substation
New
transformers
MVA
1x400
1x300
1x200
1x300
1x300
Total cost
in Euros
Remarks
new transformer only
new transformer only
Total
new transformer only
Total
3,150,000
2,730,000
1,730,000
4,460,000
2,730,000
4,130,000
6,860,000
*position in geographical map (Figure 2.4.4)
2.13
2.4.
Remarks on PSS/E Transmission System Model and GTmax Model Harmonization
To investigate regional market conditions in region, GTmax software tool is used. Unlike PSS/E,
with this software tool you can not analyze network conditions. Only in rough congestion analyses
is possible. In order to check, whether production distribution, that is result of GTmax analyses is
feasible from transmission system capabilities point of view, results from GTmax analyses has been
used as input for PSS/E regional models, and then full security analysis of thus gained models has
been performed.
Load-demand
GTmax load-demand level is distributed only in few GTmax network buses, unlike PSS/E model
where load distribution corresponds to the real system conditions (substation by substation). Also,
GTmax load-demand level includes transmission network losses, unlike PSS/E model where these
losses are calculated separately. In other words, it was not possible to achieve full load-demand
correspondence between GTmax and PSS/E model.
In order to achieve at least partial correspondence between GTmax load-demand and PSS/E load
(on system by system level), existing load distribution in PSS/E model is scaled so level of PSS/E
load increased for transmission network losses corresponds to GTmax load-demand level. Also, for
some systems, production of units installed on voltage levels lower than 110 kV is included in loaddemand level (load-demand is reduced for this amount). This was done system by system.
Network topology
In order for PSS/E network models to correspond to GTmax model, first step was to adjust network
topology of both PSS/E and GTmax models. Topology of GTmax models for 2010 and 2015 are
shown on Figure 2.4.1 and Figure 2.4.3 and corresponding real network topologies of regional
network shown on Figure 2.4.2 and Figure 2.4.4.
As it can be seen, the internal system topologies are quite rough, unlike topology of system
interconnections (tie lines), which are almost identical (that was one of prerequisites of GTmax
model).
Transmission system model for 2015 was gained using existing model for 2010. All new
interconnection lines that are predicted to be in operation in GTmax's model are also included in
PSS/E model. Also, all internal network reinforcements that are in long term plans of regional TSOs
are modeled too.
2.14
UKRAINE
CENTREL
SLOVENIA
Romania
Rosiori
0 ->
270 00
7
<- 2
Gutinas
Sibiu
13
<- 50
1 3 ->
50
Osijek
1350
<- 1 ->
330
1330 ->
<- 1330
->
50 50
13 - 13
<
Tuzla
12
<- 20 ->
13
30
13
<- 30 ->
13
30
Varna
Sofia
Bulgaria
Chervena
Mogila
Skopje
00
12
120
0
Plovdiv
Dubrovo
12
2
<- 0 ->
12
20
Maritsa
Blagoevgrad
->
690 0
9
<- 6
->
00 0
31 310
<-
1200
Vetren
0 ->
122 220
<- 1
Macedonia
Bitola
Tirana/Elbasan
->
1330 0
3
< - 13
->
30 0
19 183
<-
ia
Winter ->
<- Winter
V. Dejes
->
30
13 1330
<-
Thermal / Overcurrent
Protection Limits in MW
>
031 300
<-
an
Alb
Legend:
30
<- 0 ->
28
0
Kozlodui
Nis
Montenegro
Isaccea
1500 ->
<- 1300
300
<- 3 >
00
Ribarevina
1330
-> Kosovo B
<- 1 3
30
Podgorica
Tantareni
Pljevlja
Trebinje
0 ->
163 30
6
<- 1
Split
>
030 300
<-
Bucharest
1500
->
<- 1 3
00
->
0 0
66 66
<-
Mostar
0 ->
135 50
3
<- 1
B. Basta
1250 ->
<- 1250
Sarajevo
0 ->
125 250
<- 1
B&H
Portile
de Fier
1330->
<- 1680
Djerdap
1065 ->
< - 1065
B. Luka
Arad
Belgrade
1250
->
<- 12
50
900
< - ->
80
0
Rijeka
Serbia
13
< - 00 13 >
00
Croatia
Zagreb
27
0
< - 0 ->
27
00
1330 ->
<- 1330
>
0270 700
2
<-
1350 ->
<- 1350
33
<- 0 ->
33
0
80
<- 0 ->
17
50
TURKEY
1350
->
< - 13
50
GREECE
Figure 2.4.1: GTmax network topology in year 2010
2.15
Figure 2.4.2: Real network topology in year 2010
2.16
UKRAINE
CENTREL
SLOVENIA
Romania
Rosiori
0 ->
270 00
7
<- 2
Gutinas
Sibiu
13
<- 50
1 3 ->
50
Osijek
1350
<- 1 ->
330
1330 ->
30
13
<-
->
50 50
13 - 13
<
Tuzla
12
<- 20 ->
13
30
Varna
Sofia
Bulgaria
Chervena
Mogila
Skopje
120
0
Plovdiv
Dubrovo
12
2
< - 0 ->
12
20
Maritsa
Blagoevgrad
->
690 0
9
<- 6
->
00 0
31 310
<-
1200
Vetren
0 ->
122 220
<- 1
Macedonia
Bitola
Tirana/Elbasan
13
<- 30 ->
13
30
00
12
ia
Winter ->
<- Winter
V. Dejes
->
30
13 1330
<-
Thermal / Overcurrent
Protection Limits in MW
>
031 300
<-
an
Alb
Legend:
30
<- 0 ->
28
0
->
1330 0
3
< - 13
->
30 0
19 183
<-
Montenegro
Podgorica
Kozlodui
Nis
Ribarevina
1330
-> Kosovo B
<- 1 3
30
Isaccea
1500 ->
<- 1300
300
<- 3 >
00
Pljevlja
Trebinje
0 ->
163 30
6
<- 1
Split
>
030 300
<-
1500
->
<- 1 3
00
->
0 0
66 66
<-
Mostar
0 ->
135 50
3
<- 1
B. Basta
Bucharest
Tantareni
1250 ->
<- 1250
Sarajevo
0 ->
125 250
<- 1
B&H
Portile
de Fier
1330->
<- 1680
Djerdap
1065 ->
< - 1065
B. Luka
Arad
Belgrade
1250
->
<- 12
50
900
< - ->
80
0
Rijeka
Serbia
13
< - 00 13 >
00
Croatia
Zagreb
27
0
< - 0 ->
27
00
1330 ->
<- 1330
>
0270 700
2
<-
1350 ->
<- 1350
33
<- 0 ->
33
0
80
<- 0 ->
17
50
TURKEY
1350
->
< - 13
50
GREECE
Figure 2.4.3: GTmax network topology in year 2015
2.17
Figure 2.4.4: Real network topology in year 2015
2.18
Production distribution
In PSS/E model, all production units connected to 110 kV or higher voltage network are modeled as
they are in reality, generation units plus corresponding step-up transformers. Some of the
production units or plants connected to the network lower than 110 kV are modeled as negative load
in corresponding buses, but most of them are included in corresponding loads in system substations.
Also, in GTmax models, most of the production units are modeled as whole plants, so in order to
achieve full compliance with PSS/E model, it was necessary to distribute this production on the real
units themselves, as they are in PSS/E model. In following figures and tables production
distribution from GTmax to PSS/E model was shown, jurisdiction by jurisdiction, but also how
these units are connected to the GTmax network model, and appropriate network representation in
PSS/E model.
Albania
GTmax
Powerplant
Fier 1
Balsh
Ulez+Shkopet
Bistrica
Fierza
Komani
Vau Dejes
Vlore
Bus#
10021
10022
10022
10052
10052
10052
10052
10053
10054
10490
10490
10490
10490
10501
10502
10503
10504
10511
10512
10513
10514
10521
10522
10523
10524
10525
10551
10552
10553
10554
PSS/E
Bus Name
AFIER 9 11.000
AFIERV9 6.3000
AFIERV9 6.3000
AULEZ 9 6.3000
AULEZ 9 6.3000
AULEZ 9 6.3000
AULEZ 9 6.3000
ASHKP19 6.3000
ASHKP29 6.3000
ABISTRIG 6.3000
ABISTRIG 6.3000
ABISTRIG 6.3000
ABISTRIG 6.3000
AFIERZ91 13.800
AFIERZ92 13.800
AFIERZ93 13.800
AFIERZ94 13.800
AKOMAN91 13.800
AKOMAN92 13.800
AKOMAN93 13.800
AKOMAN94 13.800
AVDEJA91 10.500
AVDEJA92 10.500
AVDEJA93 10.500
AVDEJA94 10.500
AVDEJA95 10.500
ABABICG1 13.800
ABABICG2 13.800
ABABICG3 13.800
ABABICG4 13.800
Remarks
ID
1
4
5
1
2
3
4
1
1
G1
G2
G3
G4
1
2
3
4
1
2
3
4
1
2
3
4
5
1
2
3
4
Figure 2.4.5 – GTmax model - Albania
2.19
Figure 2.4.6 – PSS/E model - Albania
2.20
Bulgaria
GTmax
Powerplant
BELMEKEN
SOFIA HEAT
BOBOV DOL
SESTRIMO
CHAIRA
PERNIK CHP
MOMINA KLISURA
KOZLODUY 1
KOZLODUY 2
RUSE 12
RUSE 34
RUSE 56
SOFIA IND P
VARNA THERMAL
VARNA CHP
PLOVDIV CHP
MARITSA EAST 2 1
MARITSA EAST 2 N
MARITSA EAST 3
BRIKEL
PESTERA
ORPHEY
Bus#
13200
13201
13201
13202
13202
13206
13207
13208
13209
13209
13300
13301
13302
13303
13304
13305
13305
13310
13310
13306
13307
13307
13308
13309
13404
13405
13500
13500
13501
13502
13503
13504
13505
13505
13600
13601
13603
13604
13605
13602
13606
13607
13607
13700
13700
13701
13702
13703
13704
13705
13706
13707
13708
13709
13710
13711
13712
13713
13714
13715
13715
13716
13717
13800
13800
13801
13801
13802
13803
13803
13803
PSS/E
Bus Name
BELM_G0 10.500
BELM_G12 10.500
BELM_G12 10.500
BELM_G34 10.500
BELM_G34 10.500
TSF.I.G5 13.800
TSF.I.G4 6.3000
TSF.I.G3 6.3000
TSF.IG12 6.3000
TSF.IG12 6.3000
TBDOL_G1 15.750
TBDOL_G2 15.750
TBDOL_G3 15.750
HSES.G1 10.500
HSES.G2 10.500
HCHA.G12 19.000
HCHA.G12 19.000
CHA.G34 19.000
CHA.G34 19.000
TREP.G5 10.500
TREP.G34 6.3000
TREP.G34 6.3000
HMKL.G1 10.500
HMKL.G2 10.500
NKOZ_G9 24.000
NKOZ_G10 24.000
TRUSEG12 6.3000
TRUSEG12 6.3000
TRUSEG3 13.800
TRUSEG4 13.800
TRUSEG5 6.3000
TRUSEG6 6.3000
TSV_G12 6.3000
TSV_G12 6.3000
TVARN_G1 15.750
TVARN_G2 15.750
TVARN_G4 15.750
TVARN_G5 15.750
TVARN_G6 15.750
TVARN_G3 15.750
TDEV.G3 10.500
TDEV_G14 6.3000
TDEV_G14 6.3000
TMI1_G12 6.3000
TMI1_G12 6.3000
TMI1_G3 6.3000
TMI1_G4 6.3000
TMI2_G1 18.000
TMI2_G2 18.000
TMI2_G3 18.000
TMI2_G4 18.000
TMI2_G5 15.750
TMI2_G6 15.750
TMI2_G7 15.750
TMI2_G8 15.750
TMI3_G1 15.750
TMI3_G2 15.750
TMI3_G3 15.750
TMI3_G4 15.750
TBR.G34 6.3000
TBR.G34 6.3000
TBR.G5
6.3000
TBR.G6
6.3000
HPESHG12 10.500
HPESHG12 10.500
HPESHG34 10.500
HPESHG34 10.500
HPESH.G5 10.500
HANTG123 10.500
HANTG123 10.500
HANTG123 10.500
Remarks
ID
H0
H1
H2
H3
H4
D5
D4
D3
D1
D2
T1
T2
T3
H1
H2
H1
H2
H3
H4
D5
D3
D4
H1
H2
N9
N0
T1
T2
T3
T4
T5
T6
I1
I2
T1
T2
T4
T5
T6
T3
I1
I1
I4
T1
T2
T3
T4
T1
T2
T3
T4
T5
T6
T7
T8
T1
T2
T3
T4
I3
I4
I5
I6
H1
H2
H3
H4
H5
H1
H2
H3
2.21
13804
13805
13805
BATAK
13805
13806
13807
KRICHIM
13808
13809
DEVIN
13810
13811
TESHEL
13811
MARITSA 3
13812
13813
ALEKO
13814
13815
13819
13819
KARDJALY
13820
13820
13821
13821
STUDEN KLADENEC
13822
13822
13823
IVAILOVGRAD
13823
13823
13906
MARITSA EAST 1
13907
13908
LUKOIL
BUL SMALL HYDRO
-
HANT.G4
HBATG123
HBATG123
HBATG123
HBAT_G4
HKRI.G1
HKRI.G2
DEVIN.G1
DEVIN.G2
HTESH.G
HTESH.G
TMAR3.G3
HALEKOG3
HALEKOG2
HALEKOG1
HKAR.G12
HKAR.G12
HKAR.G34
HKAR.G34
HST.KG12
HST.KG12
HST.KG34
HST.KG34
HIW.G123
HIW.G123
HIW.G123
TMI1_G5
TMI1_G6
TMI1_G7
-
10.500
10.500
10.500
10.500
10.500
10.500
10.500
10.500
10.500
10.500
10.500
13.800
10.500
10.500
10.500
10.500
10.500
10.500
10.500
10.500
10.500
10.500
10.500
10.500
10.500
10.500
15.750
15.750
15.750
H4
H1
H2
H3
H4
H1
H2
H1
H2
H1
H2
T3
H3
H2
H1
H1
H2
H3
H4
H1
H2
H3
H4
H1
H2
H3
T5
T6
T7
- included in total consumption
- included in total consumption
Figure 2.4.7 – GTmax model – Bulgaria
2.22
Figure 2.4.8 – PSS/E model - Bulgaria
2.23
BIH
GTmax
Powerplant
UGLJEVIK
GACKO
VISEGRAD
TREBINJE HYDRO
BOCAC
GRABOVICA
SALAKOVAC
KAKANJ 7
TUZLA 4
TUZLA 5
TUZLA 6
JABLANICA
KAKANJ 5
KAKANJ 6
TUZLA 3
RAMA
MOSTAR HYDRO
CAPLJINA
JAJCE 1
JAJCE 2
PEC MLINI
DUBROVNIK G-2
M.BLATO
Bus#
14001
14002
14003
14003
14003
14004
14004
14004
14006
14007
16001
16002
16003
16004
16005
16006
16007
16008
16009
16011
16012
16013
16014
16015
16016
16025
16026
16033
18001
18002
18003
18004
18005
18006
18007
18008
18009
18010
18011
18012
18015
18016
-
PSS/E
Remarks
Bus Name
ID
TE UGLJE 20.000 1
GACKO
20.000 1
HE VISE 15.750
1
HE VISE 15.750
2
HE VISE 15.750
3
HE TREB 14.400 1
HE TREB 14.400 2
HE TREB 14.400 3
BOCACG1 10.500 1
BOCACG2 10.500 2
GRAB-G1 10.500 1
GRAB-G2 10.500 2
SAL-G1 13.800
1
SAL-G2 13.800
2
SAL-G3 13.800
3
KAK-G7 15.750 7
TUZ-G4
15.750
4
TUZ-G5
15.750
5
TUZ-G6
15.750
6
JAB-G1 6.3000
1
JAB-G2 6.3000
2
JAB-G3 6.3000
3
JAB-G4 6.3000
4
JAB-G5 6.3000
5
JAB-G6 6.3000
6
KAK-G5 13.800 5
KAK-G6 13.800 6
TUZ-G3
10.500
3
RAMA G1 15.650 1
RAMA G2 15.650 2
MOST-G1 10.500 1
MOST-G2 10.500 2
MOST-G3 10.500 3
CAPL-G1 15.700 1
CAPL-G2 15.700 2
JAJ1-G1 10.500
1
JAJ1-G2 10.500
2
JAJ2-G1 6.3000
1
JAJ2-G2 6.3000
2
JAJ2-G3 6.3000
3
MLINI-G1 10.500 1
MLINI-G2 10.500 2
- included in Croatian production
included in total consumption
Figure 2.4.9 – GTmax model - BIH
2.24
Figure 2.4.10 – PSS/E model - BIH
2.25
Croatia
GTmax
Powerplant
ZAGREB EL TO A
ZAGREB EL TO H
CAKOVEC
DUBROVNIK G-1
DUBROVNIK G-2
GOJAK
ORLOVAC
PERUCA
RIJEKA HPP
SENJ
SKLOPE
VINODOL
ZAKUCAC
JERTOVEC
KRALJEVAC
OSIJEK A
VARAZDIN
DUBRAVA
VELEBIT
PLOMIN 1
PLOMIN 2
RIJEKA THERMAL
SISAK
ZAGREB TE TO A
ZAGREB ELTO K
ZAGREB TE TO C
DJALE
CC 480 2
CC 480
GOLUBIC
MILJACKA
OSIJEK B
OZALJ 1
OZALJ 2
Bus#
20301
20302
20304
20305
20306
20307
20308
20309
20310
20313
20314
20315
20316
20317
20318
20319
20320
20321
20322
20323
20324
20325
20326
20327
20328
20329
20330
20331
20332
20333
20334
20335
20335
20335
20335
20336
20340
20341
20342
20343
20344
20345
20346
20347
20348
20350
20351
20352
20354
20355
20353
20360
20361
20430
20431
20420
20421
-
PSS/E
Remarks
Bus Name
ID
EL-TOG1 10.500
1
EL-TOG2 10.500
2
HECAKOG1 6.3000 1
HECAKOG2 6.3000 2
HEDUBRG1 6.3000 1
HEDUBRG2 6.3000 2
HEGOJAG1 10.500 1
HEGOJAG2 10.500 2
HEGOJAG3 10.500 3
HEORLOG1 10.500 1
HEORLOG2 10.500 2
HEORLOG3 10.000 3
HEPERUG1 10.500 1
HEPERUG2 10.500 2
HERIJEG1 10.500
1
HERIJEG2 10.500
2
HESENJG1 10.500 1
HESENJG2 10.500 2
HESENJG3 10.500 1
HESKLOG1 10.500 1
HEVINOG1 10.500 1
HEVINOG2 10.500 2
HEVINOG3 10.500 3
HEZAKUG1 16.000 1
HEZAKUG2 16.000 2
HEZAKUG3 16.000 3
HEZAKUG4 16.000 4
JERTOVG1 10.500 1
JERTOVG2 10.500 2
JERTOVG3 11.000 3
JERTOVG4 11.000 4
KRALJEVG 3.7000 1
KRALJEVG 3.7000 2
KRALJEVG 3.7000 3
KRALJEVG 3.7000 4
TE-TOOG1 10.500 1
HEVARAG1 10.500 1
HEVARAG2 10.500 2
HEDUBRG1 14.400 1
HEDUBRG2 14.400 1
RHEOBRG1 15.750 1
RHEOBRG2 15.750 2
TEPLOMG1 13.800 1
TEPLOMG2 13.800 1
TERIJEG1 20.000
1
TESISAG2 15.750
1
TE-TOG1 10.500
1
TE-TOG2 10.500
2
TE-TOG4 10.500
4
TE-TOG5 10.500
5
TE-TOG3 12.500
3
HEDJALG1 10.500 1
HEDJALG2 10.500 2
KTEOSGT 15.750 1
KTEOSST 10.500
1
KTESISGT 15.750 1
KTESISST 10.500
1
- included in total consumption
- included in total consumption
will not exist in 2010
- included in total consumption
- included in total consumption
2.26
Figure 2.4.11 – GTmax model - Croatia
Figure 2.4.12 – PSS/E model - Croatia
2.27
Macedonia
GTmax
Powerplant
BITOLA 1
BITOLA 2
BITOLA 3
OSLOMEJ
NEGOTINO
VRUTOK
GLOBOCICA
SPILJE
TIKVES
KOZJAK+MAT
RAVEN
VRBEN
Bus#
26301
26302
26303
26311
26321
26331
26332
26333
26334
26341
26342
26351
26352
26353
26361
26362
26363
26364
26371
26401
-
PSS/E
Remarks
Bus Name
ID
BT 2 100 15.750
1
BT 2 400 15.750
1
BT 2 400 15.750
2
OSLOMEJ 13.800 1
TPP NEGO 15.750 1
VRUTOK
12.000 1
VRUTOK
12.000 2
VRUTOK
12.000 3
VRUTOK
12.000 4
GLOBOCIC 10.500 1
GLOBOCIC 10.500 2
SPILJE 10.500
1
SPILJE 10.500
2
SPILJE 10.500
3
TIKVES
10.500
1
TIKVES
10.500
2
TIKVES
10.500
3
TIKVES
10.500
4
KOZJAK 10.500 1
MATKA 2 10.500 1
- included in total consumption
- included in total consumption
Figure 2.4.13 – GTmax model - Macedonia
2.28
Figure 2.4.14 – PSS/E model - Macedonia
2.29
Montenegro
GTmax
PSS/E
Powerplant
Bus#
Bus Name
PLJEVLJA THERMAL 36511 JTPLJEG1 15.750
36521 JHPERUG1 10.500
36522 JHPERUG2 10.500
36523 JHPERUG3 10.500
PERUCICA
36524 JHPERUG4 10.500
36525 JHPERUG5 10.500
36526 JHPERUG6 10.500
36527 JHPERUG7 10.500
SMALL HPPS
-
Remarks
ID
G1
G1
G2
G3
G4
G5
G6
G7
- included in total consumption
Figure 2.4.15 – GTmax model - Montenegro
Figure 2.4.16 – PSS/E model - Montenegro
2.30
Serbia
GTmax
Powerplant
Bus#
35001
35001
35003
35003
35005
35005
35171
35171
DJER1 AND DJER2
35173
35173
35175
35175
35177
35177
35179
35179
35011
KOSTOLAC B
35012
35021
TENT A 1
35022
35023
35024
TENT A 2
35025
35026
35031
TENT B
35032
35041
KOSOVO B THERMAL
35042
35051
35052
BAJINA BASTA
35053
35054
35071
BAJINA
35072
35081
35082
BISTRICA KOKIN
35111
35112
35093
KOSOVO A 3-4
35094
KOSOVO A5
35095
KOSTOLAC A 1
35101
KOSTOLAC A 2
35102
35121
POTPEC
35122
35131
35141
35142
ZVORNIK
35143
35144
35151
35152
KOLUBARA 2
35153
35154
KOLUBARA 1
35155
MORAVA
35161
35181
PIROT
35182
35191
35192
35193
35194
VLASINSKE
35201
35202
35211
35212
34512
35221
GAZIVODE
35222
35241
NOVI SAD CHP
35242
PSS/E
Bus Name
JHDJERG1 15.750
JHDJERG1 15.750
JHDJERG3 15.750
JHDJERG3 15.750
JHDJERG5 15.750
JHDJERG5 15.750
JHDJE2G1 6.3000
JHDJE2G1 6.3000
JHDJE2G3 6.3000
JHDJE2G3 6.3000
JHDJE2G5 6.3000
JHDJE2G5 6.3000
JHDJE2G7 6.3000
JHDJE2G7 6.3000
JHDJE2G9 6.3000
JHDJE2G9 6.3000
JTDRMNG1 22.000
JTDRMNG2 22.000
JTENTAG1 15.750
JTENTAG2 15.750
JTENTAG3 15.000
JTENTAG4 15.000
JTENTAG5 15.000
JTENTAG6 15.000
JTENTBG1 21.000
JTENTBG2 21.000
JTKOSBG1 24.000
JTKOSBG2 24.000
JHBBASG1 15.650
JHBBASG2 15.650
JHBBASG3 15.650
JHBBASG4 15.650
JRHBBAG1 11.000
JRHBBAG2 11.000
JHBISTG1 10.500
JHBISTG2 10.500
JHKBROG1 6.3000
JHKBROG2 6.3000
JTKOSAG3 15.750
JTKOSAG4 15.750
JTKOSAG5 15.750
JTKSTAG1 10.500
JTKSTAG2 15.750
JHPOTPG2 8.8000
JHPOTPG3 8.8000
JHUVACG1 10.500
JHZVORG1 11.000
JHZVORG2 11.000
JHZVORG3 11.000
JHZVORG4 11.000
JTKOLUG1 10.500
JTKOLUG2 10.500
JTKOLUG3 10.500
JTKOLUG4 10.500
JTKOLUG5 10.500
JTMORAG1 13.500
JHPIROG1 10.500
JHPIROG2 10.500
JHVRL1G1 6.3000
JHVRL1G2 6.3000
JHVRL1G3 6.3000
JHVRL1G4 6.3000
JHVRL2G1 6.3000
JHVRL2G2 6.3000
JHVRL3G1 6.3000
JHVRL3G2 6.3000
JHVRL35 110.00
JHGAZIG1 6.3000
JHGAZIG2 6.3000
JTTNSAG1 15.750
JTTNSAG2 15.750
Remarks
ID
G1
G2
G3
G4
G5
G6
G1
G2
G3
G4
G5
G6
G7
G8
G0
G9
G1
G2
G1
G2
G3
G4
G5
G6
G1
G2
G1
G2
G1
G2
G3
G4
G1
G2
G1
G2
G1
G2
G3
G4
G5
G1
G2
G2
G3
G1
G1
G2
G3
G4
G1
G2
G3
G4
G5
G1
G1
G2
G1
G2
G3
G4
G1
G2
G1
G2
2
G1
G2
G1
G2
as negative load
2.31
ZRENJANIN CHP
KOLUBARA B
KOSOVO 2X450
KOSOVO C 2X450
PIVA
35251
35271
35272
34070
34070
36501
36502
36503
JTTZREG1
JTKOLBG1
JTKOLBG2
JTKOSB1
JTKOSB1
JHPIVAG1
JHPIVAG2
JHPIVAG3
15.750
22.000
22.000
400.00
400.00
15.750
15.750
15.750
G1
G1
G2
C1
C2
G1
G2
G3
Figure 2.4.17 – GTmax model - Serbia
Figure 2.4.18 – GTmax model – Serbia UNMIK
2.32
UNMIK
Figure 2.4.19 – PSS/E model – Serbia and UNMIK
2.33
Romania
GTmax
Powerplant
TURCENI 1
TURCENI 2
ROVINARI
RIURENI
BUCURESTI SUD
GOVORA THERMAL
GROZAVESTI CHP
PROGRESU CHP
BUCURESTI VEST
RETEZAT
MARISELU
DEVA 1
GILCEAG
SUGAG
ORADEA
ARAD CHP
RUIENI
ISALNITA 2
PORTILE 1
PORTILE 2
DROBETA
Bus#
29110
29112
29113
29114
29115
29116
29117
29119
29120
29121
29238
29455
29125
29126
29127
29128
29136
29137
29138
29139
29317
29318
29140
29141
29142
29143
29144
29145
29147
29148
29149
29150
29151
29152
29154
29155
29162
29163
29164
29165
29166
29167
29168
29169
29170
29171
29172
29173
29174
29175
29176
29177
29178
29179
29181
29183
29248
29184
29185
29189
29190
29191
29192
29193
29250
29194
29195
29199
29196
29197
29251
PSS/E
Bus Name
ID
TURCENI1 24.000 1
TURCENI3 24.000 1
TURCENI4 24.000 1
TURCENI5 24.000 1
TURCENI 24.000
1
TURCENI6 24.000 1
TURCENI7 24.000 1
ROVIN 5 24.000
1
ROVIN 6 24.000
1
ROVIN 3 24.000
1
ROVIN 4 24.000
1
ROVIN 7 24.000
1
AREFU 1 10.500
1
AREFU 2 10.500
1
AREFU 3 10.500
1
AREFU 4 10.500
1
BUC.S 5 13.800
1
BUC.S 6 13.800
1
BUC.S 3 10.500
1
BUC.S 1 10.500
1
BUC.S 2 10.500
1
BUC.S 4 10.500
1
STUPA I1 10.500
1
STUPA I2 10.500
1
STUPAII6 10.500
1
STUPAII5 10.500
1
STUPA I3 10.500
1
STUPA I4 10.500
1
GROZAV 2 10.500 1
GROZAV 1 10.500 1
PROGRS 1 10.500
1
PROGRS4 10.500
1
PROGRS3 10.500
1
PROGRS 2 10.500
1
BUC.V 2 13.800
1
BUC.V 1 13.800
1
RETEZAT1 15.750 1
RETEZAT2 15.750 1
MARISEL1 15.750 1
MARISEL2 15.750 1
MARISEL3 15.750 1
MINTIA 1 15.750
1
MINTIA 2 15.750
1
MINTIA 5 15.750
1
GALCEAG1 15.750 1
GALCEAG2 15.750 1
SUGAG 1 15.750
1
SUGAG 2 15.750
1
ORAD I 4 10.500
1
ORAD II1 10.500
1
ORAD II2 10.500
1
ORAD I 6 10.500
1
ORAD I 5 10.500
1
ORAD II3 10.500
1
ARAD 1
10.500
1
RUIENI 1 10.500
1
RUIENI 2 10.500
1
ISALNIT7 24.000
1
ISALNIT8 24.000
1
P.D.F 1 15.750
1
P.D.F 2 15.750
1
P.D.F 3 15.750
1
P.D.F 4 15.750
1
P.D.F 5 15.750
1
P.D.F.6 15.750
1
GRUIA12 6.3000
1
GRUIA34 6.3000
1
GRUIA56 6.3000
1
GRUIA78 6.3000
1
DROBETA2 10.500 1
DROBETA4 10.500 1
Remarks
2.34
CRAIOVA
BORZESTI
BORZESTI CHP
STEJARU
SUCEAVA
BACAU THERMAL
IASI II
IASI I
CERNAVODA
BRAILA 1
BRAILA 2
GALATI CHP
PALAS CHP
LOTRU CIUNGET
BRASOV THERMAL
PITESTI
BRAZI CHP
DEVA 2
PAROSENI
BRAILA 3
DOICESTI
CERNAVODA 2
CERNAVODA 3
ARCESTI
BABENI
BACAU
BERESTI
BRADISOR
CALBUCET
CALIMANESTI OLT
CALIMANESTI SIRE
CORNETU
DAESTI
DOICESTI
DRAGANESTI
DRAGASANI
FRUNZARU
GALBENI
GUIRGIU
GOVORA
IONESTI
IPOTESTI
IZBICENI
LUDUS
MOTRU
MUNTENI I
NEHOUI
29252
29253
29198
29200
29201
29202
29203
29205
29206
29207
29208
29209
29210
29211
29212
29214
29215
29216
29217
29218
29219
29220
29221
29224
29225
29310
29226
29227
29232
29233
29234
29235
29236
29237
29305
29239
29267
29268
29266
29259
29260
29262
29269
29270
29271
29263
29299
29301
29302
29332
29470
28639
28678
28147
28320
28564
28574
28752
28674
28624
28676
29261
28849
-
DROBETA3 10.500
DROBETA1 10.500
CRAI II2 15.750
CRAI II1 15.750
BORZEST7 15.750
BORZEST8 15.750
BORZE I6 10.500
BORZE I4 10.500
BORZE I5 10.500
STEJARU5 10.500
STEJARU6 10.500
STEJARU 10.500
SUCEAVA1 10.500
SUCEAVA2 10.500
BACAU 1 10.500
FAI II 1 10.500
FAI II 2 10.500
FAI I 4 10.500
FAI I 3 10.500
CERNAV.1 24.000
BRAILA 1 15.750
BRAILA 2 15.750
BARBOSI5 10.500
BARBOSI3 10.500
SMARDAN6 10.500
SMARDAN4 10.500
PALAS 1 10.500
PALAS 2 10.500
LOTRU 1 15.750
LOTRU 2 15.750
LOTRU 3 15.750
BRASOV 1 10.500
BRASOV 2 10.500
PITEST 4 10.500
PITEST 5 10.500
BRAZI 5 10.500
BRAZI 7 10.500
BRAZI 6 10.500
BRAZI 10 10.500
MINTIA 4 15.750
MINTIA 3 15.750
MINTIA 6 15.750
PAROSEN1 10.500
PAROSEN2 10.500
PAROSEN3 10.500
PAROS 4 18.000
BRAILA 3 15.750
DOICEST8 15.750
DOICEST7 15.750
CERNAV.2 24.000
CERNAV.3 24.000
ARCESTI 110.00
BABENI 110.00
BACAU S 110.00
VERNEST 110.00
BRADISO 110.00
DAIESTI 110.00
CHE DRAG 110.00
CHE DRG 110.00
FRUNZ. 110.00
IONESTI 110.00
IZBICEN 110.00
MUNTEAN 110.00
-
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
1
1
1
1
1
1
1
1
-
as negative load
as negative load
as negative load
as negative load
as negative load
included in total consumption
included in total consumption
included in total consumption
included in total consumption
as negative load
included in total consumption
as negative load
as negative load
as negative load
included in total consumption
included in total consumption
included in total consumption
as negative load
included in total consumption
as negative load
included in total consumption
included in total consumption
as negative load
included in total consumption
2.35
OTHER AUTOPRODUC
RACACIUNI
REMETI
RIMNICU VILCEA
RUSANESTI
SASCIORI
SLATINA HYDRO
STREJESTI
TARNITA
TISMANA
TOTAL UNDER 25
TOTAL UNDER 25 2
TOTAL UNDER 25 3
TOTAL UNDER 25 4
TOTAL UNDER 25 5
TURNU
VADURI
VIDRARU
ZAVIDENI
28850 REMETI 110.00
28597 VILCEL A 110.00
28645 STRAJ
110.00
28844 TURNU
110.00
28102 VADURI 110.00
28675 ZAVID
110.00
1
2
1
1
1
1
included in total consumption
included in total consumption
as negative load
as negative load
included in total consumption
included in total consumption
included in total consumption
as negative load
included in total consumption
included in total consumption
included in total consumption
included in total consumption
included in total consumption
included in total consumption
included in total consumption
as negative load
as negative load
included in total consumption
as negative load
Figure 2.4.20 – GTmax model – Romania
2.36
Figure 2.4.21 – PSS/E model - Romania
2.37
All these production facilities are analyzed as they will be built till 2010. Besides in sensitivity
analyses for average hydrology high load scenarios in 2010 and 2015, additional production
capacities are analyzed. In following tables is shown in what way these units are modeled.
BIH
GTmax
Powerplant
Bus#
14010
14011
14012
14015
14015
14015
BUKBSRB
PSS/E
Bus Name
HEBBG1
15.750
HEBBG2
15.750
HEBBG3
15.750
HESRBG
6.3000
HESRBG
6.3000
HESRBG
6.3000
Remarks
ID
G1
G2
G3
G3
G2
G1
GLAVATICEVO
included in total consumption
Bulgaria
GTmax
Powerplant
BELENE
Bus#
12450
PSS/E
Bus Name
CAREVEC 400.00
ID
1
Remarks
GTmax
Powerplant
KOMARNICA
ANDRIJEVO I ZLAT
KOSTANICA
Bus#
36140
36998
36999
PSS/E
Bus Name
JNIKGR51 110.00
JHANDR21 220.00
JHKOST11 400.00
ID
1
1
1
Bus#
34070
34070
36999
PSS/E
Bus Name
JTKOSB1 400.00
JTKOSB1 400.00
JPRIZ22 220.00
ID
C1
C2
1
Montenegro
Remarks
Serbia (UNMIK)
GTmax
Powerplant
KOSOVO 3X274
KOSOVO C 5X450
ZHUR
Remarks
Exchange programs-desired interchange
Because of different modeling approach of production and intersystem exchanges between GTmax
and PSS/E, there are some peculiarities that need to be implemented in PSS/E model. This causes
differences between exchange programs in GTmax and PSS/E. So, NPP Krsko is installed in system
of Slovenia, but 50 % it is owned by Croatia. In GTmax this is modeled simply as Croatian power
plant that has half power installed. But in PSS/E model, this plant is modeled as it is, in system of
Slovenia, so Croatian part of energy produced is modeled as export from Slovenia to Croatia.
Similar situation is with HPP Dubrovnik in Croatia, which is owned by Bosnia and Herzegovina. In
case of HPP Piva, there is long term contract between Serbia and Montenegro, by which this plant
operates as plant owned by Serbia in exchange for 105 MW exchange.
In order to achieve correspondence between exchange programs in GTmax and PSS/E model,
following assumptions are made:
Albania:
Bulgaria:
BIH:
Croatia:
Macedonia:
Romania:
Serbia:
Montenegro:
PSS/E program
PSS/E program
PSS/E program
PSS/E program
PSS/E program
PSS/E program
PSS/E program
PSS/E program
=
=
=
=
=
=
=
=
GTmax program
GTmax program
GTmax program – production (DUBROVNIK G-2)
GTmax program + production (DUBROVNIK G-2) – 50% production (KRSKO)
GTmax program
GTmax program
GTmax program (Serbia) + GTmax program (UNMIK) – production (PIVA)
GTmax program + production (PIVA)
In sensitivity cases analyses, new generation facilities are modeled. One of them is HPP Buk Bijela
and HPP Srbinje in Bosnia and Herzegovina. 1/3 of the energy produced by this plants is owned by
Montenegro, so in these cases (2015 average hydrology high load scenarios) exchange program of
BIH and Montenegro is calculated as:
BIH:
Montenegro:
PSS/E program =
PSS/E program =
GTmax program – production (DUBROVNIK G-2)+ 1/3production(BUKBSRB)
GTmax program + production (PIVA)-1/3production(BUKBSRB)
2.38
3 ANALYZED GENERATION, DEMAND AND EXCHANGE
SCENARIOS
3.1
This chapter shortly describes analyzed generation, demand and exchange scenarios in WASP and
GTMax and explains how they are modeled in PSS/E. Total number of 10 scenarios were analyzed
from transmission network prospective.
WASP Scenario B results, were discussed and it was decided that WASP Scenario B “Case 1A”
would be used as the Reference Case for GTmax Scenario C, and for PSS/E analyses as Reference
cases. These Reference Cases include medium demand forecast, most likely fuel price forecast for
all fuels and life extension and rehabilitation program as scheduled by the utilities. Within the
extensive GTMax analyses in 2010 and 2015, the distribution of the new generation units to
specific jurisdictions has been done. The specific (named) new generation units have defined sites
and their distribution throughout the region have been done according to defined sites. Distribution
of non-specific new generation units has been solved by taking into account the power balance for
each jurisdiction or their specific needs. On the basis of the WASP results for the Reference Case
and defined distribution of specific and non-specific new generation units, detailed GTMax
simulation of the weekly operation of the regional power system have been done in average, dry
and wet hydrology conditions. The generation of each power plant in peak hour of 2010 and 2015
in average, dry and wet hydrology conditions have been obtained as one of the GTMax results and
have been used as input data for transmission network analyses done using PSS/E software
package. Five scenarios were related to year 2010, and other five scenarios were related to year
2015.
Three scenarios for each year represent base cases and other two extra situations characterized by
high demand and power imports.
3.1. Year 2010 Scenarios
Following tables 3.1.1-3.1.5 include generation and demand data dependent on analyzed
hydrological situations, load level and power imports, related to year 2010. For each country one
row represents GTMax data (white rows) and one row (shaded ones) equivalent data in PSS/E
model according to explanations from Chapter 2. Differences in demand, hydro power plants and
thermal power plants production are caused by plants connected to voltage levels below 110 kV
which are not included into PSS/E model, so their overall production is included reducing demand,
hydro and thermal production. This is the most obvious in Romanian power system characterized
by large number of small hydro power plants connected to low voltage levels so Romanian balance
in PSS/E model is quite different than on GTMax model.
Total power system balances (production minus demand) are different in GTMax and PSS/E
models for Bosnia and Herzegovina (HPP Dubrovnik production is included into Croatian balance
on PSS/E model), Croatia (HPP Dubrovnik is included, NPP Krsko in Slovenia is excluded from
the balance), Montenegro (HPP Piva is included), and Serbia and UNMIK (HPP Piva is excluded
from their balance).
WASP results for the Reference Case show that, for the period 2005-2010, the following new
capacity would be added to the regional power system:
•
Cernavoda nuclear unit #2
•
Kolubara lignite unit #1, and
•
One 500-MW Kosovo lignite plant
and these were included in the PSS/E model for 2010.
3.2
For 2010 year, additional scenarios are analyzed as Sensitivity cases. One of them is the High Load
Forecast with new generation facilities implemented: HPP Zhur connected to the UNMIK node and
there are also one 300 MW and one 500 MW combined cycle plant in Croatia and additional 500
MW unit connected to UNMIK node. This High Demand Forecast Case includes high demand
forecast, most likely fuel price forecast for all fuels and life extension and rehabilitation program as
scheduled by the utilities. On the basis of the WASP results for the High Demand Forecast Case
and defined distribution of specific and non-specific new generation units, detailed GTMax
simulation of the weekly operation of the regional power system have been done in average
hydrology condition. The generations of each power plant in peak hour of 2010 and 2015 in average
hydrology condition have been obtained as one of the GTMax results and have been used as input
data for transmission network analyses done using PSS/E software package.
As second Sensitivity case, additional energy exchanges in the region are analyzed. This
Import/Export Case includes medium demand forecast, most likely fuel price forecast for all fuels,
life extension and rehabilitation program as scheduled by the utilities and net import of 1,500MW
into the region. Following additional exchanges are simulated:
- Import 750 MW from UCTE.
- Import 500 MW from Turkey.
- Export 500 MW to Greece.
- Import 750 MW from Ukraine.
On the basis of the WASP results for the Import/Export Case and defined distribution of specific
and non-specific new generation units, detailed GTMax simulation of the weekly operation of the
regional power system have been done in average hydrology condition. The generations of each
power plant in peak hour of 2010 and 2015 in average hydrology condition have been obtained as
one of the GTMax results and have been used as input data for transmission network analyses done
using PSS/E software package.
After comparison of the GTMax results for this case with the results for Base Case it can be seen
that there is no new TPPs connected to UNMIK node or Mladost node in 2010 (without new TPPs
on Kosovo and without Kolubara B units).
3.3
Table 3.1.1: Demand, Generation and Exchanges in SE Europe for base case - average hydrological scenario in
2010
Country
Demand HPP Generation
TPP and NPP
Total Generation
(MW)
(MW)*
Generation (MW)
(MW)
1338
757
140
897
Albania
1338
757
140
897
Bosnia and
2077
1591
826
2417
Herzegovina
2029
1439
826
2265
6193
554
6426
6980
Bulgaria
6113
474
6426
6900
3217
1018
749
1767
Croatia
3186
1092
411
1503
1229
232
730
962
Macedonia
1218
220
730
950
687
228
0
228
Montenegro
687
540
0
540
7797
2996
5730
8726
Romania
7022
2256
5696
7952
Sebia and
7112
2582
5090
7672
UNMIK
7112
2270
5090
7360
29649
9958
19691
29649
Region Total
28705
9048
19319
28367
* pumped storage HPP’s included
** half of NPP Krsko (Slovenia) production (scheduled for Croatian power system)
Surplus(+)/Deficit(-)
-441
-441
341
236
787
787
-1450
-1683
-268
-268
-459
-147
930
930
560
248
0
-338**
Table 3.1.2: Demand, Generation and Exchanges in SE Europe for base case - dry hydrological scenario in 2010
Country
Demand HPP Generation
TPP and NPP
Total Generation
(MW)
(MW)*
Generation (MW)
(MW)
1338
753
200
953
Albania
1338
753
200
953
Bosnia and
2077
1636
1338
2974
Herzegovina
2050
1504
1338
2843
6193
373
6426
6799
Bulgaria
6103
283
6426
6709
3217
924
1632
2555
Croatia
3189
1001
1294
2295
1229
296
730
1026
Macedonia
1224
290
730
1020
687
179
0
179
Montenegro
687
467
0
467
7797
1820
5776
7596
Romania
7128
1185
5742
6927
Sebia and
7112
2476
5090
7566
UNMIK
7112
2188
5090
7278
29649
8457
21192
29649
Region Total
28830
7672
20820
28492
* pumped storage HPP’s included
** half of NPP Krsko (Slovenia) production (scheduled for Croatian power system)
Surplus(+)/Deficit(-)
-384
-384
898
793
606
606
-661
-894
-203
-203
-508
-220
-200
-200
454
166
0
-338*
3.4
Table 3.1.3: Demand, Generation and Exchanges in SE Europe for base case - wet hydrological scenario in 2010
Country
Demand HPP Generation
TPP and NPP
Total Generation
(MW)
(MW)*
Generation (MW)
(MW)
1338
759
140
899
Albania
1338
759
140
899
Bosnia and
2077
1630
570
2200
Herzegovina
2019
1468
570
2038
6193
686
6346
7032
Bulgaria
6103
596
6346
6942
3217
1249
749
1998
Croatia
3187
1324
411
1735
1229
367
730
1097
Macedonia
1214
352
730
1082
687
244
0
244
Montenegro
687
586
0
586
7797
3744
4684
8428
Romania
6706
2653
4684
7337
Sebia and
7112
2662
5090
7751
UNMIK
7112
2319
5090
7409
29649
11341
18309
29649
Region Total
28366
10057
17971
28028
* pumped storage HPP’s included
** half of NPP Krsko (Slovenia) production (scheduled for Croatian power system)
Surplus(+)/Deficit(-)
-439
-439
124
19
839
839
-1219
-1452
-132
-132
-443
-101
632
632
639
297
0
-338**
Table 3.1.4: Demand, Generation and Exchanges in SE Europe for sensitivity case – high load – average
hydrological scenario in 2010
Country
Demand HPP Generation
TPP and NPP
Total Generation
(MW)
(MW)*
Generation (MW)
(MW)
1414
790
140
930
Albania
1414
790
140
930
Bosnia and
2114
1693
667
2360
Herzegovina
2061
1535
667
2202
6492
518
6832
7350
Bulgaria
6425
450
6832
7282
3371
1001
1507
2508
Croatia
3350
1085
1169
2254
1262
271
730
1001
Macedonia
1252
261
730
991
704
228
0
228
Montenegro
704
540
0
540
8320
3237
5026
8263
Romania
7533
2484
4992
7476
Sebia and
7346
2774
5608
8382
UNMIK
7346
2462
5608
8070
31022
10512
20510
31022
Region Total
30085
9379
20138
29747
* pumped storage HPP’s included
** half of NPP Krsko (Slovenia) production (scheduled for Croatian power system)
Surplus(+)/Deficit(-)
-484
-484
246
141
858
858
-863
-1096
-261
-261
-476
-163
-57
-57
1036
724
0
-338**
3.5
Table 3.1.5: Demand, Generation and Exchanges in SE Europe for sensitivity case – power import – average
hydrological scenario in 2010
Country
Demand HPP Generation
TPP and NPP
Total Generation Surplus(+)/Deficit(-)
(MW)
(MW)*
Generation (MW)
(MW)
1338
757
140
897
-440
Albania
1338
757
140
897
-440
Bosnia and
2077
1734
826
2560
484
Herzegovina
2019
1572
826
2398
379
6193
561
6346
6907
714
Bulgaria
6113
481
6346
6827
714
3217
1218
749
1967
-1250
Croatia
3188
1294
411
1705
-1483
1229
266
730
996
-233
Macedonia
1199
235
730
965
-233
687
228
0
228
-459
Montenegro
687
540
0
540
-147
7797
2936
4756
7692
-105
Romania
6977
2150
4722
6872
-105
Sebia and
7112
2582
4320
6902
-210
UNMIK
7112
2270
4320
6590
-522
29649
10282
17867
28149
-1500
Region Total
28633
8788
17495
26795
-1838**
* pumped storage HPP’s included
** 1500 MW of import plus half of NPP Krsko (Slovenia) production (scheduled for Croatian power system)
3.2. Year 2015 Scenarios
Following tables 3.2.1-3.2.5 include generation and demand data dependent on analyzed
hydrological situations, load level and power imports, related to year 2015, using the same
assumptions explained in previous subchapter.
WASP results for the Reference Case show that, for the period 2011-2015, the following new
capacity would be added to the regional power system:
•
Cernavoda nuclear unit #3
•
Kolubara lignite unit #2
•
One 300-MW and two 500-MW Kosovo lignite plants
•
Two 100-MW CHP plants, and
•
Two 300-MW and one 500-MW combined cycle plants
On the basis of this kind of analyses the following distribution of new non-specific generation units
have been arranged:
two 100 MW CHP units have been placed in Romania
two 300 MW and one 500 MW combined cycle units have been placed in Croatia
all these are included in PSS/E model for 2015.
Like for 2010, and for 2015 year, additional scenarios are analyzed as Sensitivity cases. One of
them is the High Load Forecast with new generation facilities implemented:
- HPP Buk Bijela with HPP Srbinje in B&H. Within the results from GTMax it can be seen that
this hydro system is partially connected to Sarajevo node in B&H and partially to Montenegro
node, due to the partial ownership of this hydro system.
- HPP Glavaticevo connected to Mostar node in B&H.
- HPP Dabar connected to Trebinje node in B&H.
- HPP Komarnica connected to Montenegro node.
3.6
- HPP Kostanica connected to Montenegro node.
- HPP Andrijevo and Zlatica connected to Montenegro node.
- NPP Belene nuclear plant connected to Varna node in Bulgaria,
- one 100 MW CHP unit connected to Mladost node in Serbia,
- one 500 MW combined cycle plant connected to node Zagreb in Croatia
- and additional two 300 MW and two 500 MW units connected to UNMIK node.
After comparison of the GTMax results for Export/Import Case (exchanges in/from/to SE Europe
are the same as in 2010) with the results for Base Case in 2015 it can be seen that there is no
Chernavoda Unit 3, no CHP unit connected to SIBIU node, no second 300 MW unit connected to
Zagreb node, only one 500 MW unit connected to UNMIK node, but instead of one now there are
two 300 MW units connected to UNMIK node.
As second Sensitivity case for 2015, additional energy exchanges in the region are analyzed. This
was called Import/Export Case, and following additional exchanges are simulated (same as for
2010):
- Import 750 MW from UCTE.
- Import 500 MW from Turkey.
- Export 500 MW to Greece.
- Import 750 MW from Ukraine.
Table 3.2.1: Demand, Generation and Exchanges in SE Europe for base case - average hydrological scenario in
2015
Country
Demand HPP Generation
TPP and NPP
Total Generation
(MW)
(MW)*
Generation (MW)
(MW)
1614
978
140
1118
Albania
1614
978
140
1118
Bosnia and
2410
1648
826
2474
Herzegovina
2358
1491
826
2317
6688
533
6854
7387
Bulgaria
6619
465
6854
7319
3752
986
1595
2581
Croatia
3721
1060
1257
2317
1438
379
720
1099
Macedonia
1427
367
720
1087
694
230
191
421
Montenegro
694
542
191
733
9056
3424
5680
9104
Romania
7973
2547
5474
8021
Sebia and
7499
2584
6383
8967
UNMIK
7499
2272
6383
8655
33151
10762
22389
33151
Region Total
31906
9722
21845
31568
* pumped storage HPP’s included
** half of NPP Krsko (Slovenia) production (scheduled for Croatian power system)
Surplus(+)/Deficit(-)
-496
-496
64
-41
699
699
-1171
-1404
-340
-340
-273
39
48
48
1468
1156
0
-338**
3.7
Table 3.2.2: Demand, Generation and Exchanges in SE Europe for base case - dry hydrological scenario in 2015
Country
Demand HPP Generation
TPP and NPP
Total Generation
(MW)
(MW)*
Generation (MW)
(MW)
1614
751
143
894
Albania
1614
751
143
894
Bosnia and
2410
1656
1617
3274
Herzegovina
2382
1523
1617
3141
6688
600
6854
7454
Bulgaria
6598
510
6854
7364
3752
1177
1721
2898
Croatia
3723
1253
1383
2636
1438
274
720
994
Macedonia
1429
265
720
985
694
107
191
298
Montenegro
694
395
191
586
9056
2033
6654
8687
Romania
8188
1371
6448
7819
Sebia and
7499
2268
6383
8651
UNMIK
7499
1980
6383
8363
33151
8868
24283
33151
Region Total
32127
8049
23739
31789
* pumped storage HPP’s included
** half of NPP Krsko (Slovenia) production (scheduled for Croatian power system)
Surplus(+)/Deficit(-)
-720
-720
864
759
766
766
-853
-1086
-444
-444
-396
-108
-369
-369
1152
864
0
-338**
Table 3.2.3: Demand, Generation and Exchanges in SE Europe for base case - wet hydrological scenario in 2015
Country
Demand HPP Generation
TPP and NPP
Total Generation
(MW)
(MW)*
Generation (MW)
(MW)
1614
984
140
1124
Albania
1614
984
140
1124
Bosnia and
2410
1743
573
2316
Herzegovina
2353
1581
573
2154
6688
899
6744
7643
Bulgaria
6598
809
6744
7553
3752
1468
1595
3063
Croatia
3721
1543
1257
2800
1438
358
523
881
Macedonia
1427
347
523
870
694
246
191
437
Montenegro
694
588
191
779
9056
3445
5488
8933
Romania
8074
2635
5316
7951
Sebia and
7499
2662
6092
8754
UNMIK
7499
2320
6092
8412
33151
11805
21346
33151
Region Total
31980
10806
20836
31642
* pumped storage HPP’s included
** half of NPP Krsko (Slovenia) production (scheduled for Croatian power system)
Surplus(+)/Deficit(-)
-490
-490
-94
-199
955
955
-688
-921
-557
-557
-257
85
-123
-123
1255
913
0
-338**
3.8
Table 3.2.4: Demand, Generation and Exchanges in SE Europe for sensitivity case – high load – average
hydrological scenario in 2015
Country
Demand HPP Generation
TPP and NPP
Total Generation
(MW)
(MW)*
Generation (MW)
(MW)
1781
758
140
898
Albania
1781
758
140
898
Bosnia and
2503
1989
630
2619
Herzegovina
2433
1814
630
2444
7327
552
7674
8226
Bulgaria
7259
484
7674
8158
4067
1007
1845
2852
Croatia
4040
1085
1507
2592
1516
179
720
899
Macedonia
1513
176
720
896
736
922
191
1113
Montenegro
736
1234
191
1425
10232
3584
5608
9192
Romania
9175
2698
5436
8135
Sebia and
8025
2472
7917
10389
UNMIK
8025
2160
7917
10077
36188
11463
24725
36188
Region Total
34962
10408
24215
34624
* pumped storage HPP’s included
** half of NPP Krsko (Slovenia) production (scheduled for Croatian power system)
Surplus(+)/Deficit(-)
-883
-883
116
11
899
899
-1215
-1448
-617
-617
377
689
-1040
-1040
2364
2052
0
-338**
Table 3.2.5: Demand, Generation and Exchanges in SE Europe for sensitivity case – power import – average
hydrological scenario in 2015
Country
Demand HPP Generation
TPP and NPP
Total Generation Surplus(+)/Deficit(-)
(MW)
(MW)*
Generation (MW)
(MW)
1614
886
140
1026
-588
Albania
1614
886
140
1026
-588
Bosnia and
2410
1720
826
2546
136
Herzegovina
2364
1568
826
2395
31
6688
840
6854
7694
1006
Bulgaria
6578
730
6854
7584
1006
3752
1133
1307
2440
-1312
Croatia
3721
1207
969
2176
-1545
1438
379
720
1099
-340
Macedonia
1417
357
720
1077
-340
694
230
191
421
-273
Montenegro
694
542
191
733
39
9056
3030
5054
8084
-972
Romania
8161
2256
4934
7190
-972
Sebia and
7499
2584
5757
8341
842
UNMIK
7499
2272
5757
8029
530
33151
10802
20849
31651
-1500
Region Total
32048
9819
20391
30210
-1838**
* pumped storage HPP’s included
** import of 1500 MW plus half of NPP Krsko (Slovenia) production (scheduled for Croatian power system)
3.9
4 LOAD FLOW AND CONTINGENCY ANALYSIS – REFERENCE
CASES
4.1
Introduction
In this chapter load-flow and security (n-1) analysis for reference cases defined in Chapter 2 is
described. The analyzed network models are:
Year
2010
Hydrology
average
dry
wet
average
2015
dry
wet
Topology
2010
2010
2015
2010
2015
2010
2015
The load-flow analysis includes line loading and voltage profile analysis, analysis of losses and also
analysis of power flows through interconnection lines.
The system reliability and adequacy is checked using “n-1” contingency criterion. List of
contingencies includes:
• all interconnection lines;
• all 400 and 220 kV lines in analyzed region, except lines which outage cause “island”
operation (in case of parallel and double circuit lines, outage of one line is considered);
• all transformers 400/x kV in analyzed region (in case of parallel transformers, outage of one
transformer is considered).
Current thermal limits are used as rated limits of lines and transformers, as described in Chapter 2.
Voltage limits are defined in Chapter 2, also.
Every branch with current above its thermal limit is treated as overloaded. States with overloaded
branches and/or voltages below or above defined voltage limits are treated as "insecure".
4.1 Scenario 2010 – average hydrology – 2010 topology
This part of the Study presents results of static load flow and voltage profile analyses that are
conducted for complete network topology and (n-1) contingencies in the scenario which is denoted
as Scenario 2010 average hydrology.
4.1.1 Line loadings
Area totals and power exchanges for the 2010-base case-average hydrology scenario are shown in
Figure 4.1.1 and
Table 4.1.1. Power flows along regional interconnection lines and system balances are shown in
Figure 4.1.2. Power flows along interconnection lines are also given in Table 4.1.2, while Figure
4.1.3 shows histogram of tie lines loadings.
4.2
SVK
91
54
1
AUT
1
4 50
UKR
27
HUN
53
0
94
36
6
32
9
30
SLO
ITA
CRO
ROM
333
25
0
4
67
60
SRB
330
BIH
27
2
8
10
3 76
MNT
40
18
10
BUL
4
2
0
133
ALB
272
5
17
MKD
TUR
24
3
41
GRE
2 50
Figure 4.1.1 - Area exchanges in analyzed electric power systems for 2010-base case-average hydrology scenario
Table 4.1.1 - Area totals in analyzed electric power systems for 2010-base case-average hydrology scenario
Country
Albania
Bulgaria
Bosnia and Herzegovina
Croatia
Macedonia
Romania
Serbia and UNMIK
Montenegro
TOTAL - SE EUROPE
Generation
(MW)
896.7
6900.4
2266.1
1502.9
950.3
7939.9
7355.9
539.8
28351.9
Load
(MW)
1287.3
5977.3
1971.3
3136.7
1198.2
6728.3
6873.1
669.2
27841.4
Bus Shunt
(Mvar)
0
0
0
0
0
0
0
0.6
0.6
Line Shunt
(Mvar)
0
14.4
0
0
0
80.0
13.6
1.6
109.5
Losses
(MW)
50.4
121.6
58.6
49.0
20.1
201.2
220.8
15.5
737.1
Net Interchange
(MW)
-441.0
787.0
236.2
-1682.8
-268.0
930.4
248.4
-147.0
-336.7
4.3
LEVIC 1400.0
CENTREL
200
143
251
116
GABICK1400.0
450.0 MW
MSAJO 4
410.0
167
52.9
JSOMB31
396.6
332
71.3
407.8
224.4
UMUKACH
412.0
90.
3.1 7
282
NADAB
74.5
404.4
404.1
247
13.0
249 ARAD
59.7
402.9
ROM
198
98.7
ISACCEA 414.3
930.0 MW
JSUBO31 398.8
334
JSMIT21
103
403.2
JHDJE11
250
42.3
250
41.6
P.D.FIE
TANTAREN
404.3
276
39.2
415.6
UGLJEVIK 405.1
SCG
233.7
147
16.6
JHPIVA21
148 236.1
14.3
34.1
18.8
JHPERU21
33.9 229.6
26.7
18.6
76.9
18.7
JPODG211
120
JPODG121228.5
CRG
-147.0 MW
787.0 MW
416.4
412.4
AZEMLA1
411
135
KARDIA K
417 416.8
112
QES/H
68.8
28.2
AHS_FLWR417.6
K
413.2
415.6
9
23 6.0
7
23.7
22.6
K-KEXRU
417.9
4HAMITAX
11.0
10.9 415.6
80.1
23.7 68.8
4BABAESK
71.6
415.7
0.0 MW
-0.1 MW
KARACQOU
250 419.1
50.0
FILIPPOI
1
7. 2.7
2
.6
12 5.9
9
7
63. 6
25.
.8
31 .9
45
LAGAD K 414.1
398.2
MI_3_4_1 420.4
BLAGOEV414.5
412.4
68.7
33.2
7
63. 1
48.
ALB
-441.0 MW
413.8
5
17 .0
38
STIP 1
DUBROVO
386.0
6
17 7
.
48
408.3
24
38. 1
9
BITOLA 2
C_MOGILA
SK 4
407.3
31.9
49.4
26.0
21.7
6.9
29.4
SK 1 223.3
.0 9
11 1.
1
27.3
167
BUL
JLESK21
386.9
1
37 0.0
.2
7.0
15.3
27.7
128
395.4
AVDEJA2
227.4AFIERZ2228.4
MKD
KV: 110 , 220 , 400
DOBRUD4 421.3
415.8
377 SOFIA_W4
108
411.9
JVRAN31
0.000
-268.0 MW
AKASHA1
374
138
26.0
28.6
6.8 7
18.
6.8 7
18.
24.6
48.5
AEC_400
JNIS2 1
395.7
JPRIZ22
223.0 JTKOSA2
226.9JTKOSB1406.9
1
45 0.
.8 3
397.2
101.4 MW
SRB
248.4 MW
6.9
29.4
RP TREB
233.1
JVARDI22
53.2 230.5
61.0
7.0
5.8
SA 20
232.2
53.0
59.2
24.7
20.6
VISEGRA
232.4
AVDEJA1
GIS REGIONAL MODEL - AVERAGE HYDRO
2010
ROSIORI 410.3
9
12 1
6.
408.6
128 MO-4
16.2
234.8
BIH
236.2 MW
281
63.8
9
12 1
6.
MO-4
228.5
TE TUZL
105
4
68.
125
14.3
291
46.9
GRADACAC
226.0
RP TREB
404.1
SHAW POWER
TECHNOLOGIES
INC. R
UKR
450.0 MW
130
42.5
225.8
4
10 9
.
36
PRIJED2
290
0.6
3
58. 7
18.
.7
51 .5
28
51.5
19.5
CRO
-1682.8 MW
408.4
166
57.2
MSAFA 4
404.2
HE ZAKUC
233.3
10.6
29.3
.1
32 10
1
68 278
.1
51.2
12.5
51.4
20.8
226.6
103
39.6
DAKOVO
227.2MEDURIC
58.0
23.4
404.2
MRACLIN
105
12.4
ERNESTIN
.9 6
44 6.
1
TUMBRI
PEHLIN 233.7
409.8
77.
4
4
77 6.8
46 .4
.8
5
5
10 .5 10 .5
62
62
ZERJAVIN
406.7
ZERJAVIN 230.4
6
17 3
.
29
6
17 3
.
29
32.7
19.3
228.9
77
.
2. 5
3
77
.
2. 5
3
LCIRKO2
403.2
407.3
KONJSKO
10.7
7.7
MBEKO 4
406.3
409.5
MPECS 4
LKRSKO1
MELINA
35.6
UMUKACH2
27.8
32.0
135
SLO
32
29 .6
.0
35.4
21.0
106
42.7
106
42.7
215.0 MW
66
403.0 92 .6
.0
LDIVAC2
230.8
MHEVI 4
45.2
25.4
28.0
49.0
14.9 229.0
28.6
LMARIB1 403.9
66.7
68.2
231.3
28.0
45.4
159
41.5
15.0
14.5
MTLOK 2
230.5
HUN
176
53.8
IPDRV121
159
62.7
42.7
23.2
MKISV 2
226.3
-1249.4 MW
176
53.8
401.4
MGYOR 2
43.1
7.5
58.
22. 4
2
58.
22. 4
2
163
36.
3
IRDPV111
407.8
9
90. 7
59.
LPODLO2229.9
MSAJI 408.2
4
199
40.8
58.4
53.4
58.4
53.4
234.3
ONEUSI2
233.5
LDIVAC1
MGOD
OWIEN 2
350
UZUKRA01
299
772.5
11.0
68.8
OKAINA1 402.6
166
41.0
OOBERS2238.7
346
803
130
42.5
69.6
21.6
405.6
199
97.3
UCTE
221.6 MW
MAISA 7
715.1
MGYOR 4
0
25 .1
69.5
94
104
403.5
OWIEN 1
GRE
TUR
BUS -VOLTAGE(KV)
BRANCH -MW/MVAR
EQUIPMENT -MW/MVAR
Figure 4.1.2 - Power flows along interconnection lines in the region for 2010 - base case - average hydrology scenario
4.4
Table 4.1.2 - Power flows along regional interconnection lines for 2010 - base case-average hydrology scenario
Power Flow
Interconnection line
OHL 400 kV
OHL 220 kV
OHL 220 kV
OHL 400 kV
OHL 400 kV
OHL 400 kV
OHL 400 kV
OHL 400 kV
OHL 220 kV
OHL 220 kV
OHL 220 kV
OHL 220 kV
OHL 220 kV
OHL 220 kV
OHL 220 kV
OHL 220 kV
OHL 220 kV
OHL 400 kV
OHL 400 kV
OHL 400 kV
OHL 400 kV
OHL 400 kV
OHL 400 kV
OHL 2x400 kV ckt.1
OHL 2x400 kV ckt.2
OHL 400 kV
OHL 2x400 kV ckt.1
OHL 2x400 kV ckt.2
OHL 2x400 kV ckt.1
OHL 2x400 kV ckt.2
OHL 2x400 kV ckt.1
OHL 2x400 kV ckt.2
OHL 400 kV
OHL 400 kV
OHL 220 kV
OHL 220 kV
OHL 400 kV
OHL 400 kV
OHL 400 kV
OHL 2x220 kV ckt.1
OHL 2x220 kV ckt.2
OHL 400 kV
OHL 400 kV OHL 400 kV
OHL 400 kV
OHL 400 kV
OHL 400 kV
OHL 220 kV
OHL 220 kV
Zemlak (ALB)
Fierze (ALB)
V.Dejes (ALB)
V.Dejes (ALB)
Ugljevik (B&H)
Mostar (B&H)
Ugljevik (B&H)
Trebinje (B&H)
Trebinje (B&H)
Prijedor (B&H)
Prijedor (B&H)
Gradacac (B&H)
Tuzla (B&H)
Mostar (B&H)
Visegrad (B&H)
Sarajevo 20 (B&H)
Trebinje (B&H)
Blagoevgrad (BUL)
M.East 3 (BUL)
M.East 3 (BUL)
M.East 3 (BUL)
C.Mogila (BUL)
Dobrudja (BUL)
Kozloduy (BUL)
Kozloduy (BUL)
Sofia West (BUL)
Zerjavinec (CRO)
Zerjavinec (CRO)
Ernestinovo (CRO)
Ernestinovo (CRO)
Tumbri (CRO)
Tumbri (CRO)
Melina (CRO)
Ernestinovo (CRO)
Zerjavinec (CRO)
Pehlin (CRO)
Dubrovo (MCD)
Bitola (MCD)
Skopje (MCD)
Skopje (MCD)
Skopje (MCD)
Arad (ROM)
Nadab (ROM)
Rosiori (ROM)
Portile De Fier (ROM)
Subotica (SER)
Ribarevine (MON)
Pljevlja (MON)
Pljevlja (MON)
Kardia (GRE)
Prizren (SER)
Podgorica (MON)
Podgorica (MON)
Ernestinovo (CRO)
Konjsko (CRO)
S. Mitrovica (SER)
Podgorica (MON)
Plat (CRO)
Mraclin (CRO)
Medjuric (CRO)
Djakovo (CRO)
Djakovo (CRO)
Zakucac (CRO)
Vardiste (SER)
Piva (MON)
Perucica (MON)
Thessaloniki (GRE)
Filippi (GRE)
Babaeski (TUR)
Hamitabat (TUR)
Stip (MCD)
Isaccea (ROM)
Tantarena (ROM)
Tantarena (ROM)
Nis (SER)
Heviz (HUN)
Heviz (HUN)
Pecs (HUN)
Pecs (HUN)
Krsko (SLO)
Krsko (SLO)
Divaca (SLO)
S.Mitrovica (SER)
Cirkovce (SLO)
Divaca (SLO)
Thessaloniki (GRE)
Florina (GRE)
Kosovo B (UNMIK)
Kosovo A (UNMIK)
Kosovo A (UNMIK)
Sandorfalva (HUN)
Bekescaba (HUN)
Mukacevo (UKR)
Djerdap (SER)
Sandorfalva (HUN)
Kosovo B (UNMIK)
Bajina Basta (SER)
Pozega (SER)
MW
-410.7
-10.0
7.0
-24.6
105.4
291.3
-276.4
-18.6
-104.8
51.7
51.5
58.3
103.9
127.6
-53.0
-147.1
34.1
31.9
240.9
11.0
12.7
175.7
199.5
-129.5
-129.5
377.1
-105.4
-105.4
-77.4
-77.4
-176.0
-176.0
66.8
-332.1
-44.9
32.7
-68.7
-63.7
26.0
6.9
6.9
248.6
281.9
90.9
249.7
-32.0
-296.0
-46.0
34.9
Mvar
-135.0
37.2
-15.3
-48.5
-68.4
-46.9
39.2
76.9
40.0
-28.5
-19.5
18.7
36.9
-16.2
59.2
-16.6
18.8
-49.4
-38.9
-11.9
-7.2
-48.7
-40.8
-6.1
-6.1
108.2
-62.5
-62.5
-46.8
-46.8
29.3
29.3
53.5
71.3
16.6
19.3
-33.2
-48.1
-21.7
-29.4
-29.4
-59.7
-74.5
-59.7
41.6
-135.1
-15.0
10.7
35.3
% of thermal
rating
32
15
6
4
10
22
21
9
36
18
17
21
36
33
26
38
14
8
35
5
5
25
15
10
10
55
9
9
7
7
16
16
10
25
16
13
6
6
3
9
9
21
24
9
19
10
22
17
19
4.5
Figure 4.1.3 shows that the tie lines in the region are mostly loaded less than 25% of their thermal
limits for the analyzed hydrological base case scenario in year 2010. Among total number of forty
nine 400 kV and 220 kV interconnection lines in the region only seven are loaded between 25% and
50% of their thermal ratings. Only one line (OHL 400 kV Sofia – Nis between Bulgaria and Serbia)
is loaded more than 50% of its thermal rating, which is set at lower value (692.8 MVA) on the
Bulgarian side compared to the line rating on the Serbian side (1330.2 MVA).
45
40
Frequency
35
30
25
20
15
10
5
0
x<25
25<x<50
50<x<75
75<x<100
x>100
Bin
Figure 4.1.3 - Histogram of interconnection lines loadings for 2010-base case-average hydrology scenario ("Frequency"
denotes number of lines and "Bin" denotes loading range in % of thermal limit)
Table 4.1.3 lists all network elements loaded over 80% of their thermal limits. As it can be seen
from this output list, most of the elements loaded over 80% are transformers in some substations
and internal 110 kV and 220 kV lines. Thus, certain internal network reinforcements are necessary
to sustain given load-demand level and production pattern.
Two 220/110 kV transformers in the Fierze substation in Albania are slightly overloaded in this
scenario, while the third transformer in the same substation is loaded near permitted values. It is
expected that the transformers in that substation will be replaced with larger transformer units.
There are two 110 kV internal lines in Romania which are loaded over 80% of their thermal limits,
but none of them is overloaded. These lines are related to the Bojuren and Domnesti nodes.
Two 220 kV lines and thirteen 110 kV lines in the Serbian power system are highly loaded when all
branches are available in the analyzed scenario. Highly loaded 220 kV lines are connected to the
Obrenovac substation, while 110 kV lines are located mostly in the area of Belgrade. Four 110 kV
lines are overloaded, ranging between 108% Ithermal and 118% Ithermal.
Power systems of Bulgaria, Bosnia and Herzegovina, Croatia, Macedonia and Montenegro do not
have or have very few highly loaded branches in 110 kV networks.
Figure 4.1.4 shows histogram of 400 kV and 220 kV regional internal lines and 400/x kV and 220/x
kV transformers loadings. 47% of observed branches are loaded below 25% of their thermal ratings,
34% are loaded between 25% and 50%, 16% are loaded between 50% and 75% and only 2% of
observed branches are loaded between 75% and 100% of their thermal ratings. Two branches
(transformers 220/110 kV Fierze in Albania, 102% - 106% Sn) are overloaded if all branches are in
operation for the analyzed scenario.
4.6
Table 4.1.3 - Network elements loaded over 80% of thermal limits for 2010-base case-average hydrology scenario
AREA
ALB
BIH
ROM
ELEMENT
LOADING
MVA
RATING
MVA
PERCENT
116.8
95.7
91.4
254.8
385.7
173.1
120.0
90.0
90.0
300
400.0
200.0
93.7
106.4
101.5
84.9
96.4
86.6
Transformers
TR 220/110 kV AFIER 2-AFIER 5 ckt.1
TR 220/110 kV AFIER 2-AFIER 5 ckt.2
TR 220/110 kV AFIER 2-AFIER 5 ckt.3
TR 400/110 kV UGLJEV 1
TR 400/220 kV MINTIA-MINTIA B
TR 220/110 kV FUNDENI-FUNDE2B
400
350
Frequency
300
250
200
150
100
50
0
x<25
25<x<50
50<x<75
75<x<100
x>100
Bin
Figure 4.1.4 - Histogram of 400 kV and 220 kV regional lines loadings for 2010-base case-average hydrology scenario
("Frequency" denotes number of lines and "Bin" denotes loading range in % of thermal limit)
4.1.2 Voltage Profile in the Region
Voltage profile in the region within this scenario which is defined by given generation and demand
pattern is seen as satisfactory despite several appearances of certain bus voltage deviations. The
deviations are shown in Table 4.1.4, which includes only 400 kV and 220 kV network buses.
Table 4.1.4 - Bus voltage deviations for 2010-base case-average hydrology scenario, complete network
Country
Node
ALBANIA
BOSNIA AND HERZEGOVINA
400 kV VARNA4
400 kV BURGAS
400 kV MARITSA EAST2
400 kV TECMIG5
400 kV TECMIG7
400 kV DOBRUD4
400 kV MARITSA EAST 3_4_1
400 kV TECMIG6
220 kV SESTRIMO
220 kV BPC_220
220 kV MIZIA2
220 kV AEC_220
220 kV TECVARNA
-
BULGARIA
CROATIA
MACEDONIA
MONTENEGRO
ROMANIA
SERBIA AND UNMIK
Voltages
pu
1.054
1.051
1.055
1.051
1.051
1.053
1.051
1.051
1.104
1.102
1.101
1.103
1.104
-
kV
421.4
420.5
422.1
420.5
420.5
421.3
420.5
420.5
242.8
242.5
242.1
242.6
242.9
-
4.7
Bus voltage magnitudes below permitted limits are not found in the analyzed scenario. Bus voltage
magnitudes that are found above permitted limits (110% Vnominal in 110 kV and 220 kV networks
and 105% Vnominal in 400 kV network) are detected only in Bulgaria. There are eight 400 kV buses
and five 220 kV buses with voltages slightly above permitted limits. Figure 4.1.5 shows histogram
of voltages in monitored 400 kV and 220 kV substations.
120
Frequency
100
80
60
40
20
x>1.1
1.05<x<1.1
1.025<x<1.05
1<x<1.025
0.975<x<1
0.95<x<0.975
0.9<x<0.95
x<0.9
0
Bin
Figure 4.1.5 - Histogram of voltages in monitored substations for 2010-base case-average hydrology scenario
("Frequency" denotes number of busses and "Bin" denotes voltage range in p.u.)
It should be emphasized that these results represent only a situation when additional devices
(transformer automatic tap changers, switched shunts, etc.) are not used for voltage regulation.
Impacts of such devices, which exist in many points of the SEE regional transmission network,
need more comprehensive and thorough analysis.
4.1.3 Security (n-1) analysis
Results of security (n-1) analysis for the 2010-base case-average hydrology scenario are presented
in Table 4.1.5 and Table 4.1.6.
Insecure states for given generation and demand pattern are detected mostly in the power systems of
Romania and Serbia, although there is one contingency in Albania which leads to insecure state.
The most critical branch in the analyzed load/generation scenario is the transformer 400/220 kV in
the Mintia B substation in Romania. It is due to a high level of generation in DEVA 1 power plant
(850 MW), which is connected to 220 kV busbars in that substation. Noted transformer becomes
overloaded if one among eleven 400 kV and 220 kV lines or 400/220 kV transformers in the
Romanian or Serbian power system goes out of operation. Secure operating conditions can be
achieved if the generation level in DEVA 1 (Mintia) power plant is significantly lowered. Other
way to achieve more secure operation with the same level of production in DEVA 1 power plant is
to change network topology (connect generators in DEVA 1 to the second busbars or to connect
two busbar systems in Mintia substation). Loss of 400/220 kV transformer in the Mintia B
substation can jeopardize system security due to possible overloading of two 220 kV lines in
Romania. Loss of one 400 kV line in Romania (Mintia-Arad) is critical because the 400 kV line
Mintia-Sibiu can be overloaded. Single outages of 400/110 kV transformers in the stations Brasov
4.8
and Dirste in Romania are also found critical, since the second transformer 400/110 kV in the
Brasov substation is permanently out of operation in the model.
Loss of OHL 220 kV in the Belgrade area can cause overloading of the parallel line. Loss of one
400/110 kV transformer in the Nis substation is critical due to possible overloading of the other
parallel one.
Loss of 220 kV line between the Rashbul and Tirana substations can cause overloading of 220 kV
line between the Elbasan and Fier substations in Albania.
The heaviest line overloading (147% Ithermal) in the analyzed scenario is related to a 220 kV line in
Romania around the Mintia substation. The heaviest transformer overloading (137% Sn) is related to
the transformer 400/110 kV in the Dirste substation (Romania) when the transformer 400/110 kV in
the Brasov substation is outaged (the parallel one is permanently out of operation in the model).
Figure 4.1.6 shows geographical positions of critical elements in the analyzed scenario. A green
color reveals 220 kV elements (line 220 kV or transformer 220/x kV), while a red one reveals 400
kV elements (line 400 kV or transformer 400/x kV).
According to the obtained and presented results, it may be concluded that a re-dispatching of
generation in the Romanian power system, especially of the DEVA 1 power plant (to decrease its
generation level from the initially assumed 850 MW in this scenario), as well as certain
reinforcements in the internal networks of Romania, Albania and Serbia are necessary shall this
generation/load pattern be made more secure. Re-dispatching of Romanian power plants may be
avoided if network topology is changed, especially in Mintia substation. None of the identified
congestions is located at the border lines.
Table 4.1.5 - Lines overloadings for 2010–base case-average hydrology scenario, single outages
Outage
Overloaded line(s)
OHL 220 kV AKASHA2-ARRAZH2
OHL 400 kV MINTIA-ARAD
TR 400/220 kV MINTIA B
TR 400/220 kV MINTIA B
OHL 220 kV JBGD172-JBGD8 22 ckt.1
OHL 220 kV AELBS12-AFIER 2
OHL 400 kV MINTIA-SIBIU
OHL 220 kV PESTIS-MINTIA A
OHL 220 kV PESTIS-MINTIA B
OHL 220 kV JBGD172-JBGD8 22 ckt.2
Loadings
MVA
%
253.1 113.2
408.2 106.5
431.1 147.3
351.4 119.3
434.8 122.2
Country
ALBANIA
ROMANIA
SERBIA
Table 4.1.6 - Transformers overloadings for 2010–base case-average hydrology scenario, single outages
Outage
Overloaded branch(es)
OHL 400 kV TANTAREN-SIBIU
OHL 220 kV HAJD OT-MINTIA B
OHL 220 kV PESTIS-MINTIA A
OHL 220 kV MINTIA A-TIMIS
OHL 220 kV MINTIA B-AL.JL
OHL 220 kV CLUJ FL-AL.JL
TR 400/220 kV MINTIA A
TR 400/220 kV MINTIA B
TR 400/220 kV ARAD
TR 400/220 kV BUC.S ckt.1
400/110 kV BRASOV
400/110 kV DIRSTE
400/110 kV NIS ckt.1
TR 400/220 kV MINTIA B
TR 400/220 kV MINTIA B
TR 400/220 kV MINTIA B
TR 400/220 kV MINTIA B
TR 400/220 kV MINTIA B
TR 400/220 kV MINTIA B
TR 400/220 kV MINTIA B
TR 400/220 kV MINTIA A
TR 400/220 kV MINTIA B
TR 400/220 kV BUC.S ckt.2
400/110 kV DIRSTE
400/110 kV BRASOV
400/110 kV NIS ckt.2
Loadings
MVA
%
412.1 103.0
404.4 101.1
511.2 127.8
412.3 103.1
474.8 118.7
496.0 124.0
530.8 132.7
459.5 114.9
401.2 100.3
410.2 102.5
341.3 136.5
337.5 135.0
312.3 104.1
Country
ROMANIA
SERBIA
4.9
SVK
AUT
HUN
SLO
ROM
CRO
SRB
BIH
MNG
BUL
MKD
TUR
ALB
GRE
critical elements
400 kV line or 400/x kV transformer
220 kV line or 220/x kV transformer
Figure 4.1.6 - Geographical positions of the critical elements for 2010-base case-average hydrology scenario
4.10
4.2 Scenario 2010 – dry hydrology
This part of the Study presents the results of static load flow and voltage profile analyses that are
conducted for complete network topology and (n-1) contingencies in the scenario which is denoted
as GTmax run for year 2010 - dry hydrology.
4.2.1 Lines loadings
Figure 4.2.1 shows power exchanges between areas for 2010-dry hydrology scenario. Power flows
along interconnection lines in the region together with balances of the systems are shown in Figure
4.2.2. Area totals are shown in Table 4.2.1. Figure 4.2.3 shows histogram of tie lines loadings. It is
concluded that most of the tie lines are loaded less than 25% of their thermal limits.
UKR
30
9
28
AUT
0
42
450
SVK
HUN
16
8
68
31
47
SLO
7
14
CRO
ROM
195
2
52
4
17
164
SRB
89
BIH
0
36
273
27
MNG
7
16
34
BUL
33
27
ALB
122
74
MKD
95
ITA
27
TUR
0
25
GRE
Figure 4.2.1 - Area exchanges in analyzed electric power systems for 2010-dry hydrology scenario
Table 4.2.1 - Area totals in analyzed electric power systems for 2010-dry hydrology scenario
AREA
ALBANIA
BULGARIA
BIH
CROATIA
MACEDONIA
ROMANIA
SERBIA
MONTENEGRO
TOTALS
GENERATION
953.3
6709.6
2842.4
2294.6
1021.1
6742
7311.7
467.8
28342.5
LOAD
1290.5
5970
1989
3147
1206
6703.8
6944
671
27921.3
LOSSES
46.8
133.6
60.3
41.6
18.1
238
201.6
16.9
756.9
INTERCHANGE
-384
606
793
-894
-203
-199.9
166
-220
-335.9
4.11
LEVIC 1400.0
CENTREL
200
140
251
113
GABICK1400.0
450.0 MW
MSAJO 4
410.0
224.4
UMUKACH
412.0
37.
70. 1
5
222.5
MSAFA 4
38.0 ARAD
43.0
401.5
ROM
-369.2 MW
JSUBO31 396.0
JHDJE11
P.D.FIE
69.5
72.5
69.5
71.5
228
145
230 SOFIA_W4
92.4
411.0
TANTAREN
406.0
53.9
16.2
SCG
231.7
138
29.5
JHPIVA21
139 236.0
26.3
103
43.5
JHPERU21
102 221.1
45.0
CRG
-107.7 MW
AEC_400
JNIS2 1
396.5
147
JPODG211
181
JPODG121219.0
204
180
AVDEJA2
210.7AFIERZ2208.0
373.2
SK 1 222.9
SK 4
407.5
MKD
BITOLA 2
414.2
410.5
1
18. 1
72.
ALB
357
271
KARDIA K
364 414.9
257
QES/H
5.0
14.0
AHS_FLWR416.1
AZEMLA1
K
MI_3_4_1 419.3
412.2
FILIPPOI
8
17 7.2
7
415.5
KV: 110 , 220 , 400
19.8
21.7
K-KEXRU
417.8
4HAMITAX
8.5
8.5 415.4
75.0
19.8 63.7
4BABAESK
70.7
415.5
0.0 MW
0.2 MW
KARACQOU
250 418.1
50.0
1
13 1.3
.8
.3
11 9.1
8
.4
35 . 4
49
LAGAD K 413.3
381.4
412.8
BLAGOEV413.3
410.7
1
18. 7
49.
-720.0 MW
C_MOGILA
.9
47 .8
59
STIP 1
DUBROVO
358.9
.0
48 .0
34
407.7
5 9
8. 6.
1
202
204
BUL
766.0 MW
JVRAN31
0.000
-444.0 MW
AKASHA1
DOBRUD4 417.1
414.7
8
28 1.9
.5
72.8
50.8
204
180
347
20.0
7
18. 0
17.
7
18. 0
17.
8
25 2.
.1 8
384.0
JLESK21
0.000
JPRIZ22
214.1 JTKOSA2
227.0JTKOSB1412.4
5.0
48.0
148
147
756.2 MW
SRB
864.0 MW
345
48.8
RP TREB
228.9
JVARDI22
37.3 229.6
54.0
18.6
27.5
SA 20
230.5
37.2
51.9
18.6
27.5
VISEGRA
231.4
AVDEJA1
GIS REGIONAL MODEL - DRY HYDRO
2010 TOPOLOGY 2010
413.7
UGLJEVIK 400.9
RP TREB
396.9
SHAW POWER
TECHNOLOGIES
INC. R
ISACCEA 405.4
400.9
73.7
47.1
225.5
TE TUZL
402.4
38.0
16.6
208
JSMIT21
85.1
205
165
404.6
130 MO-4
19.2
232.4
BIH
758.8 MW
ROSIORI 402.1
.1
76 .0
14
.1
76 .0
14
MO-4
64.8
NADAB
71.6
402.2
212
121
71.
8
8
71 5.8
85 .8
.8
JSOMB31
393.4
207
40.3
404.5
153
4
81.
127
18.2
319
24.0
220.1
2
11 0
.
46
317
16.4
GRADACAC
64.7
52.7
76.1
35.8
4.9
6.8
34.7
26.6
219.6
4.9
1.1
.9
34 .3
34
PRIJED2
402.2
39.1
47.0
17
45. 9
5
DAKOVO
222.3MEDURIC
2
60. 7
22.
CRO
-1086.4 MW
HE ZAKUC
231.4
39.0
71.8
.5
30 41
1
405.9
KONJSKO
10.6
29.2
35.3
45.4
401.0
MRACLIN
408.2
5
24 4.0
.0
TUMBRI
153
29.5
ERNESTIN
110
47.0
6
15 0
2.
6
15 0
2.
10.7
7.6
227.1
.3
.3
35 .2 35 .2
85
85
ZERJAVIN
403.4
ZERJAVIN 226.9
402.1
PEHLIN 229.2
10.8
7.6
UKR
450.0 MW
30.7
165
LCIRKO2
71
37 .7
.8
71
37 .7
.8
401.0
5
6. .7
5
MELINA
35.6
UMUKACH2
27.7
MBEKO 4
404.6
408.3
MPECS 4
LKRSKO1
10
2. .7
6
35.5
20.9
35.4
21.9
SLO
97
400.9 30 .7
.7
LDIVAC2
229.3
35.4
21.9
215.0 MW
59.9
27.1
11.4
77.1
MHEVI 4
6.5
4.2
11.4
73.6
22.9 228.9
26.6
LMARIB1 402.4
97.8
7.2
230.3
180
24.1
23.0
12.7
MTLOK 2
230.5
HUN
156
26.9
IPDRV121
180
3.3
52.2
20.7
MKISV 2
226.3
-1250.0 MW
156
26.9
400.9
MGYOR 2
52.7
5.8
39.
10. 5
5
39.
10. 5
5
154
45.
7
IRDPV111
407.6
8
36. 1
13
LPODLO2228.9
MSAJI 408.1
4
214
9.8
39.5
41.8
39.5
41.8
234.2
ONEUSI2
233.5
LDIVAC1
MGOD
OWIEN 2
350
UZUKRA01
297
772.5
8.5
63.7
OKAINA1 401.5
156
49.2
OOBERS2238.7
346
804
76.1
35.8
106
24.2
405.4
199
94.6
UCTE
223.5 MW
MAISA 7
714.8
MGYOR 4
0
25 .4
105
90
101
403.3
OWIEN 1
GRE
TUR
BUS -VOLTAGE(KV)
BRANCH -MW/MVAR
EQUIPMENT -MW/MVAR
Figure 4.2.2 - Power flows along interconnection lines in the region with balances of the systems for 2015-dry hydrology scenario – 2010 topology
4.12
Histogram
50
Frequency
45
40
35
30
25
20
15
10
5
0
25
50
75
100
More
Bin
Figure 4.2.3 - Histogram of interconnection lines loadings for 2010-dry hydrology scenario ("Frequency" denotes
number of lines and "Bin" denotes loading range in % of thermal limit)
Following Table 4.2.2 shows all network elements loaded over 80% of their thermal limits. As it
can be seen only transformers 220/110 kV in substation Fier in Albania are highly loaded. Also,
transformer 220/110 kV in substation Fundeni in Romania is loaded over 80 % of its thermal limit.
Figure 4.2.4 shows histogram of branch loadings in the system. As for the conclusion regarding
thermal loadings in this scenario it can be said that almost all of the network elements are loaded
less then 75% of their thermal limits, but there are some elements highly loaded. The elements
loaded over 80% are transformers 220/110 kV in substation Fier and transformer 220/110 kV in
substation Fundeni in Romania, so some internal network reinforcements are necessary to sustain
this load-demand level and production pattern. It is expected that transformers in Fierza substation
will be replaced with more powerful transformer units.
Table 4.2.2 - Network elements loaded over 80% of their thermal limits for 2010-dry hydrology scenario
BRANCH LOADINGS ABOVE
AREA
80.0 % OF RATING:
LOADING
MVA
Transformers
kV AFIER 1
106.8
kV AFIER 2
87.5
kV AFIER 3
83.5
kV FUNDE2 1
168.1
ELEMENT
TR
TR
TR
TR
ALB
ROM
220/110
220/110
220/110
220/110
RATING
MVA
PERCENT
120
90
90
200
89
97.2
92.8
84.1
Histogram
400
350
Frequency
300
250
200
150
100
50
0
x<25
25<x<50 50<x<75 75<x<100
x>100
Bin
Figure 4.2.4 - Histogram of branch loadings for 2010-dry hydrology scenario ("Frequency" denotes number of branches
and "Bin" denotes loading range in % of thermal limit)
4.13
4.2.2 Voltage Profile in the Region
Figure 4.2.5 shows histogram of voltages in monitored substations. Voltages in almost all
monitored substations are found within permitted limits.
Histogram
140
120
Frequency
100
80
60
40
20
x>1.1
1.05<x<1.1
1.025<x<1.05
1<x<1.025
0.975<x<1
0.95<x<0.975
0.9<x<0.95
x<0.9
0
Bin
Figure 4.2.5 - Histogram of voltages in monitored substations for 2010-dry hydrology scenario ("Frequency" denotes
number of busses and "Bin" denotes voltage range in p.u.)
In the rest of the monitored network the voltage profile is satisfying and that most of the substations
have magnitudes in range 0.975-1.05 p.u.
4.2.3 Security (n-1) analysis
Results of security (n-1) analysis for 2010-dry hydrology scenario are presented in Table 4.2.3.
Figure 4.2.6 shows the geographical position of the critical elements in monitored systems.
It can be concluded that all identified insecure states are located in internal networks that belong to
monitored power systems of Romania and Serbia. In the most critical cases in Romanian system,
the critical elements are transformers 400/110 kV in substations Dirste and Brasov. The most
critical elements in Serbian system are lines 220 kV Beograd 8 – Beograd 17 and Obrenovac –
Beograd 3.
Some of the overloadings identified can be relieved by certain dispatch actions (splitting busbars,
changing lower voltage network topology in order to redistribute load-demand or change of
generation units engagement), like in the case of most severe overloading in Romanian network
happens on transformer 400/110 kV Dirste when transformer 400/110 kV in Brasov is outaged, but
this is a consequence of the fact that second transformer unit 400/110 kV in Brasov is out of
4.14
operation. Switching on of this transformer clears this critical outage. The similar situation is with
outage of the 400 kV line Obrenovac-Beograd 3 in Serbia. Splitting off the 220 kV busbars in
substation Beograd 3 relieves this overloading, but voltage profile in Serbian network remains
critical, so additional dispatching actions are necessary too.
All in all, certain reinforcement of internal network is necessary (more about this in chapter 6) in
order to make this regime more secure. None of the identified congestions is located at border lines.
Table 4.2.3 - Network overloadings for 2010-dry hydrology scenario, single outages
Area
1
CS
CS
RO
RO
CS
RO
contingency
2
OHL 220kV JBGD172 -JBGD8 22 1
OHL 400kV JBGD8 1 -JOBREN11 1
TR 400/110 BRASOV
1
TR 400/110 DIRSTE
1
TR 400/110 JNIS2 1
TR 400/220 BUC.S
1
RO
TR 400/220 MINTIA
1
RO
TR 400/220 MINTIA
1
overloadings
/
Area
out of limits voltages
3
4
CS
HL 220kV JBGD172-JBGD8 22
CS
HL 220kV JBGD3 21-JOBREN2
RO
TR 400/110kV/kV DIRSTE
RO
TR 400/110kV/kV BRASOV
CS
TR 400/110kV/kV JNIS2 1
RO
TR 400/220kV/kV BUC.S
RO
TR 400/220kV/kV MINTIA
RO
HL 220kV PESTIS-MINTIA A
RO
TR 400/220kV/kV MINTIA
#
5
2
1
1
1
2
2
1
1
1
limit
/
Unom
6
365.8MVA
301MVA
250MVA
250MVA
300MVA
400MVA
400MVA
277.4MVA
400MVA
Flow
rate
/
/
Voltage volt.dev.
7
8
440.8MVA
123.1%
340.1MVA
115.4%
333.3MVA
133.3%
329.9MVA
132.0%
338.2MVA
112.7%
403.8MVA
101.0%
419.5MVA
104.9%
313.9MVA
105.5%
433MVA
108.3%
SVK
AUT
HUN
SLO
ROM
CRO
SRB
BIH
MNG
BUL
MKD
TUR
ALB
critical elements
GRE
400 kV line or 400/x kV transformer
220 kV line or 220/x kV transformer
Figure 4.2.6 – Geographical position of critical elements for 2010-dry hydrology scenario
4.15
4.3 Scenario 2010 – wet hydrology
This part of the Study presents results of static load flow and voltage profile analyses that are
conducted for complete network topology and (n-1) contingencies in the scenario which is denoted
as Scenario 2010 – wet hydrology.
4.3.1 Line loadings
Area totals and power exchanges for the 2010-base case-wet hydrology scenario are shown in
Figure 4.3.1 and Table 4.3.1. Power flows along regional interconnection lines and system balances
are shown in Figure 4.3.2. Power flows along interconnection lines are also given in Table 4.3.2,
while Figure 4.3.3 shows histogram of tie lines loadings.
SVK
97
54
1
AUT
7
4 50
UKR
51
HUN
37
2
71
30
6
35
6
28
SLO
ITA
CRO
ROM
356
17
5
5
50
12
SRB
357
BIH
25
1
9
12
41 4
MNT
4
21
6
1
118
BUL
ALB
0
99
MKD
247
1
15
TUR
16
4
41
GRE
2 50
Figure 4.3.1 - Area exchanges in analyzed electric power systems for 2010-base case-wet hydrology scenario
Table 4.3.1 - Area totals in analyzed electric power systems for 2010-base case-wet hydrology scenario
Country
Albania
Bulgaria
Bosnia and Herzegovina
Croatia
Macedonia
Romania
Serbia and UNMIK
Montenegro
TOTAL - SE EUROPE
Generation
(MW)
898.6
6940.4
2037.1
1735.4
1081.8
7342.4
7406.2
586.3
28028.1
Load
(MW)
1287.3
5966.2
1961.1
3136.7
1194.0
6423.7
6875.1
669.2
27513.3
Bus Shunt
(Mvar)
0
0
0
0
0
0
0
0.6
0.6
Line Shunt
(Mvar)
0
14.3
0
0
0
79.9
13.6
1.6
109.4
Losses
(MW)
50.3
120.9
57.0
50.7
19.8
206.6
220.4
15.9
741.5
Net Interchange
(MW)
-439.0
839.0
19.0
-1452.0
-132.0
632.2
297.1
-101.0
-336.7
4.16
LEVIC 1400.0
CENTREL
200
144
251
117
GABICK1400.0
450.0 MW
MSAJO 4
410.0
174
53.0
97.
6.6 1
ROSIORI 410.7
404.5
155
6.9
155 ARAD
48.5
403.9
ROM
172
97.1
ISACCEA 415.7
631.9 MW
JSUBO31 398.7
357
JSMIT21
104
402.4
JHDJE11
175
38.0
175
37.1
P.D.FIE
TANTAREN
405.0
276
30.1
415.7
UGLJEVIK 403.7
SCG
231.3
169
16.2
JHPIVA21
171 235.8
18.1
35.3
18.0
JHPERU21
35.2 229.9
25.9
19.9
78.3
20.0
JPODG211
122
JPODG121228.8
CRG
-101.0 MW
839.0 MW
60.8
21.7
28.5
29.7
SK 1 224.2
BITOLA 2
C_MOGILA
SK 4
408.1
416.6
412.9
AZEMLA1
412
134
KARDIA K
418 416.9
111
QES/H
59.4
30.5
AHS_FLWR417.8
K
413.3
415.7
9
22 8.1
7
15.9
21.8
K-KEXRU
417.9
4HAMITAX
7.5
7.5 415.6
81.2
15.9 69.8
4BABAESK
70.9
415.7
0.0 MW
-0.1 MW
KARACQOU
250 419.1
0.0
FILIPPOI
8
6. .4
1
3
8. .1
97
5
39. 2
23.
.7
17 .7
48
LAGAD K 414.2
398.3
MI_3_4_1 420.6
BLAGOEV414.8
412.8
59.4
31.2
5
39. 9
45.
ALB
-439.0 MW
414.2
0
15 .6
40
STIP 1
DUBROVO
386.2
1
15 6
.
48
409.0
5 0
7. 1.
1
24.1
168
BUL
JLESK21
387.6
4
35 .0
.1
2.9
15.6
24.5
129
AVDEJA2
227.5AFIERZ2228.6
395.7
MKD
KV: 110 , 220 , 400
DOBRUD4 421.8
416.1
415 SOFIA_W4
105
412.3
JVRAN31
0.000
-132.0 MW
AKASHA1
411
128
60.7
28.3
3
28. 6
19.
3
28. 6
19.
23.7
50.4
AEC_400
JNIS2 1
396.4
JPRIZ22
223.3 JTKOSA2
227.3JTKOSB1407.4
43 4.2
.8
397.6
196.1 MW
SRB
297.1 MW
28.5
29.7
RP TREB
233.3
JVARDI22
80.4 230.0
60.4
2.9
6.1
SA 20
231.3
80.1
59.3
23.8
22.4
VISEGRA
231.7
.2
80 .6
9
409.2
56.0 MO-4
25.6
235.2
BIH
19.0 MW
217
NADAB
63.5
405.5
.2
80 .6
9
MO-4
226.6
TE TUZL
127
7
76.
55.5
13.7
223
61.5
225.5
.6
92 .0
32
222
7.0
GRADACAC
217
49.0
35.2
151
JSOMB31
396.5
355
75.5
407.0
AVDEJA1
GIS REGIONAL MODEL - WET HYDRO
2010 TOPOLOGY 2010
412.0
80.3
40.5
224.3
RP TREB
404.6
SHAW POWER
TECHNOLOGIES
INC. R
224.4
UMUKACH
23
38. 0
5
57.2
14.8
59.4
24.6
226.3
6
49. 5
15.
.9
59 .8
31
PRIJED2
410.6
173
56.4
UKR
450.0 MW
17.8
47.0
DAKOVO
226.9MEDURIC
57.6
21.3
CRO
-1452.0 MW
HE ZAKUC
236.9
10.6
29.3
MSAFA 4
404.9
KONJSKO
10.8
7.7
.1
35 26
1
58 277
.9
404.2
MRACLIN
126
22.2
ERNESTIN
91.5
36.0
40.4
18.7
9
17 8
.
28
9
17 8
.
28
TUMBRI
409.4
59.
3
5
59 5.0
55 .3
.0
.3
.3
93 .4 93 .4
65
65
ZERJAVIN
406.6
ZERJAVIN 230.2
407.4
PEHLIN 233.8
228.8
59
.
5. 3
7
59
.
5. 3
7
LCIRKO2
403.2
.8 4
43 5.
1
MELINA
35.6
UMUKACH2
27.8
MBEKO 4
407.0
409.5
MPECS 4
LKRSKO1
40
28 .3
.2
35.5
21.0
93.6
40.7
SLO
10
403.1 90 0
.4
LDIVAC2
230.8
93.6
40.7
215.0 MW
49.4
20.5
30.4
49.0
MHEVI 4
44.1
24.2
30.4
45.4
17.6 229.0
27.9
LMARIB1 403.9
100
67.1
231.4
169
42.5
17.8
13.9
MTLOK 2
230.5
HUN
179
53.2
IPDRV121
168
63.5
46.0
22.3
MKISV 2
226.3
-1249.7 MW
179
53.2
401.5
MGYOR 2
46.4
6.9
53.
21. 2
6
53.
21. 2
6
162
36.
5
IRDPV111
407.9
3
97. 0
56.
LPODLO2229.9
MSAJI 408.2
4
173
47.0
53.3
52.8
53.3
52.8
234.3
ONEUSI2
233.5
LDIVAC1
MGOD
OWIEN 2
350
UZUKRA01
300
772.5
7.5
69.8
OKAINA1 402.6
164
41.0
OOBERS2238.7
346
802
80.3
40.5
87.2
22.9
405.6
199
98.7
UCTE
221.7 MW
MAISA 7
715.3
MGYOR 4
0
25 .8
87.1
94
103
403.5
OWIEN 1
GRE
TUR
BUS -VOLTAGE(KV)
BRANCH -MW/MVAR
EQUIPMENT -MW/MVAR
Figure 4.3.2 - Power flows along interconnection lines in the region for 2010 - base case - wet hydrology scenario
4.17
Table 4.3.2 - Power flows along regional interconnection lines for 2010 wet hydrology scenario
Power Flow
Interconnection line
OHL 400 kV
OHL 220 kV
OHL 220 kV
OHL 400 kV
OHL 400 kV
OHL 400 kV
OHL 400 kV
OHL 400 kV
OHL 220 kV
OHL 220 kV
OHL 220 kV
OHL 220 kV
OHL 220 kV
OHL 220 kV
OHL 220 kV
OHL 220 kV
OHL 220 kV
OHL 400 kV
OHL 400 kV
OHL 400 kV
OHL 400 kV
OHL 400 kV
OHL 400 kV
OHL 2x400 kV ckt.1
OHL 2x400 kV ckt.2
OHL 400 kV
OHL 2x400 kV ckt.1
OHL 2x400 kV ckt.2
OHL 2x400 kV ckt.1
OHL 2x400 kV ckt.2
OHL 2x400 kV ckt.1
OHL 2x400 kV ckt.2
OHL 400 kV
OHL 400 kV
OHL 220 kV
OHL 220 kV
OHL 400 kV
OHL 400 kV
OHL 400 kV
OHL 2x220 kV ckt.1
OHL 2x220 kV ckt.2
OHL 400 kV
OHL 400 kV OHL 400 kV
OHL 400 kV
OHL 400 kV
OHL 400 kV
OHL 220 kV
OHL 220 kV
Zemlak (ALB)
Fierze (ALB)
V.Dejes (ALB)
V.Dejes (ALB)
Ugljevik (B&H)
Mostar (B&H)
Ugljevik (B&H)
Trebinje (B&H)
Trebinje (B&H)
Prijedor (B&H)
Prijedor (B&H)
Gradacac (B&H)
Tuzla (B&H)
Mostar (B&H)
Visegrad (B&H)
Sarajevo 20 (B&H)
Trebinje (B&H)
Blagoevgrad (BUL)
M.East 3 (BUL)
M.East 3 (BUL)
M.East 3 (BUL)
C.Mogila (BUL)
Dobrudja (BUL)
Kozloduy (BUL)
Kozloduy (BUL)
Sofia West (BUL)
Zerjavinec (CRO)
Zerjavinec (CRO)
Ernestinovo (CRO)
Ernestinovo (CRO)
Tumbri (CRO)
Tumbri (CRO)
Melina (CRO)
Ernestinovo (CRO)
Zerjavinec (CRO)
Pehlin (CRO)
Dubrovo (MCD)
Bitola (MCD)
Skopje (MCD)
Skopje (MCD)
Skopje (MCD)
Arad (ROM)
Nadab (ROM)
Rosiori (ROM)
Portile De Fier (ROM)
Subotica (SER)
Ribarevine (MON)
Pljevlja (MON)
Pljevlja (MON)
Kardia (GRE)
Prizren (SER)
Podgorica (MON)
Podgorica (MON)
Ernestinovo (CRO)
Konjsko (CRO)
S. Mitrovica (SER)
Podgorica (MON)
Plat (CRO)
Mraclin (CRO)
Medjuric (CRO)
Djakovo (CRO)
Djakovo (CRO)
Zakucac (CRO)
Vardiste (SER)
Piva (MON)
Perucica (MON)
Thessaloniki (GRE)
Filippi (GRE)
Babaeski (TUR)
Hamitabat (TUR)
Stip (MCD)
Isaccea (ROM)
Tantarena (ROM)
Tantarena (ROM)
Nis (SER)
Heviz (HUN)
Heviz (HUN)
Pecs (HUN)
Pecs (HUN)
Krsko (SLO)
Krsko (SLO)
Divaca (SLO)
S.Mitrovica (SER)
Cirkovce (SLO)
Divaca (SLO)
Thessaloniki (GRE)
Florina (GRE)
Kosovo B (UNMIK)
Kosovo A (UNMIK)
Kosovo A (UNMIK)
Sandorfalva (HUN)
Bekescaba (HUN)
Mukacevo (UKR)
Djerdap (SER)
Sandorfalva (HUN)
Kosovo B (UNMIK)
Bajina Basta (SER)
Pozega (SER)
MW
-411.6
-4.0
2.9
-23.7
126.8
223.2
-275.6
-19.9
-104.8
59.9
57.6
49.6
92.6
56.0
-80.1
-169.2
35.3
17.8
230.0
7.5
8.4
151.0
172.5
-80.2
-80.2
414.9
-93.3
-93.3
-59.3
-59.3
-178.6
-178.6
100.0
-354.6
-43.8
40.4
-59.4
-39.5
60.8
28.5
28.5
155.0
217.3
97.3
174.5
35.2
-300.0
-30.1
46.0
Mvar
-134.3
35.1
-15.6
-50.4
-76.7
-61.5
30.1
78.3
39.9
-31.8
-21.3
15.5
32.0
-25.6
59.3
-16.2
18.0
-47.0
-38.5
-11.0
-6.1
-48.6
-47.0
-9.6
-9.6
105.0
-65.4
-65.4
-55.0
-55.0
28.8
28.8
52.7
75.5
15.4
18.7
-31.2
-45.9
-21.7
-29.7
-29.7
-48.5
-63.5
-56.0
37.1
-151.3
-16.4
8.5
34.5
% of thermal
rating
32
14
5
4
11
17
21
9
36
21
19
18
32
18
32
44
14
7
34
5
5
22
13
6
6
60
9
9
6
6
16
16
11
28
15
15
6
4
5
13
13
28
18
9
13
12
22
12
21
4.18
Figure 4.3.3 shows that the tie lines in the region are mostly loaded less than 25% of their thermal
limits for the analyzed hydrological base case scenario in year 2010. Among total number of forty
nine 400 kV and 220 kV interconnection lines in the region only eight are loaded between 25% and
50% of their thermal ratings. Only one line (OHL 400 kV Sofia – Nis between Bulgaria and Serbia)
is loaded more than 50% of its thermal rating, which is set at lower value (692.8 MVA) on the
Bulgarian side compared to the line rating on the Serbian side (1330.2 MVA).
45
40
Frequency
35
30
25
20
15
10
5
0
x<25
25<x<50
50<x<75
75<x<100
x>100
Bin
Figure 4.3.3 - Histogram of interconnection lines loadings for 2010-base case-wet hydrology scenario ("Frequency"
denotes number of lines and "Bin" denotes loading range in % of thermal limit)
Table 4.3.3 lists all network elements loaded over 80% of their thermal limits. As it can be seen
from this output list, most of the elements loaded over 80% are transformers in some substations
and internal 110 kV and 220 kV lines. Thus, certain internal network reinforcements are necessary
to sustain given load-demand level and production pattern.
Two 220/110 kV transformers in the Fierze substation in Albania are slightly overloaded in this
scenario, while the third transformer in the same substation is loaded near permitted values.
There are three 220 kV and two 110 kV internal lines in Romania which are loaded over 80% of
their thermal limits, but none of them is overloaded. These lines are related to the Lotru, Sibiu,
Parosen, Bojuren and Domnesti nodes. Transformer 220/110 kV in the Fundeni substation is highly
loaded in this scenario.
Two 220 kV lines and twelve 110 kV lines in the Serbian power system are highly loaded when all
branches are available in the analyzed scenario. Highly loaded 220 kV lines are connected to the
Obrenovac substation, while 110 kV lines are located mostly in the area of Belgrade. Four 110 kV
lines are overloaded, ranging between 108% Ithermal and 117% Ithermal.
Power systems of Bulgaria, Bosnia and Herzegovina and Croatia have several highly loaded
branches in 110 kV networks. High loading of 110 kV line Komolac-Plat is caused by radial
connection of one unit in the HPP Dubrovnik to the grid. High loading of 110 kV lines between the
Resnik, Zitnjak and TETO substations in the Zagreb area are caused by disconnection of generating
units in the TETO power plant in analyzed scenario. These combined heat and electricity generating
units are normally put into the operation during winter high load period, because their main purpose
is to produce heat for consumers in the Croatian capitol Zagreb.
4.19
Figure 4.3.4 shows histogram of 400 kV and 220 kV regional internal lines and 400/x kV and 220/x
kV transformers loadings. 46% of observed branches are loaded below 25% of their thermal ratings,
35% are loaded between 25% and 50%, 16% are loaded between 50% and 75% and only 2% of
observed branches are loaded between 75% and 100% of their thermal ratings. Two branches
(transformers 220/110 kV Fierze in Albania, 101% - 106% Sn) are overloaded if all branches are in
operation for the analyzed scenario.
Table 4.3.3 - Network elements loaded over 80% of thermal limits for 2010-base case-wet hydrology scenario
AREA
ROM
ALB
ROM
ELEMENT
Lines
OHL 220 kV LOTRU-SIBIU ckt.1
OHL 220 kV LOTRU-SIBIU ckt.2
OHL 220 kV TG.JIU-PAROSEN
Transformers
TR 220/110 kV AFIER 2-AFIER 5 ckt.1
TR 220/110 kV AFIER 2-AFIER 5 ckt.2
TR 220/110 kV AFIER 2-AFIER 5 ckt.3
TR 220/110 kV FUNDENI-FUNDE2B
LOADING
MVA
RATING
MVA
PERCENT
276.0
276.0
181.6
277.4
277.4
208.1
99.5
99.5
87.3
116.5
95.5
91.1
168.7
120.0
90.0
90.0
200.0
97.1
106.1
101.3
84.4
400
350
Frequency
300
250
200
150
100
50
0
x<25
25<x<50
50<x<75
75<x<100
x>100
Bin
Figure 4.3.4 - Histogram of 400 kV and 220 kV regional lines loadings for 2010-base case-wet hydrology scenario
("Frequency" denotes number of lines and "Bin" denotes loading range in % of thermal limit)
4.3.2 Voltage Profile in the Region
Voltage profile in the region within this scenario which is defined by given generation and demand
pattern is seen as satisfactory despite several appearances of certain bus voltage deviations which
are shown in Table 4.3.4. Presented table includes only 400 kV and 220 kV network buses.
Bus voltage magnitudes below permitted limits are not found in the analyzed scenario. Bus voltage
magnitudes that are found above permitted limits (110% Vnominal in 110 kV and 220 kV networks
and 105% Vnominal in 400 kV network) are detected only in Bulgaria. There are nine 400 kV buses
and four 220 kV buses with voltages slightly above permitted limits. Figure 4.2.5 shows histogram
of voltages in monitored 400 kV and 220 kV substations.
4.20
Table 4.3.4 - Bus voltage deviations for 2010-base case-wet hydrology scenario, complete network
Voltages
Country
Node
ALBANIA
BOSNIA AND HERZEGOVINA
400 kV VARNA4
400 kV BURGAS
400 kV MARITSA EAST2
400 kV TECMIG5
400 kV TECMIG7
400 kV DOBRUD4
400 kV MARITSA EAST 3_4_1
400 kV MARITSA EAST
400 kV TECMIG6
220 kV BPC_220
220 kV MIZIA2
220 kV AEC_220
220 kV TECVARNA
-
BULGARIA
CROATIA
MACEDONIA
MONTENEGRO
ROMANIA
SERBIA AND UNMIK
pu
1.057
1.052
1.056
1.052
1.051
1.055
1.051
1.051
1.052
1.103
1.101
1.103
1.104
-
kV
421.9
420.8
422.2
420.6
420.6
421.8
420.6
420.6
420.6
242.6
242.3
242.7
243.0
-
120
Frequency
100
80
60
40
20
x>1.1
1.05<x<1.1
1.025<x<1.05
1<x<1.025
0.975<x<1
0.95<x<0.975
0.9<x<0.95
x<0.9
0
Bin
Figure 4.3.5 - Histogram of voltages in monitored substations for 2010-base case-wet hydrology scenario ("Frequency"
denotes number of busses and "Bin" denotes voltage range in p.u.)
4.3.3 Security (n-1) analysis
Results of security (n-1) analysis for the 2010-base case-wet hydrology scenario are presented in
Table 4.3.5 and Table 4.3.6.
Insecure states for given generation and demand pattern are detected in the power systems of
Romania and Serbia, although there is one contingency in Albania which leads to insecure state.
The most critical branches in the analyzed load/generation scenario are those ones connected to the
Sibiu substation 400/220 kV in Romania. Double circuit line 220 kV Lotru-Sibiu is loaded near
4.21
upper limit (99% Ithermal) when all branches are available. It becomes overloaded if one circuit goes
out of operation. Line 400 kV Mintia-Sibiu is overloaded when 400 kV line Sibiu-Iernut goes out of
operation. Transformer 400/220 kV in Sibiu is overloaded if the parallel one is unavailable.
Reasons for these overloadings are found in a rather high level of hydro generation in the LOTRU
CIUNGET power station in this scenario (478 MW). More secure operation could be achieved by
decreasing a generation level in this hydro power plant.
Double circuit 220 kV lines around the Resita substation (Portile de Fier-Resita and TimisoaraResita) are overloaded if one circuit goes out of operation, due to high generation in the PORTILE
1 hydro power plant (796 MW). Secure operation could be achieved by decreasing a generation
level in this power plant.
OHL 220 kV Targa Jiu-Paroseni is overloaded if one of the 400 kV lines Mintia-Sibiu, TantareniSibiu in Romania or Djerdap-Kostolac B in Serbia goes out of operation. Higher engagement of the
TPP Paroseni resolves these overloads.
Single outages of 400/110 kV transformers in the stations Brasov and Dirste in Romania are also
recognized as critical, since the second transformer 400/110 kV in the Brasov substation is
permanently out of operation in the model.
Loss of OHL 220 kV in the Belgrade area can cause overloading of the parallel line. Loss of one
400/110 kV transformer in the Nis substation is critical due to possible overloading of the other
parallel one.
Loss of 220 kV line between the Rrashbul and Tirana substations can cause overloading of 220 kV
line between the Elbassan and Fier substations in Albania.
The heaviest line overloading (124% Ithermal) in the analyzed scenario is related to a 400 kV line in
Romania around the Sibiu substation (Mintia-Sibiu). The heaviest transformer overloading (137%
Sn) is related to the transformer 400/110 kV in the Dirste substation (Romania) when the
transformer 400/110 kV in the Brasov substation is outaged (the parallel one is permanently out of
operation in the model).
Figure 4.3.6 shows geographical positions of critical elements in the analyzed scenario. A green
color reveals 220 kV elements (line 220 kV or transformer 220/x kV), while a red one reveals 400
kV elements (line 400 kV or transformer 400/x kV).
According to the obtained and presented results, it may be concluded that a re-dispatching of
generation in the Romanian power system, especially of the HPP LOTRU CIUNGET and HPP
PORTILE 1 (to decrease their generation level in this scenario from the initially assumed 478 MW
and 796 MW respectively) and TPP PAROSENI (to increase its generation level in this scenario
from the initially assumed 120 MW), as well as certain reinforcements in the internal networks of
Romania, Albania and Serbia are necessary shall this generation/load pattern be made more secure.
None of the identified congestions is located at the border lines.
4.22
Table 4.3.5 - Lines overloadings for 2010–base case-wet hydrology scenario, single outages
Outage
Overloaded line(s)
OHL 220 kV AKASHA2-ARRAZH2
OHL 400 kV MINTIA-SIBIU
OHL 400 kV TANTAREN-SIBIU
OHL 400 kV SIBIU-IERNUT
OHL 220 kV P.D.F.A-RESITA ckt.1
OHL 220 kV RESITA-TIMIS ckt.1
OHL 220 kV JBGD172-JBGD8 22 ckt.1
OHL 220 kV AELBS12-AFIER 2
OHL 220 kV TG.JIU-PAROSEN
OHL 220 kV TG.JIU-PAROSEN
OHL 400 kV MINTIA-SIBIU
OHL 220 kV P.D.F.A-RESITA ckt.2
OHL 220 kV RESITA-TIMIS ckt.2
OHL 220 kV JBGD172-JBGD8 22 ckt.2
Loadings
MVA
%
251.4 111.9
242.7 111.7
227.1 104.5
477.3 124.3
315.5 108.4
315.9 112.8
434.4 122.2
Country
ALBANIA
ROMANIA
SERBIA
Table 4.3.6 - Transformers overloadings for 2010–base case-wet hydrology scenario, single outages
Outage
Overloaded branch(es)
TR 400/220 kV SIBIU ckt.1
400/110 kV BRASOV
400/110 kV DIRSTE
400/110 kV NIS ckt.1
TR 400/220 kV SIBIU ckt.2
400/110 kV DIRSTE
400/110 kV BRASOV
400/110 kV NIS ckt.2
Loadings
MVA
%
527.3 131.8
332.7 133.1
329.7 131.9
313.1 104.4
Country
ROMANIA
SERBIA
SVK
AUT
HUN
SLO
ROM
CRO
SRB
BIH
MNG
BUL
MKD
TUR
ALB
GRE
critical elements
400 kV line or 400/x kV transformer
220 kV line or 220/x kV transformer
Figure 4.3.6 - Geographical positions of the critical elements for 2010-base case-wet hydrology scenario
4.23
4.4 Scenario 2015 – average hydrology – 2010 topology
This part of the Study presents results of static load flow and voltage profile analyses that are
conducted for complete network topology and (n-1) contingencies in the scenario which is denoted
as GTmax run for year 2015 - average hydrology and expected network topology for 2010.
4.4.1 Lines loadings
Figure 4.4.1 shows power exchanges between areas for 2015-average hydrology scenario. Power
flows along interconnection lines in the region together with balances of the systems are shown in
Figure 4.4.2. Area totals are shown in Table 4.4.1. Figure 4.4.3 shows histogram of tie lines
loadings. It is concluded that most of the tie lines are loaded less than 25% of their thermal limits.
UKR
21
4
12
AUT
1
47
450
SVK
HUN
13
3
1
10
56
129
SLO
6
31
ITA
CRO
ROM
311
1
72
0
13
236
SRB
372
BIH
0
39
1
38
253
MNG
30
10
293
BUL
53
6
ALB
167
54
MKD
10
TUR
3
41
GRE
Figure 4.4.1 - Area exchanges in analyzed electric power systems for 2015-average hydrology scenario – 2010 topology
Table 4.4.1 - Area totals in analyzed electric power systems for 2015-average hydrology scenario – 2010 topology
AREA
GENERATION
LOAD
LOSSES INTERCHANGE
ALB
1116.1
1531
81.2
-496
BUL
7332.7
6483
150.7
699
BIH
2316.6
2279
78.7
-41.1
CRO
2316.3
3657
63.4
-1404.1
MKD
1087
1407
20
-340
ROM
7712.8
7317.4
347.4
47.9
SRB
8687.6
7263
268.7
1156
CRG
735
671
25
39
TOTALS
31304.1
30608.4
1035.1
-339.3
4.24
LEVIC 1400.0
CENTREL
200
136
251
109
GABICK1400.0
450.0 MW
MSAJO 4
410.0
407.4
MSAFA 4
224.4
UMUKACH
412.0
20.
106 2
36.6 ARAD
75.4
395.3
47.9 MW
JSUBO31 392.6
JHDJE11
P.D.FIE
130
13.8
130
12.8
252
139
253 SOFIA_W4
89.7
410.4
TANTAREN
401.0
281
11.1
JHPIVA21
229 235.7
41.0
7.9
22.4
JHPERU21
7.9 226.8
30.5
CRG
39.0 MW
159
JPODG211
129
JPODG121225.8
5
18 2.9
.2
19.7
20.1
49.6
124
270
34.7
6
11. 6
23.
6
11. 6
23.
49.6
124
AVDEJA2
224.3AFIERZ2223.9
383.4
SK 1 224.6
SK 4
409.2
MKD
BITOLA 2
415.0
411.2
7.8 0
69.
ALB
AZEMLA1
KARDIA K
417 415.8
178
QES/H
13.9
18.2
AHS_FLWR416.7
410
194
K
MI_3_4_1 419.2
412.7
FILIPPOI
1
19 4.3
7
415.6
KV: 110 , 220 , 400
10.1
22.8
K-KEXRU
417.8
4HAMITAX
4.2
4.2 415.3
74.0
10.1 62.7
4BABAESK
71.9
415.5
0.0 MW
0.0 MW
KARACQOU
250 418.6
50.0
5
15 .9
.0
0
6. .0
88
.8
24 .8
43
LAGAD K 413.6
390.3
412.4
BLAGOEV413.0
411.6
7.8
5
46.
-496.0 MW
C_MOGILA
.5
53 .1
50
STIP 1
DUBROVO
373.9
.6
53 .6
43
409.2
2 9
4. 7.
1
49.2
160
699.0 MW
JVRAN31
0.000
-340.0 MW
AKASHA1
DOBRUD4 416.8
413.9
BUL
JLESK21
0.000
JPRIZ22
222.3 JTKOSA2
229.5JTKOSB1415.1
5
25 3.
.6 3
389.7
13.9
44.0
158
92.8
AEC_400
JNIS2 1
396.1
19
46. 2
8
226
25.1
194
6.4
SCG
1195.0 MW
SRB
1156.0 MW
269
72.6
RP TREB
229.9
JVARDI22
90.3 228.5
54.4
11.4
34.3
SA 20
228.3
90.0
53.5
11.4
34.3
VISEGRA
230.0
AVDEJA1
GIS REGIONAL MODEL - AVERAGE HYDRO
2015 TOPOLOGY 2010
411.8
UGLJEVIK 398.2
230.2
RP TREB
396.8
SHAW POWER
TECHNOLOGIES
INC. R
ISACCEA 404.5
398.2
19.7
11.1
224.3
TE TUZL
399.2
36.5
17.6
312
JSMIT21
103
49.8
99.0
400.9
162 MO-4
9.5
231.2
BIH
-41.1 MW
ROSIORI 399.3
.3
21 .4
10
.3
21 .4
10
MO-4
96.4
NADAB
104
397.3
192
127
48.
3
9
48 5.2
95 .3
.2
JSOMB31
390.0
310
70.4
402.9
3
79. 6
88.
158
16.1
391
32.9
GRADACAC
219.0
96.2
86.5
21.3
60.4
221.4
.3
94 .3
44
4.9
0.9
.8
35 .0
34
PRIJED2
388
5.9
1
47. 5
22.
CRO
-1404.1 MW
399.4
96.6
50.3
UKR
450.0 MW
ROM
401.0
HE ZAKUC
228.0
96.3
65.9
9
12 3
15
38 282
.3
4.9
7.0
35.5
26.4
218.5
93.1
47.2
DAKOVO
221.2MEDURIC
46.9
27.3
400.2
MRACLIN
79.1
33.8
ERNESTIN
5
9. .7
8
4.2
6.5
226.8
.9
.9
75 .2 75 .2
84
84
ZERJAVIN
401.8
ZERJAVIN 226.0
0
16 2
4.
0
16 2
4.
TUMBRI
KONJSKO
10.7
29.1
24.7
51.2
LCIRKO2
48
47 .2
.2
48
47 .2
.2
400.4
401.8
PEHLIN 229.2
10.8
7.5
MBEKO 4
401.1
407.4
MPECS 4
LKRSKO1
MELINA
35.7
UMUKACH2
27.5
129
174
SLO
4.
3. 2
7
35.5
20.7
76.2
20.6
76.2
20.6
215.0 MW
55
400.8 33 .2
.0
LDIVAC2
229.3
MHEVI 4
9.5
1.1
8.8
77.9
16.1 228.8
28.4
LMARIB1 402.1
55.3
9.0
230.3
8.8
74.5
163
28.1
16.3
14.4
MTLOK 2
230.5
HUN
160
20.5
IPDRV121
163
7.0
44.2
23.0
MKISV 2
226.3
-1249.9 MW
160
20.5
400.9
MGYOR 2
44.6
7.5
56.
10. 0
4
56.
10. 0
4
159
46.
0
IRDPV111
407.3
7
20. 5
16
LPODLO2228.6
MSAJI 408.1
4
350
UZUKRA01
292
772.5
4.2
62.7
234.2
ONEUSI2
233.4
LDIVAC1
MGOD
OWIEN 2
56.0
41.4
56.0
41.4
161
50.6
OKAINA1 401.2
346
807
21.3
60.4
63.0
20.5
405.3
OOBERS2238.7
199
90.0
UCTE
224.4 MW
MAISA 7
714.0
MGYOR 4
0
25 .2
62.9
86
105
403.2
OWIEN 1
GRE
TUR
BUS -VOLTAGE(KV)
BRANCH -MW/MVAR
EQUIPMENT -MW/MVAR
Figure 4.4.2 - Power flows along interconnection lines in the region with balances of the systems for 2015-average hydrology scenario – 2010 topology
4.25
Histogram
45
40
Frequency
35
30
25
20
15
10
5
0
x<25
25<x<50
50<x<75 75<x<100
x>100
Bin
Figure 4.4.3 - Histogram of interconnection lines loadings for 2015-average hydrology scenario – 2010 topology
("Frequency" denotes number of lines and "Bin" denotes loading range in % of thermal limit)
Following Table 4.4.2 lists all network elements loaded over 80% of their thermal limits. As it can
be seen some lines 220 kV voltage level in Albania, Romania and Serbia are loaded over 80%.
Also, most of the elements loaded over 80% are transformers in some substations, again, in
Albania, Bosnia & Herzegovina, Romania and Serbia. Figure 4.4.4 shows histogram of branch
loadings in the system. As for the conclusion regarding thermal loadings in this scenario it can be
said that most of the network elements are loaded between 25-75% of their thermal limits, but there
are some elements highly loaded. Most of the elements loaded over 80% are transformers in some
substations, so some internal network reinforcements are necessary to sustain this load-demand
level and production pattern. There are some elements that are overloaded (220 kV lines Targu Jiu –
Paroseni and Urechesti-Targu Jiu in Romania, and 220/110 kV transformers in Fier substation in
Albania. This leads to conclusion that transmission network is not able to sustain this load-demand
level and this production pattern and certain network reinforcement are necessary. It should be
pointed out that higher engagement of TPP Paroseni resolves this overloads of the 220 kV lines
Targu Jiu – Paroseni and Urechesti-Targu Jiu and decreases the load of the 400/220 transformer in
Urechesti substation. Also, it is expected that transformers in Fier substation will be replaced with
more powerful transformer units.
4.26
Table 4.4.2 - Network elements loaded over 80% of their thermal limits for 2015-average hydrology scenario – 2010
topology
BRANCH LOADINGS ABOVE
AREA
80.0 % OF RATING:
LOADING
MVA
Lines
220kV AKASHA2-ARRAZH2 1
250.1
220kV P.D.F.A-CETATE1 1
205.2
220kV P.D.F.A-RESITA 1
236.2
220kV P.D.F.A-RESITA 2
236.2
220kV P.D.F.II-CETATE1 1
267.7
220kV TG.JIU-PAROSEN 1
278.2
220kV URECHESI-TG.JIU 1
278.2
220kV JBGD3 21-JOBREN2 1
261
Transformers
220/110 kV AELBS1 1
78
220/110 kV AELBS1 2
78
220/110 kV AELBS1 3
83.8
220/110 kV AFIER 1
138.2
220/110 kV AFIER 2
113.3
220/110 kV AFIER 3
108.1
220/110 kV AFIERZ 1
51.9
220/110 kV AFIERZ 2
51.9
220/110 kV AKASHA 1
84.1
220/110 kV AKASHA 2
84.1
400/110 kV UGLJEV 1
254.8
220/110 kV FUNDE2 1
193.3
220/110 kV FUNDEN 1
162.9
400/220 kV IERNUT 1
320.4
400/220 kV URECHE 1
395.4
220/110 kV JBGD3 1
166.7
220/110 kV JBGD3 2
125.6
220/110 kV JTKOSA 2
130.7
220/110 kV JTKOSA 3
133
220/110 kV JZREN2 2
123.6
400/220 kV JTKOSB 2
325.8
400/220 kV JTKOSB 3
325.8
ELEMENT
ALB
HL
HL
HL
HL
HL
HL
HL
HL
ROM
SRB
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
ALB
BIH
ROM
SRB
RATING
MVA
PERCENT
270
208.1
277.4
277.4
277.4
208.1
277.4
301
92.6
98.6
85.2
85.2
96.5
133.7
100.3
86.7
90
90
90
120
90
90
60
60
100
100
300
200
200
400
400
200
150
150
150
150
400
400
86.6
86.6
93.1
115.2
125.9
120.2
86.5
86.5
84.1
84.1
84.9
96.7
81.4
80.1
98.9
83.4
83.7
87.1
88.7
82.4
81.4
81.4
Histogram
350
300
Frequency
250
200
150
100
50
0
x<25
25<x<50 50<x<75 75<x<100
x>100
Bin
Figure 4.4.4 - Histogram of branch loadings for 2015-average hydrology scenario – 2010 network topology
("Frequency" denotes number of lines and "Bin" denotes loading range in % of thermal limit)
4.4.2 Voltage Profile in the Region
Figure 4.4.5 shows histogram of voltages in monitored substations. Voltages in almost all
monitored substations are found within permitted limits. Only in substation Elbasan and Kashar,
where voltages are around 373 kV, but this can be resolved with changing of the setting of the tap
changing transformers in these substations.
4.27
Histogram
100
90
80
Frequency
70
60
50
40
30
20
10
x>1.1
1.05<x<1.1
1.025<x<1.05
1<x<1.025
0.975<x<1
0.95<x<0.975
0.9<x<0.95
x<0.9
0
Bin
Figure 4.4.5 - Histogram of voltages in monitored substations for 2015-average hydrology scenario – 2010 network
topology ("Frequency" denotes number of busses and "Bin" denotes voltage range in p.u.)
In the rest of the monitored network the voltage profile is satisfying and that most of the substations
have magnitudes in range 0.975-1.05 p.u.
4.4.3 Security (n-1) analysis
Results of security (n-1) analysis for 2015-dry hydrology scenario and expected topology for 2010
are presented in Table 4.4.3. Figure 4.4.6 shows the geographical position of the critical elements in
monitored systems.
It can be concluded that all identified insecure situations are located in internal networks that belong
to monitored power systems of Albania, Croatia, Romania and Serbia. In most critical case in
Romanian system, the critical elements are 220 kV lines Targu Jiu – Paroseni and Urechesti-Targu
Jiu and 400/220 kV transformer in Urechesti substation, but these elements are overloaded by full
topology too, which is the main reason why they appear as critical by most outages analyzed. As it
has been stated before, this can be resolved by higher engagement of the TPP Paroseni.
Some of the overloadings identified can be relieved by changing internal network topology
(splitting busbars, changing lower voltage network topology in order to redistribute load-demand or
change of generation units engagement), like in the case of most severe overloading in Romanian
network happens on transformer 400/110 kV Dirste when transformer 400/110 kV in Brasov is
outaged, but this is a consequence of the fact that second transformer unit 400/110 kV in Brasov is
out of operation. Switching on of this transformer clears this critical outage. The similar situation is
with outage of the 400 kV line Obrenovac-Beograd 8 in Serbia. Splitting of the busbars in 220 kV
Beograd 3 in substation relieves this overloading, but voltage profile in Serbian network remains
critical, so additional dispatching actions are necessary too.
Losing the 400 kV line Kosovo B-Pec or one transformer unit in substation Kosovo B causes
overloading of the transformer units in this substation. This is consequence of the low level of
4.28
production in 220 kV network in Kosovo region, due to decommissioning of the generation units in
TPP Kosovo A.
All in all, certain reinforcement of internal network is necessary in order to make this regime more
secure (more about this in chapter 6). None of the identified congestions is located at border lines.
Table 4.4.3 - Network overloadings for 2015-average hydrology scenario, single outages – 2010 network topology
Area
1
contingency
2
RO
BASE CASE
OHL 220kV P.D.F.A -CALAFAT
1
RO
OHL 220kV P.D.F.A -RESITA
1
RO
OHL 220kV RESITA
-TIMIS
1
RO
OHL 220kV PESTIS
-MINTIA A 1
RO
OHL 220kV CLUJ FL -MARISEL
1
RO
OHL 220kV AL.JL
-GILCEAG
1
RO
OHL 400kV TANTAREN-SLATINA
1
RO
OHL 400kV TANTAREN-BRADU
1
RO
OHL 400kV TANTAREN-SIBIU
1
RO
OHL 400kV URECHESI-DOMNESTI 1
RO
OHL 400kV MINTIA
RO
OHL 400kV P.D.FIE -SLATINA
1
RO
OHL 400kV DOMNESTI-BRAZI
1
RO
OHL 400kV SUCEAVA -GADALIN
1
RO
OHL 400kV SMIRDAN -GUTINAS
1
RO
CS
CS
OHL 400kV GADALIN -ROSIORI 1
OHL 220kV JBGD172 -JBGD8 22 1
OHL 400kV JBGD8 1 -JOBREN11 1
CS
OHL 400kV JHDJE11 -JTDRMN1
1
CS
OHL 400kV JNSAD31 -JSUBO31
1
CS
OHL 400kV JTKOSB1 -JPEC
A
RO
RO
RO
CS
RO
TR
TR
TR
TR
TR
RO
TR 400/220 IERNUT
CS
TR 400/220 JTKOSB 1
RO
TR 400/220 SLATINA
400/110
400/110
400/110
400/110
400/220
-SIBIU
BRASOV
1
CLUJ E
1
DIRSTE
1
JNIS2 1
BUC.S
1
1
1
1
1
overloadings
/
out of limits voltages
Area
3
4
RO
HL 220kV TG.JIU-PAROSEN
RO
HL 220kV P.D.F.A-CETATE1
RO
TR 400/220kV/kV URECHESI
RO
HL 220kV URECHESI-TG.JIU
RO
HL 220kV P.D.F.A-RESITA
RO
HL 220kV TG.JIU-PAROSEN
RO
HL 220kV URECHESI-TG.JIU
RO
HL 220kV RESITA-TIMIS
RO
HL 220kV TG.JIU-PAROSEN
RO
TR 400/220kV/kV URECHESI
RO
HL 220kV URECHESI-TG.JIU
RO
HL 220kV TG.JIU-PAROSEN
RO
TR 400/220kV/kV URECHESI
RO
HL 220kV URECHESI-TG.JIU
RO
HL 220kV TG.JIU-PAROSEN
RO
TR 400/220kV/kV URECHESI
RO
HL 220kV URECHESI-TG.JIU
RO
HL 220kV TG.JIU-PAROSEN
RO
TR 400/220kV/kV URECHESI
RO
TR 400/220kV/kV URECHESI
RO
HL 220kV URECHESI-TG.JIU
RO
HL 220kV TG.JIU-PAROSEN
RO
TR 400/220kV/kV URECHESI
RO
HL 220kV URECHESI-TG.JIU
RO
HL 220kV TG.JIU-PAROSEN
RO
TR 400/220kV/kV URECHESI
RO
HL 220kV URECHESI-TG.JIU
RO
HL 220kV TG.JIU-PAROSEN
RO
TR 400/220kV/kV URECHESI
RO
HL 220kV URECHESI-TG.JIU
RO
HL 220kV TG.JIU-PAROSEN
RO
TR 400/220kV/kV URECHESI
RO
TR 400/220kV/kV URECHESI
RO
HL 220kV URECHESI-TG.JIU
RO
HL 220kV TG.JIU-PAROSEN
RO
HL 220kV BUC.S-B-FUNDENI
RO
HL 220kV TG.JIU-PAROSEN
RO
TR 400/220kV/kV URECHESI
RO
HL 220kV URECHESI-TG.JIU
RO
HL 220kV TG.JIU-PAROSEN
RO
HL 220kV TG.JIU-PAROSEN
CS
HL 220kV JBGD172-JBGD8 22
CS
HL 220kV JBGD3 21-JOBREN2
RO
TR 400/220kV/kV URECHESI
RO
HL 220kV URECHESI-TG.JIU
RO
HL 220kV TG.JIU-PAROSEN
CS
HL 220kV JBGD3 21-JOBREN2
RO
TR 400/220kV/kV URECHESI
RO
HL 220kV URECHESI-TG.JIU
RO
HL 220kV TG.JIU-PAROSEN
RO
HL 220kV URECHESI-TG.JIU
RO
HL 220kV TG.JIU-PAROSEN
CS
TR 400/220kV/kV JTKOSB1
CS
TR 400/220kV/kV JTKOSB1
CS
TR 400/220kV/kV JTKOSB1
RO
TR 400/110kV/kV DIRSTE
RO
HL 220kV TG.JIU-PAROSEN
RO
TR 400/110kV/kV BRASOV
CS
TR 400/110kV/kV JNIS2 1
RO
TR 400/220kV/kV BUC.S
RO
HL 400kV GADALIN-CLUJ E
RO
HL 220kV TG.JIU-PAROSEN
RO
HL 220kV STEJARU-GHEORGH
CS
TR 400/220kV/kV JTKOSB1
CS
TR 400/220kV/kV JTKOSB1
RO
TR 400/220kV/kV URECHESI
#
5
1
1
1
1
2
1
1
2
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
1
1
1
1
1
1
1
1
1
1
1
2
3
1
1
1
2
2
1
1
1
2
3
1
limit
/
Unom
6
208.1MVA
208.1MVA
400MVA
277.4MVA
277.4MVA
208.1MVA
277.4MVA
277.4MVA
208.1MVA
400MVA
277.4MVA
208.1MVA
400MVA
277.4MVA
208.1MVA
400MVA
277.4MVA
208.1MVA
400MVA
400MVA
277.4MVA
208.1MVA
400MVA
277.4MVA
208.1MVA
400MVA
277.4MVA
208.1MVA
400MVA
277.4MVA
208.1MVA
400MVA
400MVA
277.4MVA
208.1MVA
320MVA
208.1MVA
400MVA
277.4MVA
208.1MVA
208.1MVA
365.8MVA
301MVA
400MVA
277.4MVA
208.1MVA
301MVA
400MVA
277.4MVA
208.1MVA
277.4MVA
208.1MVA
400MVA
400MVA
400MVA
250MVA
208.1MVA
250MVA
300MVA
400MVA
238.3MVA
208.1MVA
208.1MVA
400MVA
400MVA
400MVA
rate
Flow
/
/
Voltage volt.dev.
7
8
282.6MVA
133.7%
259.4MVA
126.5%
417.6MVA
104.4%
299MVA
106.6%
337.8MVA
128.0%
299MVA
142.1%
292.7MVA
104.1%
348.7MVA
129.1%
292.7MVA
138.8%
416.1MVA
104.0%
293.8MVA
105.4%
287.2MVA
140.6%
412.9MVA
103.2%
293.3MVA
104.7%
293.3MVA
139.6%
424.3MVA
106.1%
306MVA
109.7%
299.7MVA
146.3%
442.8MVA
110.7%
420.4MVA
105.1%
296.5MVA
106.2%
296.5MVA
141.6%
443.9MVA
111.0%
327.5MVA
117.9%
327.5MVA
157.2%
444MVA
111.0%
299MVA
106.8%
299MVA
142.3%
443.2MVA
110.8%
330.1MVA
118.7%
322.6MVA
158.2%
428.1MVA
107.0%
413.5MVA
103.4%
295MVA
105.0%
295MVA
140.0%
331.2MVA
102.7%
282MVA
136.7%
413.5MVA
103.4%
298.1MVA
106.7%
298.1MVA
142.2%
280.1MVA
135.6%
462.8MVA
133.3%
429.3MVA
155.8%
436.7MVA
109.2%
316.6MVA
113.9%
316.6MVA
151.8%
319.6MVA
109.6%
422.5MVA
105.6%
303.6MVA
108.6%
303.6MVA
144.7%
294.1MVA
104.9%
294.1MVA
139.8%
419.7MVA
104.9%
438.6MVA
109.6%
438.6MVA
109.6%
385.8MVA
154.3%
282.5MVA
137.1%
380MVA
152.0%
332.5MVA
110.8%
483MVA
120.8%
238.1MVA
102.9%
282.5MVA
137.4%
211.8MVA
121.7%
457.2MVA
114.3%
457.2MVA
114.3%
424.9MVA
106.2%
4.29
SVK
AUT
HUN
SLO
ROM
CRO
SRB
BIH
MNG
BUL
MKD
TUR
ALB
GRE
critical elements
400 kV line or 400/x kV transformer
220 kV line or 220/x kV transformer
Figure 4.4.6 – Geographical position of critical elements for 2015-average hydrology scenario
4.30
4.5 Scenario 2015 – average hydrology – topology 2015
This part of the Study presents results of static load flow and voltage profile analyses that are
conducted for complete network topology and (n-1) contingencies in the scenario which is denoted
as GTmax run for year 2015 - average hydrology and expected network topology for 2015.
4.5.1 Lines loadings
Figure 4.5.1 shows power exchanges between areas for 2015-average hydrology scenario. Power
flows along interconnection lines in the region together with balances of the systems are shown in
Figure 4.5.2. Area totals are shown in Table 4.5.1. Figure 4.5.3 shows histogram of tie lines
loadings. It is concluded that most of the tie lines are loaded less than 25% of their thermal limits.
UKR
11
5
13
AUT
1
46
450
SVK
HUN
10
6
89
28
126
SLO
4
30
ITA
CRO
ROM
301
1
77
3
12
193
SRB
336
BIH
6
47
1
16
247
MNG
BUL
8
220
6
27
2
38
7
27
298
ALB
169
98
MKD
8
TUR
2
11
GRE
Figure 4.5.1 - Area exchanges in analyzed electric power systems for 2015-average hydrology scenario
Table 4.5.1 - Area totals in analyzed electric power systems for 2015-average hydrology scenario
AREA
GENERATION
LOAD
LOSSES INTERCHANGE
ALB
1118.1
1541
73.1
-496
BUL
7332.9
6483
150.9
699
BIH
2317.2
2279
79.2
-41
CRO
2317
3657
64
-1404
MKD
1088.4
1407
21.4
-340
ROM
7708.6
7317.4
343.1
48.1
SRB
8689.4
7279
254.4
1156
CRG
735.2
676
20.3
39
TOTALS
31306.8
30639.4
1006.4
-338.9
As it can be seen, new elements that are expected to be build till 2015 cause totally different
distribution of power flows in the southern part of the region (Albania, FYR of Macedonia, Serbia,
and Montenegro).
4.31
LEVIC 1400.0
CENTREL
200
137
251
109
GABICK1400.0
450.0 MW
MSAJO 4
410.0
UKR
450.0 MW
224.4
UMUKACH
412.0
11.
98. 1
7
ROM
JHDJE11
P.D.FIE
123
10.9
123
11.9
246
97.1
248 SOFIA_W4
43.0
411.5
TANTAREN
402.0
254
9.1
SCG
22.0
17.3
JHPERU21
22.1 230.3
25.6
225
35.2
226
JPODG211
71.4
JPODG121229.9
CRG
39.0 MW
JPRIZ22
226.6 JTKOSA2
232.4JTKOSB1419.0
SK 4
410.5
-340.0 MW
74.7
133
278
8
42.
395.2
ALB
411.7
DUBROVO
27
79 7
.7
AZEMLA1
112
106
KARDIA K
113 417.8
36.8
QES/H
K
FILIPPOI
413.5
KV: 110 , 220 , 400
415.7
4
19 3.7
7
4
14 .8
.1
8.1
23.1
K-KEXRU
417.9
4HAMITAX
3.3
3.3 415.4
74.8
8.0 63.4
4BABAESK
72.1
415.6
0.0 MW
-0.1 MW
KARACQOU
250 419.6
50.0
MI_3_4_1 419.4
8
4 . .0
89
.0
26 .1
45
LAGAD K 414.3
AHS_FLWR417.3
408.7
413.3
BLAGOEV413.9
412.0
242
8
63.
-496.0 MW
414.5
C_MOGILA
.1
98 .8
48
STIP 1
242 0
82.
AKASHA1
BITOLA 2
.4
98 .3
43
409.9
3 1
3. 7.
1
74.7
133
MKD
699.0 MW
JVRAN31
406.8
114
56.1
SK 1 226.3
DOBRUD4 417.1
414.4
BUL
JLESK21
403.8
1
10 14
.8
1
11 6.3
.4
AVDEJA2
228.4AFIERZ2229.2
402.6
20.7
39.9
53.0
26.1
223
16.1
398 8
97.
75.0
91.8
75.0
91.9
294
69.2
9
20. 4
29.
397
110
9
20. 4
29.
1
21 6.
.0 2
401.5
AEC_400
JNIS2 1
402.0
19
47. 5
1
JHPIVA21
225 237.3
43.7
14.0 13.9
30.1 68.4
222
29.1
1195.0 MW
SRB
1156.0 MW
56.5
42.4
RP TREB
232.4
JVARDI22
81.3 229.6
50.1
293
105
SA 20
229.5
81.0
48.8
20.7
39.9
VISEGRA
230.9
AVDEJA1
GIS REGIONAL MODEL - AVERAGE HYDRO
2015 TOPOLOGY 2015
412.3
UGLJEVIK 399.1
230.5
RP TREB
403.9
SHAW POWER
TECHNOLOGIES
INC. R
184
128
48.0 MW
JSUBO31 393.2
399.0
52.7
18.2
403.6
166 MO-4
7.9
232.3
BIH
-41.0 MW
ISACCEA 404.9
301
JSMIT21
98.1
223
6.4
162
15.4
MO-4
23.8 ARAD
71.1
396.1
1
4. 7
.
11
1
4. 7
.
11
409
21.2
224.6
TE TUZL
399.7
23.7
12.9
127
172
JSOMB31
390.6
299
63.9
403.3
9
89. 1
84.
406
3.5
GRADACAC
219.3
ROSIORI 400.0
4.1
61.9
221.7
.5
97 .3
44
7.6
0.8
.8
40 .7
33
PRIJED2
82.2
NADAB
98.9
397.9
25.9
50.2
218.7
7.6
7.1
40.5
26.3
221.4MEDURIC
82.0
81.1
94.3 94.4
51.1 71.9
DAKOVO
6
49. 3
22.
CRO
-1404.0 MW
401.0
87.4
49.9
MSAFA 4
402.1
HE ZAKUC
228.8
87.2
66.9
56.6
19.3
MRACLIN
PEHLIN 229.3
KONJSKO
10.7
29.2
6
12 0
15
38 255
.9
400.3
89.7
29.3
ERNESTIN
96.2
46.9
0
16 8
2.
0
16 8
2.
TUMBRI
407.6
54.
7
9
54 2.7
92 .7
.7
.7
.7
68 .6 68 .6
84
84
ZERJAVIN
402.1
ZERJAVIN 226.2
402.1
6.5
6.0
226.8
54
44 .6
.6
54
44 .6
.6
LCIRKO2
400.5
1
9. .4
8
MELINA
35.7
UMUKACH2
27.5
10.8
7.5
MBEKO 4
401.6
407.5
MPECS 4
LKRSKO1
6.
4. 5
2
35.5
20.7
68.9
20.8
SLO
63
400.9 35 .8
.4
LDIVAC2
229.3
68.9
20.8
215.0 MW
49.4
27.1
9.8
77.0
MHEVI 4
9.1
1.4
9.7
73.6
17.5 228.8
28.1
LMARIB1 402.2
63.9
11.4
230.3
17.6
14.1
MTLOK 2
230.5
HUN
160
21.9
IPDRV121
166
25.0
45.7
22.6
MKISV 2
226.3
-1249.9 MW
160
21.9
400.9
166
4.0
MGYOR 2
46.2
7.1
52.
10. 9
5
52.
10. 9
5
158
45.
9
IRDPV111
407.3
5
11. 8
15
LPODLO2228.7
MSAJI 408.1
4
185
7.2
52.9
41.6
52.9
41.6
234.2
ONEUSI2
233.4
LDIVAC1
MGOD
OWIEN 2
350
UZUKRA01
293
772.5
3.3
63.4
OKAINA1 401.3
160
50.3
OOBERS2238.7
346
806
4.1
61.9
71.3
21.1
405.3
199
90.7
UCTE
224.0 MW
MAISA 7
714.1
MGYOR 4
0
25 .0
71.2
87
104
403.2
OWIEN 1
GRE
TUR
BUS -VOLTAGE(KV)
BRANCH -MW/MVAR
EQUIPMENT -MW/MVAR
Figure 4.5.2 - Power flows along interconnection lines in the region with balances of the systems for 2015-average hydrology scenario
4.32
Histogram
45
40
Frequency
35
30
25
20
15
10
5
0
x<25
25<x<50
50<x<75 75<x<100
x>100
Bin
Figure 4.5.3 - Histogram of interconnection lines loadings for 2015-average hydrology scenario ("Frequency" denotes
number of lines and "Bin" denotes loading range in % of thermal limit)
Following PSS/E output (Table 4.5.2) lists all network elements loaded over 80% of their thermal
limits. Figure 4.5.4 shows histogram of branch loadings in the system.
Table 4.5.2 - Network elements loaded over 80% of their thermal limits for 2015-average hydrology scenario
BRANCH LOADINGS ABOVE
AREA
ALB
ROM
SRB
ALB
ROM
SRB
BIH
80.0 % OF RATING:
LOADING
MVA
RATING
MVA
PERCENT
237.7
205
232.2
232.2
267.5
272.6
272.6
261.8
270
208.1
277.4
277.4
277.4
208.1
277.4
301
88
98.5
83.7
83.7
96.4
131
98.3
87
74.3
74.3
79.8
134.9
110.6
105.6
50.3
50.3
82.1
82.1
193
162.6
390.9
166.6
125.8
123.5
252.5
90
90
90
120
90
90
60
60
100
100
200
200
400
200
150
150
300
82.5
82.5
88.7
112.5
122.9
117.3
83.8
83.8
82.1
82.1
96.5
81.3
97.7
83.3
83.9
82.3
84.2
ELEMENT
HL
HL
HL
HL
HL
HL
HL
HL
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
Lines
AKASHA2-ARRAZH2 1
P.D.F.A-CETATE1 1
P.D.F.A-RESITA 1
P.D.F.A-RESITA 2
P.D.F.II-CETATE1 1
TG.JIU-PAROSEN 1
URECHESI-TG.JIU 1
JBGD3 21-JOBREN2 1
Transformers
220/110 kV AELBS1 1
220/110 kV AELBS1 2
220/110 kV AELBS1 3
220/110 kV AFIER 1
220/110 kV AFIER 2
220/110 kV AFIER 3
220/110 kV AFIERZ 1
220/110 kV AFIERZ 2
220/110 kV AKASHA 1
220/110 kV AKASHA 2
220/110 kV FUNDE2 1
220/110 kV FUNDEN 1
400/220 kV URECHE 1
220/110 kV JBGD3 1
220/110 kV JBGD3 2
220/110 kV JZREN2 2
400/110 kV UGLJEV 1
220kV
220kV
220kV
220kV
220kV
220kV
220kV
220kV
4.33
Histogram
350
300
Frequency
250
200
150
100
50
0
x<25
25<x<50 50<x<75 75<x<100
x>100
Bin
Figure 4.5.4 - Histogram of branch loadings for 2015-average hydrology scenario ("Frequency" denotes number of lines
and "Bin" denotes loading range in % of thermal limit)
As it can be seen from these outputs, most of the network elements are loaded between 25-75% of
their thermal limits and most of the elements loaded over 80% are transformers in some substations,
so some internal network reinforcements are necessary to sustain this load-demand level and
production pattern. Also, planned network reinforcements compared to network topology 2010,
reduce load of some elements in southern part of Serbia.
4.5.2 Voltage Profile in the Region
Figure 4.5.5 shows histogram of voltages in monitored substations. Voltages in all monitored
substations are found within permitted limits. It is concluded that voltage profile is satisfying and
that most of the substations have magnitudes in range 1-1.05 p.u.
Histogram
120
Frequency
100
80
60
40
20
x>1.1
1.05<x<1.1
1.025<x<1.05
1<x<1.025
0.975<x<1
0.95<x<0.975
0.9<x<0.95
x<0.9
0
Bin
Figure 4.5.5 - Histogram of voltages in monitored substations for 2015-average hydrology scenario ("Frequency"
denotes number of busses and "Bin" denotes voltage range in p.u.)
4.34
4.5.3 Security (n-1) analysis
Results of security (n-1) analysis for 2015-average hydrology scenario are presented in Table 4.5.3
and Figure 4.5.6.
Like for expected topology 2010 (previous chapter), it can be concluded that all identified insecure
situations are located in internal networks that belong to monitored power systems of Albania,
Croatia, Romania and Serbia. Also, the planned network reinforcements till 2015 resolve some of
the noticed critical contingencies, especially in southern part of Serbia. The rest of the conclusions
are the same as for the analyzed topology 2010, and that is that certain level of network
reinforcement is necessary to make this regime more secure.
SVK
AUT
HUN
SLO
ROM
CRO
SRB
BIH
MNG
BUL
MKD
TUR
ALB
GRE
critical elements
400 kV line or 400/x kV transformer
220 kV line or 220/x kV transformer
Figure 4.5.6 – Geographical position of critical elements for 2015-average hydrology scenario
4.35
Table 4.5.3 - Network overloadings for 2015-average hydrology scenario, single outages
Area
1
contingency
2
AL
BA
RO
BASE CASE
OHL 220kV AELBS12 -AFIER 2
OHL 400kV VISEGRA -HE VG
OHL 220kV P.D.F.A -CALAFAT
1
1
1
RO
OHL 220kV P.D.F.A -RESITA
1
RO
RO
OHL 220kV RESITA -TIMIS
1
OHL 220kV CRAIOV B-ISALNI A 1
RO
OHL 220kV PESTIS
RO
OHL 220kV CLUJ FL -AL.JL
1
RO
OHL 220kV AL.JL
-GILCEAG
1
RO
OHL 400kV TANTAREN-SLATINA
1
RO
OHL 400kV TANTAREN-BRADU
1
RO
OHL 400kV TANTAREN-SIBIU
1
RO
OHL 400kV URECHESI-DOMNESTI 1
RO
OHL 400kV MINTIA
RO
RO
OHL 400kV P.D.FIE -SLATINA
OHL 400kV SUCEAVA -GADALIN
1
1
RO
OHL 400kV SMIRDAN -GUTINAS
1
RO
RO
CS
CS
OHL
OHL
OHL
OHL
1
1
1
1
CS
OHL 400kV JBOR 21 -JHDJE11
1
CS
OHL 400kV JHDJE11 -JTDRMN1
1
CS
OHL 400kV JNSAD31 -JSUBO31
1
RO
RO
RO
RO
TR
TR
TR
TR
BRASOV
CLUJ E
DIRSTE
BUC.S
1
1
1
1
RO
TR 400/220 IERNUT
1
CS
TR 400/220 JBGD8
RO
TR 400/220 SLATINA
400kV
400kV
220kV
400kV
400/110
400/110
400/110
400/220
SIBIU
GADALIN
JBGD172
JBGD8 1
-MINTIA A 1
-SIBIU
-IERNUT
-CLUJ E
-JBGD8 22
-JOBREN11
1
1
1
overloadings
/
Area
out of limits voltages
3
4
RO
HL 220kV TG.JIU-PAROSEN
AL
HL 220kV AKASHA2-ARRAZH2
BA
HL 220kV RP KAKAN-KAKANJ5
RO
HL 220kV P.D.F.A-CETATE1
RO
TR 400/220kV/kV URECHESI
RO
HL 220kV URECHESI-TG.JIU
RO
HL 220kV P.D.F.A-RESITA
RO
HL 220kV TG.JIU-PAROSEN
RO
HL 220kV RESITA-TIMIS
RO
TR 400/220kV/kV URECHESI
RO
TR 400/220kV/kV URECHESI
RO
HL 220kV URECHESI-TG.JIU
RO
HL 220kV TG.JIU-PAROSEN
RO
HL 220kV TG.JIU-PAROSEN
RO
TR 400/220kV/kV URECHESI
RO
HL 220kV URECHESI-TG.JIU
RO
HL 220kV TG.JIU-PAROSEN
RO
TR 400/220kV/kV URECHESI
RO
TR 400/220kV/kV URECHESI
RO
HL 220kV URECHESI-TG.JIU
RO
HL 220kV TG.JIU-PAROSEN
RO
TR 400/220kV/kV URECHESI
RO
HL 220kV URECHESI-TG.JIU
RO
HL 220kV TG.JIU-PAROSEN
RO
TR 400/220kV/kV URECHESI
RO
HL 220kV URECHESI-TG.JIU
RO
HL 220kV TG.JIU-PAROSEN
RO
TR 400/220kV/kV URECHESI
RO
HL 220kV URECHESI-TG.JIU
RO
HL 220kV TG.JIU-PAROSEN
RO
TR 400/220kV/kV URECHESI
RO
HL 220kV TG.JIU-PAROSEN
RO
TR 400/220kV/kV URECHESI
RO
HL 220kV URECHESI-TG.JIU
RO
HL 220kV TG.JIU-PAROSEN
RO
HL 220kV TG.JIU-PAROSEN
RO
HL 220kV TG.JIU-PAROSEN
CS
HL 220kV JBGD172-JBGD8 22
CS
HL 220kV JBGD3 21-JOBREN2
RO
TR 400/220kV/kV URECHESI
RO
HL 220kV URECHESI-TG.JIU
RO
HL 220kV TG.JIU-PAROSEN
RO
TR 400/220kV/kV URECHESI
RO
HL 220kV URECHESI-TG.JIU
RO
HL 220kV TG.JIU-PAROSEN
CS
HL 220kV JBGD3 21-JOBREN2
RO
TR 400/220kV/kV URECHESI
RO
HL 220kV URECHESI-TG.JIU
RO
HL 220kV TG.JIU-PAROSEN
RO
TR 400/110kV/kV DIRSTE
RO
HL 220kV TG.JIU-PAROSEN
RO
TR 400/110kV/kV BRASOV
RO
TR 400/220kV/kV BUC.S
RO
HL 220kV TG.JIU-PAROSEN
RO
HL 220kV STEJARU-GHEORGH
CS
HL 220kV JBGD3 21-JOBREN2
CS
HL 220kV JBGD3 22-JBGD8 22
RO
TR 400/220kV/kV URECHESI
#
5
1
1
1
1
1
1
2
1
2
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
1
1
1
2
1
limit
/
Unom
6
208.1MVA
270MVA
316MVA
208.1MVA
400MVA
277.4MVA
277.4MVA
208.1MVA
277.4MVA
400MVA
400MVA
277.4MVA
208.1MVA
208.1MVA
400MVA
277.4MVA
208.1MVA
400MVA
400MVA
277.4MVA
208.1MVA
400MVA
277.4MVA
208.1MVA
400MVA
277.4MVA
208.1MVA
400MVA
277.4MVA
208.1MVA
400MVA
208.1MVA
400MVA
277.4MVA
208.1MVA
208.1MVA
208.1MVA
365.8MVA
301MVA
400MVA
277.4MVA
208.1MVA
400MVA
277.4MVA
208.1MVA
301MVA
400MVA
277.4MVA
208.1MVA
250MVA
208.1MVA
250MVA
400MVA
208.1MVA
208.1MVA
301MVA
365.8MVA
400MVA
Flow
rate
/
/
Voltage volt.dev.
7
8
277.6MVA
131.0%
364.3MVA
142.6%
338.8MVA
103.8%
259.1MVA
125.9%
413.4MVA
103.3%
293.7MVA
104.4%
333.5MVA
125.7%
293.7MVA
139.2%
343.7MVA
126.7%
412.5MVA
103.1%
412.7MVA
103.2%
289.3MVA
103.5%
283MVA
138.1%
264.9MVA
124.7%
420.3MVA
105.1%
301MVA
107.6%
294.9MVA
143.5%
437.9MVA
109.5%
416.1MVA
104.0%
291.3MVA
104.0%
291.3MVA
138.6%
438.7MVA
109.7%
321.3MVA
115.3%
321.3MVA
153.7%
439.9MVA
110.0%
294MVA
104.7%
294MVA
139.5%
437.6MVA
109.4%
323.1MVA
115.8%
316MVA
154.4%
424.6MVA
106.2%
276.9MVA
133.8%
409.4MVA
102.4%
292.8MVA
104.5%
292.8MVA
139.3%
277.6MVA
134.5%
277.8MVA
134.5%
463.6MVA
133.1%
432.1MVA
156.1%
414.3MVA
103.6%
291.3MVA
103.8%
291.3MVA
138.4%
430.1MVA
107.5%
309.2MVA
110.8%
309.2MVA
147.7%
319.4MVA
109.2%
418.4MVA
104.6%
298.5MVA
106.4%
298.5MVA
141.9%
385.1MVA
154.1%
277.8MVA
134.4%
379.5MVA
151.8%
482.6MVA
120.6%
277.8MVA
134.6%
210.4MVA
120.3%
314.5MVA
108.5%
372MVA
107.5%
421MVA
105.2%
4.5.4 Summary of Impacts - 2015 topology versus 2010 topology
Compared to the expected topology 2010, analyzed in previous chapter, it can be seen that the
network losses are smaller as a consequence of building of new elements for 2015 network
topology, especially in the cases of Albania, Serbia and Montenegro. Overall reduction of loses is
around 30 MW.
Also, planned network reinforcements compared to network topology 2010, reduce load of some
elements in southern part of Serbia.
4.36
Compared to the 2010 topology, voltage profile is somewhat better, especially in southern and
central part of Serbia, as well as in Albania.
Realization of the planed investments till 2015 has impact on secure operation of the network.
Some of the insecure states identified with 2010 topology are relieved and do not exist with
expected 2015 network topology. Also, level of overloadings due to outages is decreased.
All in all overall network performance is better, especially in the region where new planed
investments are to be realized (Albania, southern Serbia and Montenegro).
4.37
4.6 Scenario 2015 – dry hydrology – 2010 topology
This part of the Study presents the results of static load flow and voltage profile analyses that are
conducted for complete network topology and (n-1) contingencies in the scenario which is denoted
as GTmax run for year 2015 - dry hydrology and expected network topology for 2010.
4.6.1 Lines loadings
Figure 4.6.1 shows power exchanges between areas for 2015-dry hydrology scenario. Power flows
along interconnection lines in the region together with balances of the systems are shown in Figure
4.6.2. Area totals are shown in Table 4.6.1. Figure 4.6.3 shows histogram of tie lines loadings. It is
concluded that most of the tie lines are loaded less than 25% of their thermal limits.
UKR
37
1
18
AUT
3
41
450
SVK
HUN
10
3
43
73
31
SLO
8
25
ITA
CRO
ROM
207
4
69
70
366
SRB
91
BIH
6
15
0
23
229
MNG
8
27
20
383
BUL
82
13
ALB
143
48
MKD
20
TUR
0
36
GRE
Figure 4.6.1 - Area exchanges in analyzed electric power systems for 2015-dry hydrology scenario – topology 2010
Table 4.6.1 - Area totals in analyzed electric power systems for 2015-dry hydrology scenario – topology 2010
AREA
GENERATION
LOAD
LOSSES
INTERCHANGE
ALBANIA
894.7
1521.9
92.7
-720
BULGARIA
7363.9
6450
147.9
766
BIH
3140.9
2304
78.2
758.8
CROATIA
2636.7
3665
58.2
-1086.4
MACEDONIA
985.1
1410
19.1
-444
ROMANIA
7724.6
7798.4
295.5
-369.2
SERBIA
8397.9
7296
237.9
864
MONTENEGRO
586.3
672
22.1
-107.7
TOTALS
31730.1
31117.3
951.6
-338.5
4.38
LEVIC 1400.0
CENTREL
200
140
251
113
GABICK1400.0
450.0 MW
223.5 MW
106
24.2
405.4
OKAINA1 401.5
234.2
39.1
47.0
450.0 MW
224.4
UMUKACH
412.0
37.
70. 1
5
ROM
400.9
230 SOFIA_W4
92.4
411.0
76.1
35.8
103
43.5
JHPERU21
102 221.1
45.0
148
147
147
JPODG211
181
JHPIVA21
236.0
CRG
-107.7 MW
384.0
JPRIZ22214.1
347
20.0
7
18. 0
17.
204
180
373.2AVDEJA2
210.7 AFIERZ2208.0
SK 1
222.9
BITOLA 2
C_MOGILA
407.5SK 4
MKD
410.5
414.2
AZEMLA1
357
271
KARDIA K
364 414.9
257
QES/H
5.0
14.0
AHS_FLWR416.1
K
FILIPPOI
8
17 7.2
7
415.5
GIS REGIONAL MODEL - DRY HYDRO
2015 TOPOLOGY 2010
KV: 110 , 220 , 400
11
13 .3
.8
19.8
21.7
K-KEXRU
417.8
8.5
63.7
19.8
4BABAESK
70.7
415.5
4HAMITAX
8.5 415.4
75.0
412.2
0.0 MW
0.2 MW
KARACQOU
250 418.1
50.0
419.3
.3 1
11 9.
8
1
18. 7
49.
.4
35 .4
49
LAGAD K 413.3
381.4
MI_3_4_1
BLAGOEV413.3
410.7
5.0
48.0
1
18. 1
72.
ALB
-720.0 MW
412.8
.9
47 .8
59
STIP 1
DUBROVO
358.9
.0
48 .0
34
407.7
5 9
8. 6.
1
202
204
766.0 MW
JVRAN31
0.000
-444.0 MW
AKASHA1
DOBRUD4 417.1
8
28 1.
.5 9
72. 8
50.8
JTKOSA2 227.0
JTKOSB1
412.4
414.7
BUL
JLESK21
0.000
JPODG121219.0
7
18. 0
17.
204
180
AEC_400
JNIS2 1
396.5
21 4
9.8
139
26.3
AVDEJA1
SHAW POWER
TECHNOLOGIES
INC. R
413.7
8.5
63.7
138
29.5
756.2 MW
SRB
864.0 MW
8
25 2.
.1 8
RP TREB
396.9
37.3
54.0
345
48.8
228.9
228
145
TANTAREN
406.0
SCG
JVARDI22
229.6
37.2
51.9
18.6
27.5
RP TREB
69.5
71.5
400.9
18.6
27.5
SA 20
230.5
P.D.FIE
69.5
72.5
53.9
16.2
73.7
47.1
BIH
758.8 MW
212
121
-369.2 MW
JSUBO31 396.0
208
JSMIT21
85.1
231.7
VISEGRA
231.4
ISACCEA 405.4
.1
76 .0
14
404.6
130 MO-4
19.2
232.4
TE TUZL
38.0 ARAD
43.0
401.5
.1
76 .0
14
MO-4
220.1
UGLJEVIK
ROSIORI 402.1
402.4
38.0
16.6
JHDJE11
225.5
64.8
NADAB
71.6
402.2
30.7
165
JSOMB31
393.4
207
40.3
404.5
153
4
81.
127
18.2
319
24.0
GRADACAC
64.7
52.7
76.1
35.8
222.5
2
11 0
.
46
317
16.4
2
60.
7
22.
4 .9
1.1
.9
34 .3
34
PRIJED2
231.4
39.0
71.8
UKR
UMUKACH2
179
45.
5
4.9
6.8
34.7
26.6
219.6
CRO
-1086.4 MW
402.2
10.6
29.2
MSAFA 4
405.9
HE ZAKUC
10.8
7.6
.5
30 41
1
71
37 .7
.8
DAKOVO
222.3 MEDURIC
ERNESTIN
226.9
401.0
MRACLIN
KONJSKO
35.6
27.7
35.3
45.4
10.7
7.6
ZERJAVIN
6
15 0
2.
TUMBRI
229.2
35.4
21.9
35.4
21.9
402.1
6
15 0
2.
PEHLIN
.3
.3
35 .2 35 .2
85
85
ZERJAVIN
403.4
5
6. .7
5
MELINA
408.2
5
24 4.0
.0
10
2. .7
6
227.1
205
165
229.3
LCIRKO2
401.0
71
37 .7
.8
11.4
77.1
MPECS 4
LKRSKO1
71.
8 8
71 5.8
85 .8
.8
11.4
73.6
SLO
97.8
7.2
230.3
35.5
20.9
MBEKO 4
404.6
153
29.5
IPDRV121
MSAJO 4
410.0
408.3
110
47.0
LDIVAC2
MHEVI 4
215.0 MW
97
30 . 7
.7
228.9
LMARIB1 402.4
59.9
27.1
400.9
230.5
6.5
4.2
180
24.1
22.9
26.6
MTLOK 2
MGYOR 2
HUN
156
26.9
180
3.3
23.0
12.7
MKISV 2
226.3
407.6
350
UZUKRA01
297
772.5
-1250.0 MW
156
26.9
400.9
52.2
20.7
39.
5
10.
5
154
45.
7
IRDPV111
52.7
5.8
MSAJI 4408.1
4
346
804
8
36.
131
39.
10. 5
5
233.5
LPODLO2228.9
0
2 5 .4
90
403.3
MGOD
ONEUSI2
LDIVAC1
105
101
OWIEN 2
39.5
41.8
156
49.2
39.5
41.8
OOBERS2238.7
MAISA 7
714.8
MGYOR 4
199
94.6
UCTE
OWIEN 1
GRE
TUR
BUS -VOLTAGE(KV)
BRANCH -MW/MVAR
EQUIPMENT -MW/MVAR
Figure 4.6.2 - Power flows along interconnection lines in the region with balances of the systems for 2015-dry hydrology scenario – 2010 topology
4.39
Histogram
40
35
Frequency
30
25
20
15
10
5
0
x<25
25<x<50
50<x<75 75<x<100
x>100
Bin
Figure 4.6.3 - Histogram of interconnection lines loadings for 2015-dry hydrology scenario – topology 2010
("Frequency" denotes number of lines and "Bin" denotes loading range in % of thermal limit)
Following Table 4.6.2 lists all network elements loaded over 80% of their thermal limits. As it can
be seen some lines 220 kV voltage level in Albania, Romania and Serbia are loaded over 80%.
Also, the most of the elements loaded over 80% are transformers in some substations, again, in
Albania, Bosnia & Herzegovina, Romania and Serbia. Figure 4.6.4 shows histogram of branch
loadings in the system.
As for the conclusion regarding thermal loadings in this scenario it can be said that the most of the
network elements are loaded less then 75% of their thermal limits, but there are some elements
highly loaded. Most of the elements loaded over 80% are transformers in some substations, so some
internal network reinforcements are necessary to sustain this load-demand level and production
pattern. There are some elements that are overloaded (220 kV line Kashar – Rashbul and
transformers 220/110 kV in substation Fierza and one transformer 220/110 kV in substation
Elbasan 1in Albania). This leads to conclusion that transmission network is not able to sustain this
load-demand level and this production pattern needs reinforcement as necessary. It should be
pointed out that it is expected that transformers in substation Fierza will be replaced with more
powerful transformer units.
4.40
Table 4.6.2 - Network elements loaded over 80% of their thermal limits for 2015-dry hydrology scenario – 2010
topology
BRANCH LOADINGS ABOVE
80.0 % OF RATING:
LOADING
MVA
Lines
HL 220kV AKASHA2-ARRAZH2 1
275.3
HL 220kV BUC.S-B-FUNDENI 1
261.2
HL 220kV JBGD3 21-JOBREN2 1
293.6
Transformers
TR 220/110 kV AELBS1 1
84.7
TR 220/110 kV AELBS1 2
84.7
TR 220/110 kV AELBS1 3
91
TR 220/110 kV AFIER 1
146.8
TR 220/110 kV AFIER 2
120.3
TR 220/110 kV AFIER 3
114.8
TR 220/110 kV AFIERZ 1
59
TR 220/110 kV AFIERZ 2
59
TR 220/110 kV AKASHA 1
90.8
TR 220/110 kV AKASHA 2
90.8
TR 220/110 kV ARRAZH 1
84.7
TR 220/110 kV ARRAZH 2
84.7
TR 220/110 kV ATIRAN 3
98.2
TR 400/110 kV UGLJEV 1
242.4
TR 220/110 kV FUNDE2 1
199.9
TR 220/110 kV FUNDEN 1
168.5
TR 400/220 kV IERNUT 1
325.2
TR 220/110 kV JBGD3 1
171.1
TR 220/110 kV JBGD3 2
129.8
TR 220/110 kV JPRIS4 1
121.6
TR 220/110 kV JPRIS4 2
121.6
TR 220/110 kV JTKOSA 2
133.3
TR 220/110 kV JTKOSA 3
135.7
TR 400/220 kV JTKOSB 1
323.1
TR 400/220 kV JTKOSB 2
337.6
TR 400/220 kV JTKOSB 3
337.6
AREA
ELEMENT
ALB
ROM
SRB
ALB
BIH
ROM
SRB
RATING
MVA
PERCENT
270
320
301
102
81.6
97.5
90
90
90
120
90
90
60
60
100
100
100
100
120
300
200
200
400
200
150
150
150
150
150
400
400
400
94.1
94.1
101.1
122.3
133.7
127.6
98.3
98.3
90.8
90.8
84.7
84.7
81.9
80.8
99.9
84.2
81.3
85.6
86.5
81.1
81.1
88.9
90.5
80.8
84.4
84.4
Histogram
350
300
Frequency
250
200
150
100
50
0
x<25
25<x<50 50<x<75 75<x<100
x>100
Bin
Figure 4.6.4 - Histogram of branch loadings for 2015-dry hydrology scenario – 2010 network topology ("Frequency"
denotes number of lines and "Bin" denotes loading range in % of thermal limit)
4.6.2 Voltage Profile in the Region
Figure 4.6.5 shows histogram of voltages in monitored substations. Voltages in almost all
monitored substations are found within permitted limits. Only in substations Elbasan and Kashar
voltages are below limits (around 359 kV), but this can be resolved with changing of the setting of
the tap changing transformers in these substations. Also, as a consequence of high imports, voltages
in network of Albania are very near low limits.
4.41
Histogram
120
Frequency
100
80
60
40
20
x>1.1
1.05<x<1.1
1.025<x<1.05
1<x<1.025
0.975<x<1
0.95<x<0.975
0.9<x<0.95
x<0.9
0
Bin
Figure 4.6.5 - Histogram of voltages in monitored substations for 2015-dry hydrology scenario – 2010 network
topology ("Frequency" denotes number of busses and "Bin" denotes voltage range in p.u.)
In the rest of the monitored network the voltage profile is satisfying and that most of the substations
have magnitudes in range 0.975-1.05 p.u.
4.6.3 Security (n-1) analysis
Results of security (n-1) analysis for 2015-dry hydrology scenario and expected topology for 2010
are presented in Table 4.6.3. Figure 4.6.6 shows the geographical position of the critical elements in
monitored systems.
It can be concluded that all identified insecure situations are located in internal networks that belong
to monitored power systems of Albania, Romania and Serbia. The most critical element in Albanian
system is 220 kV line Elbasan-Fierza. In most critical case in Romanian system, the critical
elements are transformers 400/110 kV in substations Dirste and Brasov. The most critical element
in Serbian system are line 220 kV Obrenovac – Beograd 3.
Some of the overloadings identified can be relieved by dispatch actions (splitting busbars, changing
lower voltage network topology in order to redistribute load-demand or change of generation units
engagement), like in the case of most severe overloading in Romanian network happens on
transformer 400/110 kV Dirste when transformer 400/110 kV in Brasov is outaged, but this is a
consequence of the fact that second transformer unit 400/110 kV in Brasov is out of operation.
Switching on of this transformer clears this critical outage. The similar situation is in case of outage
of the 400 kV line Obrenovac – Beograd 8 in Serbia. Splitting off the 220 kV busbars in substation
Beograd 3 relieves this overloading, but voltage profile in Serbian network remains critical, so
additional dispatching actions are necessary too.
All in all, certain reinforcement of internal network is necessary in order to make this regime more
secure. None of the identified congestions is located at border lines.
4.42
Table 4.6.3 - Network overloadings for 2015-dry hydrology scenario, single outages – 2010 network topology
Area
1
contingency
2
AL
AL
AL
AL
RO
CS
CS
CS
CS
CS
CS
CS
CS
BASE CASE
OHL 220kV
OHL 220kV
OHL 220kV
OHL 220kV
OHL 220kV
OHL 220kV
OHL 220kV
OHL 220kV
OHL 220kV
OHL 220kV
OHL 220kV
OHL 220kV
OHL 220kV
CS
OHL 400kV JBGD8 1 -JOBREN11 1
CS
CS
CS
CS
OHL
OHL
OHL
OHL
CS
OHL 400kV JTKOSB1 -JPEC
RO
RO
CS
TR 400/110 BRASOV
1
TR 400/110 DIRSTE
1
TR 400/110 JNIS2 1
AL
TR 400/220 AELBS2 1
RO
RO
RO
TR 400/220 BRAZI
TR 400/220 BUC.S
TR 400/220 IERNUT
CS
TR 400/220 JBGD8
CS
CS
CS
TR 400/220 JBGD8 2
TR 400/220 JNIS2 1
TR 400/220 JPANC2 1
CS
TR 400/220 JTKOSB 1
RO
RO
TR 400/220 MINTIA
TR 400/220 ROSIORI
400kV
400kV
400kV
400kV
AFIERZ2 -ABURRE2
ATIRAN2 -AKASHA2
AELBS12 -AFIER 2
AFIER 2 -ARRAZH2
FUNDENI -BUC.S-B
JBBAST2 -JBGD3 21
JBGD172 -JBGD8 22
JBGD3 22-JBGD8 21
JGLOGO2 -JPRIZ22
JHIP 2 -JPANC22
JNSAD32 -JOBREN2
JNSAD32 -JZREN22
JOBREN2 -JSABA32
JBGD8 1
JHDJE11
JKRAG21
JPANC21
-JBGD201
-JTDRMN1
-JTKOLB1
-JTDRMN1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
A
1
A
1
A
1
1
1
1
1
1
overloadings
/
Area
out of limits voltages
3
4
AL
HL 220kV AKASHA2-ARRAZH2
AL
HL 220kV AKASHA2-ARRAZH2
AL
HL 220kV AKASHA2-ARRAZH2
AL
HL 220kV AKASHA2-ARRAZH2
AL
HL 220kV AELBS12-AFIER 2
RO
HL 220kV BUC.S-B-FUNDENI
CS
HL 220kV JBGD3 21-JOBREN2
CS
HL 220kV JBGD172-JBGD8 22
CS
HL 220kV JBGD3 21-JOBREN2
AL
HL 220kV AKASHA2-ARRAZH2
CS
HL 220kV JBGD3 21-JOBREN2
CS
HL 220kV JBGD3 21-JOBREN2
CS
HL 220kV JBGD3 21-JOBREN2
CS
HL 220kV JBGD3 21-JOBREN2
CS
HL 220kV JBGD3 21-JOBREN2
CS
HL 220kV JBGD3 22-JBGD8 22
CS
HL 220kV JBGD3 21-JOBREN2
CS
HL 220kV JBGD3 21-JOBREN2
CS
HL 220kV JBGD3 21-JOBREN2
CS
HL 220kV JBGD3 21-JOBREN2
CS
TR 400/220kV/kV JTKOSB1
CS
TR 400/220kV/kV JTKOSB1
CS
TR 400/220kV/kV JTKOSB1
RO
TR 400/110kV/kV DIRSTE
RO
TR 400/110kV/kV BRASOV
CS
TR 400/110kV/kV JNIS2 1
AL
TR 400/220kV/kV AELBS22
AL
HL 220kV AKASHA2-ARRAZH2
RO
HL 220kV BUC.S-B-FUNDENI
RO
TR 400/220kV/kV BUC.S
RO
HL 220kV STEJARU-GHEORGH
CS
HL 220kV JBGD3 21-JOBREN2
CS
HL 220kV JBGD3 22-JBGD8 22
CS
HL 220kV JBGD3 21-JOBREN2
CS
TR 400/110kV/kV JNIS2 1
CS
HL 220kV JBGD3 21-JOBREN2
CS
TR 400/220kV/kV JTKOSB1
CS
TR 400/220kV/kV JTKOSB1
RO
HL 220kV PESTIS-MINTIA A
RO
TR 400/220kV/kV IERNUT
#
5
1
1
1
1
1
1
1
2
1
1
1
1
1
1
1
2
1
1
1
1
1
2
3
1
1
2
2
1
1
2
1
1
2
1
1
1
2
3
1
1
limit
/
Unom
6
270MVA
270MVA
270MVA
270MVA
270MVA
320MVA
301MVA
365.8MVA
301MVA
270MVA
301MVA
301MVA
301MVA
301MVA
301MVA
365.8MVA
301MVA
301MVA
301MVA
301MVA
400MVA
400MVA
400MVA
250MVA
250MVA
300MVA
300MVA
270MVA
320MVA
400MVA
208.1MVA
301MVA
365.8MVA
301MVA
300MVA
301MVA
400MVA
400MVA
277.4MVA
400MVA
Flow
rate
/
/
Voltage volt.dev.
7
8
242.3MVA
102.0%
261.1MVA
126.2%
254.3MVA
106.6%
361.1MVA
180.3%
194.4MVA
104.8%
342.8MVA
106.8%
327.8MVA
111.5%
466.5MVA
133.2%
312.8MVA
106.4%
231.3MVA
128.2%
304.5MVA
103.4%
311.5MVA
105.2%
312.6MVA
106.1%
314.3MVA
106.6%
511.9MVA
184.1%
354.7MVA
106.4%
306.4MVA
103.9%
324.4MVA
110.4%
306.5MVA
103.9%
314.9MVA
107.7%
413.9MVA
103.5%
432.5MVA
108.1%
432.5MVA
108.1%
401MVA
160.4%
394.4MVA
157.7%
363.8MVA
121.3%
307.5MVA
102.5%
259.2MVA
132.1%
346.4MVA
109.9%
506.1MVA
126.5%
191.8MVA
108.6%
342.9MVA
117.8%
385.9MVA
110.9%
320.8MVA
109.4%
303.3MVA
101.1%
300.3MVA
101.8%
471.5MVA
117.9%
471.5MVA
117.9%
324.3MVA
111.1%
425.1MVA
106.3%
SVK
AUT
HUN
SLO
ROM
CRO
SRB
BIH
MNG
BUL
MKD
TUR
ALB
critical elements
GRE
400 kV line or 400/x kV transformer
220 kV line or 220/x kV transformer
Figure 4.6.6 – Geographical position of critical elements for 2015-dry hydrology scenario – topology 2010
4.43
4.7 Scenario 2015 – dry hydrology – topology 2015
This part of the Study presents the results of static load flow and voltage profile analyses that are
conducted for complete network topology and (n-1) contingencies in the scenario which is denoted
as GTmax run for year 2015 - dry hydrology and expected network topology for 2015.
4.7.1 Lines loadings
Figure 4.7.1 shows power exchanges between areas for 2015-dry hydrology scenario. Power flows
along interconnection lines in the region together with balances of the systems are shown in Figure
4.7.2. Area totals are shown in Table 4.7.1. Figure 4.7.3 shows histogram of tie lines loadings. It is
concluded that most of the tie lines are loaded less than 25% of their thermal limits.
UKR
44
0
19
AUT
6
40
450
SVK
HUN
81
33
94
29
SLO
8
24
ITA
CRO
ROM
199
3
73
78
328
SRB
63
BIH
89
221
52
MNG
BUL
18
304
33
9
35
1
24
291
ALB
145
90
MKD
18
TUR
86
GRE
Figure 4.7.1 - Area exchanges in analyzed electric power systems for 2015-dry hydrology scenario
As it can be seen, new elements that are expected to be build till 2015 cause totally different
distribution of power flows in the southern part of the region (Albania, FYR of Macedonia, Serbia
and Montenegro).
Compared to the expected topology 2010, analyzed in previous chapter, it can be seen that the
network losses are decreased as a consequence of building of new elements for 2015 network
topology, especially in the cases of Albania, Serbia and Montenegro. Overall reduction of loses is
around 40 MW.
4.44
LEVIC 1400.0
CENTREL
200
141
251
113
GABICK1400.0
450.0 MW
222.8 MW
112
24.7
405.4
OKAINA1 401.6
234.2
JSOMB31
393.8
224.4
UMUKACH
412.0
44.
66. 3
7
ROM
220
96.1
221 SOFIA_W4
38.3
412.5
61.0
38.2
205
122
414.3
104
JPODG211
86.1
JPODG121229.3
2 4.2
21.2
1
1. 5.
1 5
343
109
402.0AVDEJA2
226.7 AFIERZ2224.6
143
100
9.2
29.1
143
101
374
67.8
3
28. 1
27.
342
127
JTKOSA2 232.1
JTKOSB1
419.1
3
28. 1
27.
15
8. .6
3
402.7
JPRIZ22224.5
SK 1
226.2
410.6SK 4
-444.0 MW
MKD
143
139
393.7
C_MOGILA
.2
90 .1
38
411.9
414.4
DUBROVO
60.5
19.4
QES/H
K
415.8
24
90 1
.9
KV: 110 , 220 , 400
10
12 .2
.6
1
18 6.3
7
17.8
22.1
K-KEXRU
417.9
7.6
64.8
17.7
4BABAESK
71.1
415.6
4HAMITAX
7.6 415.4
76.1
413.6
0.0 MW
-0.1 MW
KARACQOU
250 419.6
50.0
FILIPPOI
419.5
.2 5
10 0.
9
.1
3 6 .3
50
LAGAD K 414.4
AHS_FLWR417.3
KARDIA K
86.6 417.8
43.8
MI_3_4_1
BLAGOEV414.5
412.0
231
8
65.
86.1
114
414.0
.0
90 .7
54
STIP 1
408.2
AZEMLA1
766.0 MW
JVRAN31
407.4
ALB
-720.0 MW
BUL
JLESK21
405.0
410.0
230 4
84.
AKASHA1
242
3
51.
DOBRUD4 417.4
415.4
6 8
7. 5.
1
143
139
BITOLA 2
206
9. 9
104
42.5
AEC_400
JNIS2 1
403.5
7.6
64.8
JHPERU21
85.7 229.7
19.1
CRG
-108.0 MW
79.8 79.8
39.8 61.1
86.4
14.2
JHPIVA21
238.3
26.6 26.5
20.2 58.8
136
31.7
124
48.3
135
35.7
756.0 MW
SRB
864.0 MW
1
3. 24
2
30.5
47.2
AVDEJA1
GIS REGIONAL MODEL - DRY HYDRO
2015 TOPOLOGY 2015
TANTAREN
407.8
SCG
JVARDI22
231.1
30.5
44.9
60.5
42.2
233.6
77.6
54.8
401.8
372
93.9
RP TREB
P.D.FIE
77.6
55.8
32.7
15.6
28.0
37.4
SA 20
232.1
ISACCEA 405.7
-368.9 MW
JSUBO31 396.4
28.0
37.4
408.1
133 MO-4
15.1
234.1
BIH
759.0 MW
27.6 ARAD
40.6
402.0
401.7
232.1
VISEGRA
232.6
402.7
199
JSMIT21
80.8
24.3
12.1
MO-4
TE TUZL
ROSIORI 402.4
.0
61 .0
12
131
14.5
333
7.2
UGLJEVIK
225.8
53.3
NADAB
68.5
402.6
.0
61 .0
12
330
32.3
GRADACAC
220.5
53.2
49.4
27.6
19.3
JHDJE11
RP TREB
407.8
SHAW POWER
TECHNOLOGIES
INC. R
450.0 MW
61.0
38.2
198
35.1
404.9
162
5
76.
232.3
UKR
UMUKACH2
182
46.
2
222.8
4
11 2
.
46
7.1
0. 7
.0
39 .4
33
PRIJED2
2
62.
8
22.
CRO
-1085.9 MW
404.1
31.9
46.5
MSAFA 4
407.7
HE ZAKUC
31.8
72.5
.2
29 38
1
76
35 .7
.1
219.9
7.1
7.3
38.7
25.7
408.4
76
35 .7
.1
DAKOVO
222.5 MEDURIC
ERNESTIN
227.0
401.1
MRACLIN
KONJSKO
10.6
29.2
36.1
45.0
12.6
7.1
ZERJAVIN
6
15 2
3.
TUMBRI
229.3
10.8
7.6
29.4
162
402.3
6
15 2
3.
PEHLIN
.5
.5
29 .3 29 .3
85
85
ZERJAVIN
403.6
2
6. .3
5
MELINA
29.6
22.0
29.6
22.0
227.2
3
25 2.8
.0
12
3. . 6
1
LCIRKO2
401.1
9.2
0.1
229.4
MPECS 4
LKRSKO1
76.
8 8
76 3.2
83 .8
.2
12.2
76.3
SLO
105
9.5
230.3
12.2
72.8
215.0 MW
10
32 5
.9
35.6
27.7
MBEKO 4
404.9
408.4
161
24.8
IPDRV121
MHEVI 4
112
46.9
LDIVAC2
MSAJO 4
410.0
LMARIB1 402.5
61.8
27.1
401.0
229.0
6.2
4.6
183
21.4
230.5
HUN
156
28.1
182
0.7
23.9
26.3
MTLOK 2
35.5
20.9
-1250.0 MW
156
28.1
401.0
53.4
20.3
24.1
12.5
MKISV 2
226.3
407.7
MGYOR 2
37.
0
10.
7
153
45.
6
IRDPV111
54.0
5.6
MSAJI 4408.1
4
350
UZUKRA01
297
772.5
0
44.
128
37.
10. 0
7
233.5
LPODLO2228.9
403.3
MGOD
ONEUSI2
LDIVAC1
0
25 .8
90
OWIEN 2
37.0
41.9
155
48.9
37.0
41.9
OOBERS2238.7
112
100
346
804
MAISA 7
714.8
MGYOR 4
199
95.0
UCTE
OWIEN 1
GRE
TUR
BUS -VOLTAGE(KV)
BRANCH -MW/MVAR
EQUIPMENT -MW/MVAR
Figure 4.7.2 - Power flows along interconnection lines in the region with balances of the systems for 2015-dry hydrology scenario
4.45
Table 4.7.1 - Area totals in analyzed electric power systems for 2015-dry hydrology scenario
AREA
GENERATION
LOAD
LOSSES
INTERCHANGE
ALBANIA
893.6
1542
71.6
-720
BULGARIA
7363.9
6450
147.8
766
BIH
3141.2
2305
77.1
759
CROATIA
2637.7
3665
58.7
-1085.9
MACEDONIA
985.2
1409
20.2
-444
ROMANIA
7722.9
7798.4
293.5
-368.9
SERBIA
8396.3
7308
224.3
864
MONTENEGRO
586.3
678
16.4
-108
TOTALS
31727.1
31155.4
909.6
-337.8
Histogram
45
40
Frequency
35
30
25
20
15
10
5
0
x<25
25<x<50
50<x<75 75<x<100
x>100
Bin
Figure 4.7.3 - Histogram of interconnection lines loadings for 2015-dry hydrology scenario ("Frequency" denotes
number of lines and "Bin" denotes loading range in % of thermal limit)
Following Table 4.7.2 lists all network elements loaded over 80% of their thermal limits. Figure
4.7.4 shows histogram of branch loadings in the system.
Table 4.7.2 - Network elements loaded over 80% of their thermal limits for 2015-dry hydrology scenario
BRANCH LOADINGS ABOVE
AREA
ALB
ROM
SRB
ALB
BIH
ROM
SRB
80.0 % OF RATING:
LOADING
MVA
Lines
HL 220kV AKASHA2-ARRAZH2 1
242.7
HL 220kV BUC.S-B-FUNDENI 1
260.6
HL 220kV JBGD3 21-JOBREN2 1
294.6
Transformers
TR 220/110 kV AELBS1 1
75.4
TR 220/110 kV AELBS1 2
75.4
TR 220/110 kV AELBS1 3
81
TR 220/110 kV AFIER 1
136.9
TR 220/110 kV AFIER 2
112.2
TR 220/110 kV AFIER 3
107.1
TR 220/110 kV AFIERZ 1
52.1
TR 220/110 kV AFIERZ 2
52.1
TR 220/110 kV AKASHA 1
83.5
TR 220/110 kV AKASHA 2
83.5
TR 400/110 kV UGLJEV 1
240.7
TR 220/110 kV FUNDE2 1
199.6
TR 220/110 kV FUNDEN 1
168.2
TR 400/220 kV IERNUT 1
324.2
TR 220/110 kV JBGD3 1
170.9
TR 220/110 kV JBGD3 2
130
ELEMENT
RATING
MVA
PERCENT
270
320
301
89.9
81.4
97.9
90
90
90
120
90
90
60
60
100
100
300
200
200
400
200
150
83.8
83.8
90
114.1
124.7
119
86.9
86.9
83.5
83.5
80.2
99.8
84.1
81.1
85.5
86.6
4.46
Histogram
350
300
Frequency
250
200
150
100
50
0
x<25
25<x<50 50<x<75 75<x<100
x>100
Bin
Figure 4.7.4 - Histogram of branch loadings for 2015-dry hydrology scenario ("Frequency" denotes number of lines and
"Bin" denotes loading range in % of thermal limit)
As it can be seen from these outputs, most of the network elements are loaded less then 75% of
their thermal limits and most of the elements loaded over 80% are transformers in some substations,
so some internal network reinforcements are necessary to sustain this load-demand level and
production pattern. There are some elements that are overloaded (transformers 220/110 kV in
substation Fierza in Albania). This leads to conclusion that transmission network is not able to
sustain this load-demand level and this production pattern needs reinforcement as necessary. It
should be pointed out that it is expected that transformers in substation Fierza will be replaced with
more powerful transformer units. Also, planned network reinforcements compared to network
topology 2010, reduce loading of some elements in southern part of Serbia.
4.7.2 Voltage Profile in the Region
Figure 4.7.5 shows histogram of voltages in monitored substations. Voltages in all monitored
substations are found within permitted limits. It is concluded that voltage profile is satisfying and
that most of the substations have magnitudes in range 1-1.05 p.u. Compared to the 2010 topology,
voltage profile is somewhat better, especially in southern and central part of Serbia, as well as in
Albania.
4.47
Histogram
100
90
Frequency
80
70
60
50
40
30
20
10
x>1.1
1.05<x<1.1
1.025<x<1.05
1<x<1.025
0.975<x<1
0.95<x<0.975
0.9<x<0.95
x<0.9
0
Bin
Figure 4.7.5 - Histogram of voltages in monitored substations for 2015-dry hydrology scenario ("Frequency" denotes
number of busses and "Bin" denotes voltage range in p.u.)
4.7.3 Security (n-1) analysis
Results of security (n-1) analysis for 2015-dry hydrology scenario are presented in Table 4.7.3.
Figure 4.7.6 shows the geographical position of the critical elements in monitored systems.
Like for expected topology 2010 (previous chapter), it can be concluded that all identified insecure
situations are located in internal networks that belong to monitored power systems of Albania,
Romania and Serbia. Also, the planned network reinforcements till 2015 resolve some of the
noticed critical contingencies, especially in southern part of Serbia. The rest of the conclusions are
the same as in case of the analyzed topology 2010, and that is that certain level of network
reinforcement is necessary to make this regime more secure.
4.48
SVK
AUT
HUN
SLO
ROM
CRO
SRB
BIH
MNG
BUL
MKD
TUR
ALB
GRE
critical elements
400 kV line or 400/x kV transformer
220 kV line or 220/x kV transformer
Figure 4.7.6 – Geographical position of critical elements for 2015-dry hydrology scenario
4.49
Table 4.7.3 - Network overloadings for 2015-dry hydrology scenario, single outages
Area
1
AL
RO
RO
RO
CS
CS
CS
CS
CS
CS
CS
contingency
2
220kV AELBS12 -AFIER 2
220kV FUNDENI -BUC.S-B
220kV STEJARU -GHEORGH
400kV DOMNESTI-BRAZI
220kV JBBAST2 -JBGD3 21
220kV JBGD172 -JBGD8 22
220kV JBGD3 22-JBGD8 21
220kV JHIP 2 -JPANC22
220kV JNSAD32 -JOBREN2
220kV JNSAD32 -JZREN22
220kV JOBREN2 -JVALJ32
OHL
OHL
OHL
OHL
OHL
OHL
OHL
OHL
OHL
OHL
OHL
CS
OHL 400kV JBGD8 1 -JOBREN11 1
CS
CS
CS
CS
AL
RO
RO
RO
RO
RO
OHL 400kV JBGD8 1
OHL 400kV JHDJE11
OHL 400kV JKRAG21
OHL 400kV JPANC21
TR 400/110 AZEMLA
TR 400/110 BRASOV
TR 400/110 DIRSTE
TR 400/220 BRAZI
TR 400/220 BUC.S
TR 400/220 IERNUT
-JBGD201
-JTDRMN1
-JTKOLB1
-JTDRMN1
1
1
1
1
1
1
CS
TR 400/220 JBGD8
1
CS
CS
RO
RO
TR
TR
TR
TR
400/220
400/220
400/220
400/220
JBGD8 2
JPANC2 1
MINTIA
1
ROSIORI 1
1
1
1
1
1
1
1
1
1
1
1
A
1
A
1
overloadings
/
Area
out of limits voltages
3
4
AL
HL 220kV AKASHA2-ARRAZH2
RO
HL 220kV BUC.S-B-FUNDENI
RO
TR 400/220kV/kV IERNUT
RO
HL 220kV BUC.S-B-FUNDENI
CS
HL 220kV JBGD3 21-JOBREN2
CS
HL 220kV JBGD172-JBGD8 22
CS
HL 220kV JBGD3 21-JOBREN2
CS
HL 220kV JBGD3 21-JOBREN2
CS
HL 220kV JBGD3 21-JOBREN2
CS
HL 220kV JBGD3 21-JOBREN2
CS
HL 220kV JBGD3 21-JOBREN2
CS
HL 220kV JBGD3 21-JOBREN2
CS
HL 220kV JBGD3 22-JBGD8 22
CS
HL 220kV JBGD3 21-JOBREN2
CS
HL 220kV JBGD3 21-JOBREN2
CS
HL 220kV JBGD3 21-JOBREN2
CS
HL 220kV JBGD3 21-JOBREN2
AL
HL 220kV AKASHA2-ARRAZH2
RO
TR 400/110kV/kV DIRSTE
RO
TR 400/110kV/kV BRASOV
RO
HL 220kV BUC.S-B-FUNDENI
RO
TR 400/220kV/kV BUC.S
RO
HL 220kV STEJARU-GHEORGH
CS
HL 220kV JBGD3 21-JOBREN2
CS
HL 220kV JBGD3 22-JBGD8 22
CS
HL 220kV JBGD3 21-JOBREN2
CS
HL 220kV JBGD3 21-JOBREN2
RO
HL 220kV PESTIS-MINTIA A
RO
TR 400/220kV/kV IERNUT
#
5
1
1
1
1
1
2
1
1
1
1
1
1
2
1
1
1
1
1
1
1
1
2
1
1
2
1
1
1
1
limit
/
Unom
6
270MVA
320MVA
400MVA
320MVA
301MVA
365.8MVA
301MVA
301MVA
301MVA
301MVA
301MVA
301MVA
365.8MVA
301MVA
301MVA
301MVA
301MVA
270MVA
250MVA
250MVA
320MVA
400MVA
208.1MVA
301MVA
365.8MVA
301MVA
301MVA
277.4MVA
400MVA
Flow
rate
/
/
Voltage volt.dev.
7
8
376.7MVA
151.9%
342.5MVA
106.6%
406.2MVA
101.5%
330.3MVA
102.0%
331MVA
112.3%
467.1MVA
133.0%
314MVA
106.6%
305.8MVA
103.6%
313.5MVA
105.6%
314.5MVA
106.5%
301.6MVA
101.5%
514.9MVA
184.3%
359.3MVA
107.2%
308.3MVA
104.2%
324.2MVA
110.1%
307.8MVA
104.0%
316MVA
107.8%
263.6MVA
108.6%
401.1MVA
160.4%
394.7MVA
157.9%
346.3MVA
109.8%
506MVA
126.5%
190.9MVA
107.9%
344.1MVA
117.9%
388.2MVA
111.3%
322MVA
109.5%
301.9MVA
102.0%
325.2MVA
111.3%
425.1MVA
106.3%
4.7.4 Summary of Impacts - 2015 topology versus 2010 topology
Compared to the expected topology 2010, analyzed in previous chapter, it can be seen that the
network losses are reduced as a consequence of building of new elements for 2015 network
topology, especially in the cases of Albania, Serbia and Montenegro. Overall reduction of losses is
around 40 MW.
Also, planned network reinforcements compared to network topology 2010, reduce load of some
elements in southern part of Serbia.
Compared to the 2010 topology, voltage profile is somewhat better, especially in southern and
central part of Serbia, as well as in Albania.
Realization of the planed investments till 2015 has impact on secure operation of the network.
Some of the insecure states identified with 2010 topology are relieved and do not exist with
expected 2015 network topology. Also, levels of overloadings due to outages are decreased.
All in all overall network performance is better, especially in the region where new planed
investments are to be realized (Albania, southern Serbia and Montenegro).
4.50
4.8 Scenario 2015 – wet hydrology – 2010 topology
This part of the Study presents results of static load flow and voltage profile analyses that are
conducted for complete network topology and (n-1) contingencies in the scenario which is denoted
as Scenario 2015 wet hydrology and expected network topology for 2010.
4.8.1 Line loadings
Area totals and power exchanges for the 2015-base case-wet hydrology-topology 2010 scenario are
shown in Figure 4.8.1 and Table 4.8.1. Power flows along regional interconnection lines and system
balances are shown in Figure 4.8.2. Power flows along interconnection lines are also given in Table
4.8.2, while Figure 4.8.3 shows histogram of tie lines loadings.
SVK
11
46
1
4 50
UKR
1
AUT
69
HUN
99
54
15
86
SLO
9
26
ITA
CRO
ROM
256
13
3
1
38
366
SRB
320
BIH
26
5
1
26
28 0
MNT
368
89
31
BUL
1
0
109
ALB
201
9
10
MKD
TUR
1
0
37
GRE
2 50
Figure 4.8.1 - Area exchanges in analyzed electric power systems for 2015-base case-wet hydrology-2010 topology
scenario
4.51
Table 4.8.1 - Area totals in analyzed electric power systems for 2015-base case-wet hydrology-topology 2010 scenario
Country
Albania
Bulgaria
Bosnia and Herzegovina
Croatia
Macedonia
Romania
Serbia and UNMIK
Montenegro
TOTAL - SE EUROPE
Generation
(MW)
1124.5
7553.2
2154.2
2799.9
869.5
7953.9
8413.1
778.7
31646.9
Load
(MW)
1528.9
6446.1
2278.6
3660.4
1407.6
7704.0
7209.2
669.8
30904.6
Bus Shunt
(Mvar)
0
0
0
0
0
0
0
0.5
0.5
Line Shunt
(Mvar)
0
14.7
0
0
0
74.3
12.7
1.7
103.4
Losses
(MW)
85.8
137.4
74.8
60.9
18.9
298.9
278.4
21.6
976.6
Net Interchange
(MW)
-490.1
955.0
-199.3
-921.3
-557.0
-123.3
912.7
85.0
-338.3
Figure 4.8.3 shows that the tie lines in the region are mostly loaded less than 25% of their thermal
limits for the analyzed hydrological base case scenario in year 2015, examined on network topology
in 2010. Among total number of forty nine 400 kV and 220 kV interconnection lines in the region
only eight are loaded between 25% and 50% of their thermal ratings. Only one line (OHL 400 kV
Sarajevo 20 – Piva between Bosnia and Herzegovina and Montenegro) is loaded more than 50% of
its thermal rating.
Table 4.8.3 lists all network elements, observing only lines 400 kV and 220 kV and transformers
400/x kV and 220/x kV, that are loaded over 80% of their thermal limits. Most of the elements
loaded over 80% are transformers in some substations and internal 110 kV and 220 kV lines. Thus,
certain internal network reinforcements are necessary to sustain given load-demand level and
generation pattern.
All three 220/110 kV transformers in the Fierze substation in Albania are overloaded in this
scenario. There are eight 220/110 kV transformers in Albania which are highly loaded. Line 220 kV
from Tirana to Rashbull is loaded near its limit. There are fourteen 110 kV lines in Albania loaded
over 80% of their thermal rating and five of them are overloaded (the largest overloading is noticed
for the 110 kV line Fierze-F.arrez).
Power systems of Bulgaria and Bosnia and Herzegovina have several highly loaded branches in the
110 kV networks. Highly loaded high voltage branches are 220 kV line M.East-St.Zagora in
Bulgaria and transformer 400/110 kV in TPP Ugljevik in Bosnia and Herzegovina. Line 110 kV
Orasje-Zupanja between Croatia and Bosnia and Herzegovina is overloaded too (135 % Ithermal).
There are ten 220 kV and four 110 kV internal lines in Romania which are loaded over 80% of their
thermal limits, whereas four of them are overloaded. These lines are related to the Lotru, Sibiu,
Tg.Jiu and Parosen nodes. Transformer 400/220 kV in the Urechesti substation is slightly
overloaded in this scenario.
Three 220 kV lines and nineteen 110 kV lines in the Serbian power system are highly loaded or
overloaded when all branches are available in the analyzed scenario. Highly loaded 220 kV lines are
connected to the Obrenovac substation, while 110 kV lines are located mostly in the areas of
Belgrade, Bor, Kragujevac, Sombor and Novi Sad. One 220 kV line and nine 110 kV lines are
overloaded, ranging between 104% Ithermal and 129% Ithermal.
4.52
LEVIC 1400.0
CENTREL
200
136
251
112
GABICK1400.0
450.0 MW
MSAJO 4
410.0
87.2
49.9
224.4
UMUKACH
412.0
23.2 ARAD
75.7
394.4
ROM
211
127
-123.4 MW
JSUBO31 391.4
JHDJE11
133
47.0
133
46.1
P.D.FIE
TANTAREN
397.6
233
3.0
409.3
UGLJEVIK 395.7
SCG
226.4
218
19.6
JHPIVA21
221 232.9
34.2
18.9
32.5
JHPERU21
18.7 224.0
40.0
75.8
116
76.1
JPODG211
154
JPODG121221.7
CRG
85.0 MW
955.0 MW
SK 1 219.0
BITOLA 2
C_MOGILA
SK 4
401.1
412.5
407.9
AZEMLA1
367
232
KARDIA K
373 415.0
212
QES/H
30.9
5.0
AHS_FLWR415.4
K
411.6
415.3
4
20 3.1
7
0.8
23.3
K-KEXRU
417.8
4HAMITAX
0.0
0.1 415.3
72.9
0.7 61.5
4BABAESK
72.3
415.5
0.0 MW
0.0 MW
KARACQOU
250 418.1
50.0
FILIPPOI
0
16 .7
.3
8
0. .7
86
9
49. 0
82.
7
3. .5
37
LAGAD K 412.7
386.0
MI_3_4_1 419.1
BLAGOEV411.6
408.0
30.8
66.0
8
49.
104
ALB
-490.1 MW
410.7
8
10 .7
57
STIP 1
DUBROVO
366.9
9
10 4
.
32
402.6
0 0
0. 9.
1
91.5
195
BUL
JLESK21
375.6
3
20 1.4
.3
4.9
51.1
92.2
163
AVDEJA2
215.4AFIERZ2217.7
379.3
MKD
KV: 110 , 220 , 400
DOBRUD4 415.8
412.6
281 SOFIA_W4
152
408.6
JVRAN31
0.000
-557.0 MW
AKASHA1
278
191
333
57.1
9
17.
8.9
9
17.
8.9
83.7
111
AEC_400
JNIS2 1
388.0
JPRIZ22
215.1 JTKOSA2
221.8JTKOSB1400.0
3
28 1.
.2 5
385.2
997.8 MW
SRB
912.7 MW
332
30.0
RP TREB
228.9
JVARDI22
86.9 225.8
64.9
17.8
19.2
SA 20
226.4
86.5
64.2
17.8
19.2
VISEGRA
227.6
AVDEJA1
GIS REGIONAL MODEL - WET HYDRO
2015 TOPOLOGY 2010
ISACCEA 402.9
396.2
5.1
43.6
221.0
TE TUZL
398.4
23.2
17.9
257
JSMIT21
105
84.0
87.7
401.4
55.1 MO-4
30.1
231.5
BIH
-199.3 MW
ROSIORI 399.8
.4
77 .0
28
.4
77 .0
28
MO-4
75.4
NADAB
104
396.6
86.1
180
JSOMB31
388.7
255
67.3
401.3
7
63. 5
97.
54.5
18.8
238
71.8
219.5
.1
95 .9
36
237
21.7
GRADACAC
75.2
86.0
77.2
76.6
218.6
RP TREB
394.6
SHAW POWER
TECHNOLOGIES
INC. R
10.
100 9
20
46. 5
2
10.4
1.3
33.6
37.5
219.6
7
47. 5
16.
.0
34 .9
44
PRIJED2
403.8
87.0
67.0
UKR
450.0 MW
3.7
57.1
DAKOVO
222.9MEDURIC
10.4
6.6
CRO
-921.3 MW
HE ZAKUC
234.0
10.7
29.1
MSAFA 4
397.4
KONJSKO
10.8
7.5
.8
85 58
1
27 234
.8
401.4
MRACLIN
63.5
43.1
ERNESTIN
93.9
39.7
20.4
4.2
5
16 8
5.
5
16 8
5.
TUMBRI
406.7
45.
5
1
45 07
10 .5
7
.8
.8
52 .6 52 .6
80
80
ZERJAVIN
403.8
ZERJAVIN 227.3
403.0
PEHLIN 229.9
227.4
45
59 .3
.5
45
59 .3
.5
LCIRKO2
401.3
.4
13 2.8
MELINA
35.7
UMUKACH2
27.4
MBEKO 4
400.5
408.4
MPECS 4
LKRSKO1
20
5. .4
9
35.5
20.6
52.9
26.2
SLO
10
401.3 40 3
.7
LDIVAC2
229.5
52.9
26.2
215.0 MW
47.5
21.2
15.1
73.3
MHEVI 4
13.4
7.0
15.1
69.8
21.2 228.9
27.1
LMARIB1 402.6
103
17.3
230.5
178
13.2
21.3
13.2
MTLOK 2
230.5
HUN
165
30.5
IPDRV121
177
7.7
50.2
21.4
MKISV 2
226.3
-1250.0 MW
165
30.5
401.0
MGYOR 2
50.7
6.3
44.
12. 7
5
44.
12. 7
5
156
44.
1
IRDPV111
407.3
4
11. 9
15
LPODLO2229.0
MSAJI 408.1
4
212
2.2
44.7
43.7
44.7
43.7
234.2
ONEUSI2
233.5
LDIVAC1
MGOD
OWIEN 2
350
UZUKRA01
292
772.5
0.0
61.5
OKAINA1 401.7
158
48.0
OOBERS2238.7
346
807
77.2
76.6
97.8
23.2
405.4
199
89.9
UCTE
223.3 MW
MAISA 7
713.9
MGYOR 4
0
25 .7
97.7
89
102
403.3
OWIEN 1
GRE
TUR
BUS -VOLTAGE(KV)
BRANCH -MW/MVAR
EQUIPMENT -MW/MVAR
Figure 4.8.2 - Power flows along interconnection lines in the region for 2015 - base case - wet hydrology scenario – topology 2010
4.53
Table 4.8.2 - Power flows along regional interconnection lines for 2015 - base case-wet hydrology scenario-topology
2010
Power Flow
Interconnection line
OHL 400 kV
OHL 220 kV
OHL 220 kV
OHL 400 kV
OHL 400 kV
OHL 400 kV
OHL 400 kV
OHL 400 kV
OHL 220 kV
OHL 220 kV
OHL 220 kV
OHL 220 kV
OHL 220 kV
OHL 220 kV
OHL 220 kV
OHL 220 kV
OHL 220 kV
OHL 400 kV
OHL 400 kV
OHL 400 kV
OHL 400 kV
OHL 400 kV
OHL 400 kV
OHL 2x400 kV ckt.1
OHL 2x400 kV ckt.2
OHL 400 kV
OHL 2x400 kV ckt.1
OHL 2x400 kV ckt.2
OHL 2x400 kV ckt.1
OHL 2x400 kV ckt.2
OHL 2x400 kV ckt.1
OHL 2x400 kV ckt.2
OHL 400 kV
OHL 400 kV
OHL 220 kV
OHL 220 kV
OHL 400 kV
OHL 400 kV
OHL 400 kV
OHL 2x220 kV ckt.1
OHL 2x220 kV ckt.2
OHL 400 kV
OHL 400 kV OHL 400 kV
OHL 400 kV
OHL 400 kV
OHL 400 kV
OHL 220 kV
OHL 220 kV
Zemlak (ALB)
Fierze (ALB)
V.Dejes (ALB)
V.Dejes (ALB)
Ugljevik (B&H)
Mostar (B&H)
Ugljevik (B&H)
Trebinje (B&H)
Trebinje (B&H)
Prijedor (B&H)
Prijedor (B&H)
Gradacac (B&H)
Tuzla (B&H)
Mostar (B&H)
Visegrad (B&H)
Sarajevo 20 (B&H)
Trebinje (B&H)
Blagoevgrad (BUL)
M.East 3 (BUL)
M.East 3 (BUL)
M.East 3 (BUL)
C.Mogila (BUL)
Dobrudja (BUL)
Kozloduy (BUL)
Kozloduy (BUL)
Sofia West (BUL)
Zerjavinec (CRO)
Zerjavinec (CRO)
Ernestinovo (CRO)
Ernestinovo (CRO)
Tumbri (CRO)
Tumbri (CRO)
Melina (CRO)
Ernestinovo (CRO)
Zerjavinec (CRO)
Pehlin (CRO)
Dubrovo (MCD)
Bitola (MCD)
Skopje (MCD)
Skopje (MCD)
Skopje (MCD)
Arad (ROM)
Nadab (ROM)
Rosiori (ROM)
Portile De Fier (ROM)
Subotica (SER)
Ribarevine (MON)
Pljevlja (MON)
Pljevlja (MON)
Kardia (GRE)
Prizren (SER)
Podgorica (MON)
Podgorica (MON)
Ernestinovo (CRO)
Konjsko (CRO)
S. Mitrovica (SER)
Podgorica (MON)
Plat (CRO)
Mraclin (CRO)
Medjuric (CRO)
Djakovo (CRO)
Djakovo (CRO)
Zakucac (CRO)
Vardiste (SER)
Piva (MON)
Perucica (MON)
Thessaloniki (GRE)
Filippi (GRE)
Babaeski (TUR)
Hamitabat (TUR)
Stip (MCD)
Isaccea (ROM)
Tantarena (ROM)
Tantarena (ROM)
Nis (SER)
Heviz (HUN)
Heviz (HUN)
Pecs (HUN)
Pecs (HUN)
Krsko (SLO)
Krsko (SLO)
Divaca (SLO)
S.Mitrovica (SER)
Cirkovce (SLO)
Divaca (SLO)
Thessaloniki (GRE)
Florina (GRE)
Kosovo B (UNMIK)
Kosovo A (UNMIK)
Kosovo A (UNMIK)
Sandorfalva (HUN)
Bekescaba (HUN)
Mukacevo (UKR)
Djerdap (SER)
Sandorfalva (HUN)
Kosovo B (UNMIK)
Bajina Basta (SER)
Pozega (SER)
MW
-367.1
-31.4
-4.9
-83.7
63.7
237.9
-232.7
-75.8
-104.8
34.0
10.4
47.7
95.1
55.1
-86.5
-217.9
18.9
-3.7
205.4
0.0
-0.7
108.7
212.0
77.4
77.4
280.6
-52.8
-52.8
45.5
45.5
-165.2
-165.2
103.3
-255.3
-13.4
20.4
-30.8
-49.8
-331.6
-17.8
-17.8
23.2
75.4
11.4
133.0
86.1
-452.0
40.2
105.0
Mvar
-231.6
20.3
-51.1
-111.5
-97.5
-71.8
-3.0
115.9
40.5
-44.9
-6.6
16.5
36.9
-30.1
64.2
-19.6
32.5
-57.1
-46.2
-19.0
-16.3
-32.4
-2.2
28.0
28.0
151.6
-80.6
-80.6
-106.9
-106.9
5.8
5.8
3.0
67.3
-2.8
-4.2
-66.0
-103.8
30.0
-19.2
-19.2
-75.7
-103.7
-159.3
46.1
-180.5
4.3
5.4
42.5
% of thermal
rating
33
14
19
11
9
19
18
13
37
18
4
18
35
18
35
58
14
8
30
5
5
16
15
8
8
45
7
7
9
9
14
14
10
21
5
6
5
8
25
8
8
7
11
14
11
16
35
14
38
4.54
Figure 4.8.4 shows histogram of 400 kV and 220 kV regional internal lines and 400/x kV and 220/x
kV transformers loadings. 37% of observed branches are loaded below 25% of their thermal ratings,
36% are loaded between 25% and 50%, 19% are loaded between 50% and 75%, 7% of observed
branches are loaded between 75% and 100% of their thermal ratings and 1% of them are overloaded
(ten branches in total) if all branches are in operation for the analyzed scenario.
Table 4.8.3 - Network elements loaded over 80% of thermal limits for 2015-base case-wet hydrology-2010 topology
scenario (branches 400 kV and 220 kV)
AREA
ALB
BUL
ROM
SRB
ALB
B&H
ROM
SRB
ELEMENT
Lines
OHL 220 kV AKASHA2-ARRAZH2
OHL 220 kV MI_2_220-ST ZAGORA
OHL 220 kV MINTIA-SIBIU
OHL 220 kV P.D.F.II-CETATE1
OHL 220 kV LOTRU-SIBIU ckt.1
OHL 220 kV LOTRU-SIBIU ckt.2
OHL 220 kV URECHESI-TG.JIU
OHL 220 kV P.D.F.A-CETATE1
OHL 220 kV P.D.F.A-RESITA ckt.1
OHL 220 kV P.D.F.A-RESITA ckt.2
OHL 220 kV TG.JIU-PAROSEN
OHL 220 kV BUC.S-B-FUNDENI
OHL 220 kV JBGD3 21-JOBREN2
Transformers
TR 220/110 kV AFIERZ2-AFIERZ5 ckt.1
TR 220/110 kV AFIERZ2-AFIERZ5 ckt.2
TR 220/110 kV AELBS12-AELBS15 ckt.1
TR 220/110 kV AELBS12-AELBS15 ckt.2
TR 220/110 kV AELBS12-AELBS15 ckt.3
TR 220/110 kV AKASHA2-AKASH25 ckt.1
TR 220/110 kV ARRAZH2-ARRAZB5 ckt.1
TR 220/110 kV ARRAZH2-ARRAZB5 ckt.1
TR 220/110 kV AFIER 2-AFIER 5 ckt.1
TR 220/110 kV AFIER 2-AFIER 5 ckt.2
TR 220/110 kV AFIER 2-AFIER 5 ckt.3
TR 400/110 kV UGLJEVIK
TR 400/220 kV URECHESI
TR 400/220 kV BUC.S-BUC.S-B ckt.2
TR 400/220 kV BUC.S-BUC.S-B ckt.3
TR 220/110 kV FUNDENI
TR 220/110 kV FUNDENI-FUNDE2B
TR 400/220 kV JBGD8-JBGD8 22
TR 400/220 kV JTKOSB1-JTKOSB2 ckt.1
TR 400/220 kV JTKOSB1-JTKOSB2 ckt.2
TR 400/110 kV JJAGO41-JJAGO45
TR 220/110 kV JBGD3 21-JBGD 351
TR 220/110 kV JBGD3 22-JBGD 352
TR 220/110 kV JTKOSA2-JTKOSA5 ckt.2
TR 220/110 kV JTKOSA2-JTKOSA5 ckt.3
TR 220/110 kV JZREN22-JZREN25
LOADING
MVA
RATING
MVA
PERCENT
259.3
191.0
324.2
269.0
303.6
303.6
280.4
206.0
239.1
239.1
280.4
285.4
313.3
270.0
228.6
381.1
277.4
277.4
277.4
277.4
208.1
277.4
277.4
208.1
320.0
301.0
96.0
83.5
85.1
97.0
109.4
109.4
101.1
99.0
86.2
86.2
134.7
89.2
104.1
54.6
54.6
81.4
81.4
87.5
87.2
80.3
80.3
141.1
115.6
110.4
270.0
414.4
340.0
340.0
175.9
208.6
338.0
324.6
324.6
247.4
197.2
134.5
131.9
134.2
121.9
60.0
60.0
90.0
90.0
90.0
100.0
100.0
100.0
120.0
90.0
90.0
300.0
400.0
400.0
400.0
200.0
200.0
400.0
400.0
400.0
300.0
200.0
150.0
150.0
150.0
150.0
91.1
91.1
90.5
90.5
97.2
87.2
80.3
80.3
117.5
128.5
122.6
90.0
103.6
85.0
85.0
88.0
104.3
84.5
81.1
81.1
82.5
98.6
89.7
87.9
89.5
81.3
4.55
45
40
Frequency
35
30
25
20
15
10
5
0
x<25
25<x<50
50<x<75
75<x<100
x>100
Bin
Figure 4.8.3 - Histogram of interconnection lines loadings for 2015-base case-wet hydrology-2010 topology scenario
("Frequency" denotes number of lines and "Bin" denotes loading range in % of thermal limit)
300
Frequency
250
200
150
100
50
0
x<25
25<x<50
50<x<75
75<x<100
x>100
Bin
Figure 4.8.4 - Histogram of 400 kV and 220 kV regional lines loadings for 2015-base case-wet hydrology-2010
topology scenario ("Frequency" denotes number of lines and "Bin" denotes loading range in % of thermal limit)
4.8.2 Voltage Profile in the Region
Voltage profile in the region within this scenario which is defined by given generation and demand
pattern is seen as satisfactory despite several appearances of certain bus voltage deviations. The
deviations are shown in Table 4.8.4, which includes only 400 kV and 220 kV network buses.
Bus voltage magnitudes which are found below permitted limits (90% Vnominal in 220 kV network
and 95% Vnominal in 400 kV network) are detected in Albania (three 400 kV and two 220 kV nodes)
and Serbia (three 400 kV nodes). Bus voltage magnitudes that are found above permitted limits
(110% Vnominal in 110 kV and 220 kV networks and 105% Vnominal in 400 kV network) are detected
only in Bulgaria in one node. Figure 4.8.5 shows histogram of bus voltage magnitudes in monitored
400 kV and 220 kV substations.
4.56
Table 4.8.4 - Bus voltage deviations for 2015-base case-wet hydrology-2010 topology scenario, complete network
Country
Voltages
Node
ALBANIA
BOSNIA AND HERZEGOVINA
BULGARIA
CROATIA
MACEDONIA
MONTENEGRO
ROMANIA
400 kV AELBS21
400 kV AVDEJA1
400 kV AKASHA1
220 kV AFIER 2
220 kV ABABIC2
400 kV MARITSA EAST2
400 kV JBGD201
400 kV JPANC21
400 kV JLESK21
kV
366.7
379.3
366.9
188.3
191.0
420.3
378.9
377.5
375.6
x>1.1
1.05<x<1.1
1.025<x<1.05
1<x<1.025
0.975<x<1
0.95<x<0.975
0.9<x<0.95
90
80
70
60
50
40
30
20
10
0
x<0.9
Frequency
SERBIA AND UNMIK
pu
0.917
0.948
0.917
0.856
0.868
1.051
0.947
0.944
0.939
Bin
Figure 4.8.5 - Histogram of voltages in monitored substations for 2015-base case-wet hydrology-2010 topology
scenario ("Frequency" denotes number of busses and "Bin" denotes voltage range in p.u.)
It should be emphasized that these results represent only a situation when additional devices
(transformer automatic tap changers, switched shunts, etc.) are not used for voltage regulation.
Impacts of such devices, which exist in many points of the SEE regional transmission network,
need more comprehensive and thorough analysis.
4.8.3 Security (n-1) analysis
Results of security (n-1) analysis for the 2015-base case-wet hydrology-2010 topology scenario are
presented in Table 4.8.5 and Table 4.8.6. Critical contingencies which are included there are related
only to the branches which are not overloaded in the base case when all branches are available.
4.57
Insecure states for given generation and demand pattern are detected in the power systems of
Romania, Serbia, Albania, although there is one contingency in Bulgaria which leads to insecure
state.
The most often overloaded lines in the Romanian power system are 400 kV OHL Mintia-Sibiu, 220
kV OHL Bucuresti Sud-Fundeni and lines 220 kV connected to the Portile de Fier substation. These
lines are overloaded in several contingency cases. Lines 220 kV Lotru-Sibiu, Urechesti-Targa Jiu,
Paroseni-Targa Jiu and transformer in 400/220 kV substation Urechesti are already overloaded in
the base case when all branches are available. Possible actions to reduce the loading of the critical
branches look for a re-dispatching of the generators in Romania within this scenario, especially the
HPP LOTRU CIUNGET (dispatched at 510 MW in the base case), HPP PORTILE 1 (921 MW),
TPP ROVINARI (540 MW) and TPP PAROSENI.
Line 400 kV Mintia-Sibiu has rating of 381 MVA on the model which seems as too low value, so
contingencies in which this line is overloaded are not included in the contingency tables. It is
overloaded if one among following lines goes out of operation: 400 kV Sibiu-Iernut (loading of
Mitia-Sibiu line is 142.3 % Ithermal), 400 kV Iernut-Gadalin (107.7 % Ithermal), 400 kV RosioriGadalin (100.3 % Ithermal), 220 kV Urechesi-Tg.Jiu (117.7 % Ithermal), 220 kV Tg.Jiu-Parosen (117.6
% Ithermal) and 220 kV Baru M-Hajd Ot (100.0 % Ithermal).
Line 220 kV Buc.S-B-Fundeni is the double circuit line but one circuit is permanently out of
operation on the model which is the reason why this line is included in the contingency tables.
Installed capacity in the substations 400/220 Bucuresti Sud, Brasov, Dirste, Sibiu and Brazi in
Romania are too low to support this generation/demand scenario.
Critical lines in the Serbian power system for this scenario are detected mostly in 220 kV network
of the Beograd area. Installed transformer capacities are inadequate in the Kragujevac, Nis, Kosovo
B and Pancevo substations.
Single outages of 400/110 kV transformers in the stations Brasov and Dirste in Romania are also
found critical, since the second transformer 400/110 kV in the Brasov substation is permanently out
of operation in the model.
The heaviest line overloading (197% Ithermal) in the analyzed scenario is related to 220 kV line in
Serbia (Beograd 3-Obrenovac) when the line 400 kV Beograd 1-Obrenovac goes out of operation.
The heaviest transformer overloading (168% Sn) is related to the transformer 400/110 kV in the
Dirste substation (Romania) when the transformer 400/110 kV in the Brasov substation is outaged
(the parallel one is permanently out of operation in the model).
Figure 4.8.6 shows geographical positions of the critical elements in the analyzed scenario. A green
color reveals 220 kV elements (line 220 kV or transformer 220/x kV), while a red one reveals 400
kV elements (line 400 kV or transformer 400/x kV).
According to the obtained and presented results, it may be concluded that the network topology as
predicted to exist in 2010 is not suitable for the analyzed generation pattern. Larger investments in
the internal networks, especially in the power systems of Romania, Serbia and Albania, will be
necessary in order to support such generation pattern.
4.58
Table 4.8.5 - Lines overloadings for 2015–base case-wet hydrology-2010 topology scenario, single outages
OHL 220 kV AFIERZ2-ABURRE2
TR 400/220 kV AELBS21-AELBS22
TR 400/220 kV URECHESI
OHL 220 kV LOTRU-SIBIU ckts. 1 & 2
OHL 220 kV URECHESI-TG.JIU
OHL 220 kV TG.JIU-PAROSEN
OHL 220 kV JBGD3 21-JOBREN2
OHL 220 kV AKASHA2-ARRAZH2
OHL 220 kV AKASHA2-ARRAZH2
Loadings
MVA
%
418.5 104.6
311.4 109.4
282.1 101.1
282.1 134.7
299.7 104.1
256.7 112.9
253.7 109.1
OHL 400 kV RP TREB-JPODG221
OHL 220 kV AKASHA2-ARRAZH2
242.9
105.4
OHL 220 kV G_ORIAH-MI_2_220
OHL 400 kV TANTAREN-BRADU
OHL 400 kV DOMNESTI-BUC.S
OHL 400 kV DOMNESTI-BRAZI
OHL 220 kV P.D.F.A-CALAFAT
OHL 220 kV P.D.F.A-RESITA ckt.1
OHL 220 kV RESITA-TIMIS ckt.1
OHL 220 kV FUNDENI- BUC.S-B
TR 400/220 kV BRAZI
OHL 400 kV JHDJE11-JTDRMN1
OHL 400 kV JTKOSB1-JPEC 1
OHL 220 kV JBGD172-JBGD8 22 ckt.1
TR 400/220 kV JBGD8
OHL 220 kV MI_2_220_ST.ZAGORA
OHL 220 kV BUC.S-B-FUNDENI
OHL 220 kV BUC.S-B-FUNDENI
OHL 220 kV BUC.S-B-FUNDENI
OHL 220 kV P.D.F.A-CETATE1
OHL 220 kV P.D.F.A-RESITA ckt.2
OHL 220 kV RESITA-TIMIS ckt.2
OHL 220 kV BUC.S-B-FUNDENI
OHL 220 kV BUC.S-B-FUNDENI
OHL 220 kV P.D.F.A-RESITA ckt.1&2
OHL 220 kV AKASHA2-ARRAZH2
OHL 220 kV JBGD172-JBGD8 22 ckt.2
OHL 220 kV JBGD3 22-JBGD8 22
255.4
329.5
321.8
355.5
259.9
338.8
342.8
368.3
369.7
266.6
243.2
475.9
364.7
105.1
108.3
102.7
114.6
128.7
129.6
128.2
118.1
122.0
102.0
107.0
140.2
108.4
Outage
Base case
Overloaded line(s)
Country
ROMANIA
SERBIA
ALBANIA
B&H/MONT
/ALB
BULGARIA
ROMANIA
SER/ROM
SER/ALB
SERBIA
Table 4.8.6 - Transformers overloadings for 2015–base case-wet hydrology-2010 topology scenario, single outages
Outage
Overloaded branch(es)
TR 400/220 kV BUC.S-BUC.S-B ckt.1
400/110 kV BRASOV
400/110 kV DIRSTE
TR 400/220 kV SIBIU ckt.1
TR 400/220 kV BUC.S-BUC.S-B ckt.2
TR 400/110 kV DIRSTE
TR 400/110 kV BRASOV
TR 400/220 kV SIBIU ckt.2
TR 400/220 kV BUC.S ckt.1
TR 400/220 kV BUC.S ckt.2
TR 400/110 kV JKRAG ckt.2
TR 400/110 kV NIS ckt.2
TR 400/220 kV JBGD8 1-JBGD8 22
TR 400/110 kV JPANC ckt.2
TR 400/110 kV JTKOSB ckt.2
TR 400/220 kV BRAZI
TR 400/110 kV JKRAG ckt.1
TR 400/110 kV NIS ckt.1
OHL 220 kV JBGD3 22-JBGD8 22
TR 400/110 kV JPANC ckt.1
TR 400/110 kV JTKOSB ckt.1
Loadings
MVA
%
536.5 134.1
420.0 168.0
410.8 164.3
573.0 143.3
413.3 103.3
413.3 103.3
316.5 105.5
340.4 113.5
412.1 103.0
334.0 111.3
439.3 109.8
Country
ROMANIA
SERBIA
4.59
SVK
AUT
HUN
SLO
ROM
CRO
SRB
BIH
MNG
BUL
MKD
TUR
ALB
GRE
critical elements
400 kV line or 400/x kV transformer
220 kV line or 220/x kV transformer
Figure 4.8.6 - Geographical positions of the critical elements for 2015-base case-wet hydrology-topology 2010 scenario
4.60
4.9 Scenario 2015 – wet hydrology – topology 2015
This part of the Study presents results of static load flow and voltage profile analyses that are
conducted for complete network topology and (n-1) contingencies in the scenario which is denoted
as Scenario 2015 wet hydrology and expected network topology for 2015.
4.9.1 Line loadings
Area totals and power exchanges for the 2015-base case-wet hydrology-topology 2015 scenario are
shown in Figure 4.9.1 and Table 4.9.1. Power flows along regional interconnection lines and system
balances are shown in Figure 4.9.2. Power flows along interconnection lines are also given in Table
4.9.2, while Figure 4.9.3 shows histogram of tie lines loadings.
SVK
3
45
3
4 50
UKR
1
AUT
79
HUN
74
43
9
84
SLO
8
25
ITA
CRO
ROM
246
12
7
6
42
327
SRB
289
BIH
72
6
33
275
MNT
292
79
1
33
1
BUL
1
ALB
9
23
355
MKD
202
9
14
TUR
1
98
GRE
2 50
Figure 4.9.1 - Area exchanges in analyzed electric power systems for 2015-base case-wet hydrology-2015 topology
scenario
Table 4.9.1 - Area totals in analyzed electric power systems for 2015-base case-wet hydrology-topology 2015 scenario
Country
Albania
Bulgaria
Bosnia and Herzegovina
Croatia
Macedonia
Romania
Serbia and UNMIK
Montenegro
TOTAL - SE EUROPE
Generation
(MW)
1111.0
7553.3
2155.1
2800.9
870.8
7950.3
8398.5
774.1
31613.9
Load
(MW)
1528.9
6446.1
2278.6
3660.4
1407.6
7704.0
7209.2
669.8
30904.6
Bus Shunt
(Mvar)
0
0
0
0
0
0
0
0.5
0.5
Line Shunt
(Mvar)
0
14.7
0
0
0
74.5
13.2
1.8
104.2
Losses
(MW)
72.1
137.5
75.6
61.6
20.2
294.8
263.2
16.9
941.8
Net Interchange
(MW)
-490.0
955.0
-199.0
-921.1
-557.0
-123.0
913.0
85.0
-337.2
4.61
New elements that are expected to be built till 2015 cause significantly different distribution of
power flows in the southern part of the region (Albania, FYR of Macedonia, Serbia, Montenegro).
Compared to the expected topology 2010, analyzed in the previous chapter, it can be seen that the
network losses are smaller as a consequence of building the new elements in the 2015 network
topology, especially in the cases of Albania, Serbia and Montenegro. Overall reduction of losses is
about 35 MW.
Figure 4.9.3 shows that the tie lines in the region are mostly loaded less than 25% of their thermal
limits for the analyzed hydrological base case scenario in year 2015, examined on the planned
network topology for that year. Among total number of fifty two 400 kV and 220 kV
interconnection lines in the region only eight are loaded between 25% and 50% of their thermal
ratings. Only one line (OHL 400 kV Sarajevo 20 – Piva between Bosnia and Herzegovina and
Montenegro) is loaded more than 50% of its thermal rating.
Table 4.9.4 lists all network elements, observing only lines 400 kV and 220 kV and transformers
400/x kV and 220/x kV, that are loaded over 80% of their thermal limits. Most of the elements
loaded over 80% are transformers in some substations and internal 110 kV and 220 kV lines. Thus,
certain internal network reinforcements are necessary to sustain given load-demand level and
generation pattern.
Being compared to the same generation pattern, but with the 2010 network topology, the loading
relief of some branches and transformers in the southern Serbia and Albania may be noticed.
Figure 4.9.4 shows histogram of 400 kV and 220 kV regional internal lines and 400/x kV and 220/x
kV transformers loadings. 38% of observed branches are loaded below 25% of their thermal ratings,
36% are loaded between 25% and 50%, 20% are loaded between 50% and 75%, 5% of observed
branches are loaded between 75% and 100% of their thermal ratings and 1% of them are overloaded
(nine branches in total) if all branches are in operation for the analyzed scenario.
Being compared to the same generation pattern on the 2010 network topology, it is noticed that
there are no significant changes in the total number of branches and the ranges of internal lines
loading. Only one internal branch is loaded below permitted limit in the case of building new
interconnection lines as predicted to exist in 2015.
4.62
LEVIC 1400.0
CENTREL
200
136
251
113
GABICK1400.0
450.0 MW
105
23.7
405.4
OKAINA1 401.7
234.2
224.4
UMUKACH
412.0
JSOMB31
389.3
11.8 ARAD
72.8
395.0
ROM
203
128
ISACCEA 403.2
-123.1 MW
JSUBO31 392.0
397.0
JHDJE11
127
24.2
127
23.2
P.D.FIE
TANTAREN
398.5
208
4.4
409.7
SCG
JHPIVA21
219 235.1
40.1
5.4
18.5
JHPERU21
5.4 228.7
26.9
CRG
85.0 MW
SK 1 219.6
SK 4
400.1
-557.0 MW
91.5
108
388.8
JVRAN31
397.4
411.3
C_MOGILA
9
14 3
.
26
401.3
407.6
DUBROVO
AZEMLA1
98.1
128
KARDIA K
98.8 416.9
60.5
QES/H
95.5
12.8
23
84 9
.4
AHS_FLWR415.6
K
412.2
KV: 110 , 220 , 400
415.4
7
20 2.7
7
1.0
23.5
K-KEXRU
417.8
4HAMITAX
0.7
0.7 415.3
73.4
1.0 62.1
4BABAESK
72.5
415.5
0.0 MW
-0.1 MW
KARACQOU
250 419.1
50.0
FILIPPOI
0
15 .2
.6
2
0. .4
87
260
114
2
5. .0
38
LAGAD K 413.2
405.5
MI_3_4_1 419.2
BLAGOEV412.1
407.4
ALB
-490.0 MW
411.3
9
14 .9
60
STIP 1
259
131
AKASHA1
240
4
45.
955.0 MW
8 5
0. 8.
1
91.5
108
MKD
BITOLA 2
BUL
JLESK21
393.6
113
64.6
3
43 1.1
.9
AVDEJA2
225.8AFIERZ2227.9
394.9
26.7
23.2
28.8
24.9
151
39.6
362 8
11.
91.7
67.4
91.7
67.5
354
44.5
8
26. 2
13.
AVDEJA1
361
26.5
8
26. 2
13.
3
51 0.
.5 7
395.3
JPRIZ22
220.6 JTKOSA2
223.4JTKOSB1400.0
DOBRUD4 416.0
413.0
276 SOFIA_W4
111
409.5
21.8 21.7
43.6 79.9
133
JPODG211
111
JPODG121227.6
274
156
1
21 14
.5
133
70.9
AEC_400
JNIS2 1
393.0
0.7
62.1
216
26.6
998.0 MW
SRB
913.0 MW
95.3
71.9
RP TREB
231.5
JVARDI22
80.3 226.9
60.3
352
20.1
SA 20
227.8
80.0
59.3
26.7
23.2
VISEGRA
228.6
204
2.9
UGLJEVIK 396.6
226.9
28.6
16.2
221.4
TE TUZL
398.8
11.7
14.8
247
JSMIT21
102
150
14.8
403.9
58.9 MO-4
27.2
232.5
BIH
-199.0 MW
ROSIORI 400.3
.9
61 .0
29
.9
61 .0
29
MO-4
62.7
NADAB
100
397.1
61.8
78.0
246
62.2
401.9
5
73. 0
94.
58.4
16.0
253
59.1
.1
98 .1
37
252
9.9
GRADACAC
219.8
62.6
82.0
61.8
78.0
218.9
RP TREB
401.0
GIS REGIONAL MODEL - WET HYDRO
2015 TOPOLOGY 2015
2.7
94.
4
20
46. 8
3
12.9
1.4
38.3
37.2
219.8
0
50. 5
16.
.7
38 .5
44
PRIJED2
SHAW POWER
TECHNOLOGIES
INC. R
UKR
450.0 MW
5.1
56.8
DAKOVO
223.0MEDURIC
12.9
6.5
CRO
-921.1 MW
405.0
79.0
49.4
MSAFA 4
398.4
HE ZAKUC
234.4
78.8
67.8
132 132
20.0 39.2
401.5
MRACLIN
KONJSKO
10.7
29.1
.1
84 54
1
28 209
.3
TUMBRI
406.9
51.
2
1
51 03
10 .2
3
ERNESTIN
73.3
39.5
5
16 8
6.
5
16 8
6.
22.4
4.0
227.4
.3
.3
46 .8 46 .8
80
80
ZERJAVIN
404.0
ZERJAVIN 227.4
403.1
PEHLIN 230.0
10.8
7.5
84.4
177
LCIRKO2
51
55 .0
.6
51
55 .0
.6
401.3
.9
12 2.6
MELINA
35.7
UMUKACH2
27.5
46.5
26.3
MPECS 4
LKRSKO1
22
6. .4
0
35.5
20.7
MBEKO 4
400.9
408.5
96.8
39.5
229.6
46.5
26.3
SLO
49.8
21.1
16.0
72.9
MHEVI 4
215.0 MW
11
401.4 41 1
.7
LDIVAC2
MSAJO 4
410.0
LMARIB1 402.7
111
18.4
230.5
15.9
69.4
181
11.5
22.3 228.9
26.8
12.9
7.3
IPDRV121
181
9.3
22.4
12.9
MTLOK 2
230.5
HUN
165
31.5
401.0
51.5
21.0
MKISV 2
226.3
350
UZUKRA01
293
772.5
-1249.9 MW
165
31.5
IRDPV111
407.3
346
807
3.1
154
LPODLO2229.1
MSAJI 408.1
4
MGYOR 2
52.0
6.1
41.
12. 8
5
155
44.
1
41.
12. 8
5
ONEUSI2
233.5
LDIVAC1
MGOD
OWIEN 2
41.9
43.7
41.9
43.7
157
47.7
OOBERS2238.7
199
90.4
UCTE
222.3 MW
MAISA 7
714.0
MGYOR 4
0
25 .3
105
90
101
403.3
OWIEN 1
GRE
TUR
BUS -VOLTAGE(KV)
BRANCH -MW/MVAR
EQUIPMENT -MW/MVAR
Figure 4.9.2 - Power flows along interconnection lines in the region for 2015 - base case - wet hydrology scenario – topology 2015
4.63
Table 4.9.2 - Power flows along regional interconnection lines for 2015 - base case-wet hydrology scenario-topology
2015
Power Flow
Interconnection line
OHL 400 kV
Zemlak (ALB)
OHL 220 kV
Fierze (ALB)
OHL 220 kV
V.Dejes (ALB)
OHL 400 kV
V.Dejes (ALB)
OHL 400 kV
Ugljevik (B&H)
OHL 400 kV
Mostar (B&H)
OHL 400 kV
Ugljevik (B&H)
OHL 400 kV
Trebinje (B&H)
OHL 220 kV
Trebinje (B&H)
OHL 220 kV
Prijedor (B&H)
OHL 220 kV
Prijedor (B&H)
OHL 220 kV
Gradacac (B&H)
OHL 220 kV
Tuzla (B&H)
OHL 220 kV
Mostar (B&H)
OHL 220 kV
Visegrad (B&H)
OHL 220 kV
Sarajevo 20 (B&H)
OHL 220 kV
Trebinje (B&H)
OHL 400 kV
Blagoevgrad (BUL)
OHL 400 kV
M.East 3 (BUL)
OHL 400 kV
M.East 3 (BUL)
OHL 400 kV
M.East 3 (BUL)
OHL 400 kV
C.Mogila (BUL)
OHL 400 kV
Dobrudja (BUL)
OHL 2x400 kV ckt.1
Kozloduy (BUL)
OHL 2x400 kV ckt.2
Kozloduy (BUL)
OHL 400 kV
Sofia West (BUL)
OHL 2x400 kV ckt.1
Zerjavinec (CRO)
OHL 2x400 kV ckt.2
Zerjavinec (CRO)
OHL 2x400 kV ckt.1
Ernestinovo (CRO)
OHL 2x400 kV ckt.2
Ernestinovo (CRO)
OHL 2x400 kV ckt.1
Tumbri (CRO)
OHL 2x400 kV ckt.2
Tumbri (CRO)
OHL 400 kV
Melina (CRO)
OHL 400 kV
Ernestinovo (CRO)
OHL 220 kV
Zerjavinec (CRO)
OHL 220 kV
Pehlin (CRO)
OHL 400 kV
Dubrovo (MCD)
OHL 400 kV
Bitola (MCD)
OHL 400 kV
Skopje (MCD)
OHL 2x220 kV ckt.1
Skopje (MCD)
OHL 2x220 kV ckt.2
Skopje (MCD)
OHL 400 kV
Arad (ROM)
OHL 400 kV Nadab (ROM)
OHL 400 kV
Rosiori (ROM)
OHL 400 kV
Portile De Fier (ROM)
OHL 400 kV
Subotica (SER)
OHL 400 kV
Ribarevine (MON)
OHL 220 kV
Pljevlja (MON)
OHL 220 kV
Pljevlja (MON)
OHL 400 kV*
Skopje 4 (MCD)
OHL 400 kV*
Zemlak (ALB)
OHL 400 kV*
V.Dejes (ALB)
* new lines planned till 2015
Kardia (GRE)
Prizren (SER)
Podgorica (MON)
Podgorica (MON)
Ernestinovo (CRO)
Konjsko (CRO)
S. Mitrovica (SER)
Podgorica (MON)
Plat (CRO)
Mraclin (CRO)
Medjuric (CRO)
Djakovo (CRO)
Djakovo (CRO)
Zakucac (CRO)
Vardiste (SER)
Piva (MON)
Perucica (MON)
Thessaloniki (GRE)
Filippi (GRE)
Babaeski (TUR)
Hamitabat (TUR)
Stip (MCD)
Isaccea (ROM)
Tantarena (ROM)
Tantarena (ROM)
Nis (SER)
Heviz (HUN)
Heviz (HUN)
Pecs (HUN)
Pecs (HUN)
Krsko (SLO)
Krsko (SLO)
Divaca (SLO)
S.Mitrovica (SER)
Cirkovce (SLO)
Divaca (SLO)
Thessaloniki (GRE)
Florina (GRE)
Kosovo B (UNMIK)
Kosovo A (UNMIK)
Kosovo A (UNMIK)
Sandorfalva (HUN)
Bekescaba (HUN)
Mukacevo (UKR)
Djerdap (SER)
Sandorfalva (HUN)
Kosovo B (UNMIK)
Bajina Basta (SER)
Pozega (SER)
Vranje (SER)
Bitola (MCD)
Kosovo B (UNMIK)
MW
-98.1
31.1
28.8
150.5
73.5
253.5
-208.1
-133.1
-104.8
38.7
12.9
50.0
98.1
58.9
-80.0
-215.8
5.4
-5.1
208.0
1.0
0.0
149.4
204.0
61.9
61.9
276.1
-46.3
-46.3
51.2
51.2
-165.2
-165.2
111.0
-245.7
-12.9
22.4
-95.3
-259.0
-351.8
-26.7
-26.7
11.8
62.7
3.1
126.8
84.4
-281.4
49.2
113.7
113.6
239.8
-360.9
Mvar
-128.3
43.9
-24.9
-39.6
-94.0
-59.1
-4.4
70.9
40.2
-44.5
-6.5
16.5
37.1
-27.2
59.3
-26.6
18.5
-56.8
-46.3
-18.5
-15.6
-26.3
-2.9
29.0
29.0
111.2
-80.8
-80.8
-103.2
-103.2
6.8
6.8
4.3
62.2
-2.6
-4.0
-71.9
-130.7
20.1
-23.2
-23.2
-72.8
-100.0
-154.0
23.2
-176.8
26.7
11.1
45.4
21.5
45.4
-26.5
% of thermal
rating
12
20
13
12
9
19
16
13
36
19
5
18
35
19
32
57
9
8
30
5
5
21
15
7
7
42
7
7
9
9
14
14
10
20
4
7
8
24
27
11
11
6
10
13
10
16
22
17
38
10
19
27
4.64
Table 4.9.3 - Network elements loaded over 80% of thermal limits for 2015-base case-wet hydrology-2015 topology
scenario (branches 400 kV and 220 kV)
AREA
ALB
BUL
ROM
SRB
ALB
B&H
ROM
SRB
ELEMENT
LOADING
MVA
RATING
MVA
PERCENT
237.0
190.4
313.4
268.4
303.4
303.4
275.4
205.6
235.6
235.6
275.4
284.6
312.6
270.0
228.6
381.1
277.4
277.4
277.4
277.4
208.1
277.4
277.4
208.1
320.0
301.0
87.8
83.3
82.2
96.8
109.4
109.4
99.3
98.8
84.9
84.9
132.3
88.9
103.8
50.2
50.2
74.8
74.8
80.4
82.3
82.3
134.5
110.3
105.2
267.6
410.3
339.3
339.3
175.7
208.3
334.9
196.6
134.4
121.5
60.0
60.0
90.0
90.0
90.0
100.0
100.0
120.0
90.0
90.0
300.0
400.0
400.0
400.0
200.0
200.0
400.0
200.0
150.0
150.0
83.7
83.7
83.2
83.2
89.3
82.3
82.3
112.1
122.5
116.9
89.2
102.6
84.8
84.8
87.8
104.1
83.7
98.3
89.6
81.0
Lines
OHL 220 kV AKASHA2-ARRAZH2
OHL 220 kV MI_2_220-ST ZAGORA
OHL 220 kV MINTIA-SIBIU
OHL 220 kV P.D.F.II-CETATE1
OHL 220 kV LOTRU-SIBIU ckt.1
OHL 220 kV LOTRU-SIBIU ckt.2
OHL 220 kV URECHESI-TG.JIU
OHL 220 kV P.D.F.A-CETATE1
OHL 220 kV P.D.F.A-RESITA ckt.1
OHL 220 kV P.D.F.A-RESITA ckt.2
OHL 220 kV TG.JIU-PAROSEN
OHL 220 kV BUC.S-B-FUNDENI
OHL 220 kV JBGD3 21-JOBREN2
Transformers
TR 220/110 kV AFIERZ2-AFIERZ5 ckt.1
TR 220/110 kV AFIERZ2-AFIERZ5 ckt.2
TR 220/110 kV AELBS12-AELBS15 ckt.1
TR 220/110 kV AELBS12-AELBS15 ckt.2
TR 220/110 kV AELBS12-AELBS15 ckt.3
TR 220/110 kV AKASHA2-AKASH25 ckt.1
TR 220/110 kV AKASHA2-AKASH25 ckt.2
TR 220/110 kV AFIER 2-AFIER 5 ckt.1
TR 220/110 kV AFIER 2-AFIER 5 ckt.2
TR 220/110 kV AFIER 2-AFIER 5 ckt.3
TR 400/110 kV UGLJEVIK
TR 400/220 kV URECHESI
TR 400/220 kV BUC.S-BUC.S-B ckt.1
TR 400/220 kV BUC.S-BUC.S-B ckt.2
TR 220/110 kV FUNDENI
TR 220/110 kV FUNDENI-FUNDE2B
TR 400/220 kV JBGD8-JBGD8 22
TR 220/110 kV JBGD3 21-JBGD 351
TR 220/110 kV JBGD3 22-JBGD 352
TR 220/110 kV JZREN22-JZREN25
Frequency
50
45
40
35
30
25
20
15
10
5
0
x<25
25<x<50
50<x<75
75<x<100
x>100
Bin
Figure 4.9.3 - Histogram of interconnection lines loadings for 2015-base case-wet hydrology-2015 topology scenario
("Frequency" denotes number of lines and "Bin" denotes loading range in % of thermal limit)
4.65
300
Frequency
250
200
150
100
50
0
x<25
25<x<50
50<x<75
75<x<100
x>100
Bin
Figure 4.9.4 - Histogram of 400 kV and 220 kV regional lines loadings for 2015-base case-wet hydrology-2015
topology scenario ("Frequency" denotes number of lines and "Bin" denotes loading range in % of thermal limit)
4.9.2 Voltage Profile in the Region
Voltage profile in the region within this scenario which is defined by given generation and demand
pattern is seen as satisfactory despite several appearances of certain bus voltage deviations. The
deviations are shown in Figure 4.9.5, which includes only 400 kV and 220 kV network buses.
Table 4.9.4 - Bus voltage deviations for 2015-base case-wet hydrology-2015 topology scenario, complete network
Country
Node
ALBANIA
BOSNIA AND HERZEGOVINA
BULGARIA
CROATIA
MACEDONIA
MONTENEGRO
ROMANIA
SERBIA AND UNMIK
400 kV MARITSA EAST2
400 kV JPANC21
Voltages
pu
1.051
0.947
kV
420.4
378.8
Bus voltage magnitudes which are found below permitted limits (90% Vnominal in 220 kV network
and 95% Vnominal in 400 kV network) are detected only in Serbia (one 400 kV node). Bus voltage
magnitudes that are found above permitted limits (110% Vnominal in 110 kV and 220 kV networks
and 105% Vnominal in 400 kV network) are detected only in Bulgaria in one node. Figure 4.3.11
shows a histogram of voltages in monitored 400 kV and 220 kV substations.
Being compared to the same generation pattern on the 2010 network topology, it is noticed that
there are significant voltage profile improvements, especially in the power systems of Albania and
Serbia.
4.66
x>1.1
1.05<x<1.1
1.025<x<1.05
1<x<1.025
0.975<x<1
0.95<x<0.975
0.9<x<0.95
x<0.9
Frequency
90
80
70
60
50
40
30
20
10
0
Bin
Figure 4.9.5 - Histogram of voltages in monitored substations for 2015-base case-wet hydrology-2015 topology
scenario ("Frequency" denotes number of busses and "Bin" denotes voltage range in p.u.)
4.9.3 Security (n-1) analysis
Results of security (n-1) analysis for the 2015-base case-wet hydrology-2015 topology scenario are
presented in Table 4.9.5 and Table 4.9.6. Critical contingencies which are included there are related
only to the branches which are not overloaded in the base case when all branches are available.
Insecure states for given generation and demand pattern are detected in the power systems of
Romania, Serbia, Albania, although there is one contingency in Bulgaria and Bosnia and
Herzegovina each which leads to insecure state.
Being compared to the same generation and demand scenario on the 2010 network topology, it is
concluded that there are several new contingency cases in the power system of Albania and one new
contingency case in the power systems of Romania and Bosnia and Herzegovina (marked with red
color in Table 4.9.5 and Table 4.9.6). At the same time, several contingency cases do not appear
any more (marked in blue color in Table 4.9.5 and Table 4.9.6). These contingency cases are related
to the power systems of Albania, Serbia and Romania, although it is noticed that the influence of
the new investments is less significant in the power system of Romania than it is in Albania and
southern Serbia.
There are no contingency cases in the 2015 network topology which are related to the
interconnection lines. The same conclusion is valid for the 2010 network topology and analyzed
generation/demand scenario.
Figure 4.9.6 shows geographical positions of the critical elements in the analyzed scenario. A green
color reveals 220 kV elements (line 220 kV or transformer 220/x kV), while a red one reveals 400
kV elements (line 400 kV or transformer 400/x kV).
According to the obtained and presented results, it may be concluded that the network topology as
predicted to exist in 2015 is not suitable for the analyzed generation pattern. Larger investments in
the internal networks, especially in power systems of Romania, Serbia and Albania, are necessary
shall such generation pattern be supported.
4.67
Table 4.9.5 - Lines overloadings for 2015–base case-wet hydrology-2015 topology scenario, single outages
Outage
Overloaded line(s)
OHL 400 kV AELBS21-AZEMLA1
OHL 220 kV AELBS12-AFIER 2
TR 400/110 kV AZEMLA1-AZEMLK5
OHL 220 kV AFIERZ2-ABURRE2
TR 400/220 kV AELBS21-AELBS22
TR 400/220 kV URECHESI
OHL 220 kV LOTRU-SIBIU ckt. 1 & 2
OHL 220 kV URECHESI-TG.JIU
OHL 220 kV TG.JIU-PAROSEN
OHL 220 kV JBGD3 21-JOBREN2
OHL 220 kV AKASHA2-ARRAZH2
OHL 220 kV AKASHA2-ARRAZH2
OHL 220 kV AKASHA2-ARRAZH2
OHL 220 kV AKASHA2-ARRAZH2
OHL 220 kV AKASHA2-ARRAZH2
OHL 400 kV RP TREB-JPODG221
OHL 220 kV AKASHA2-ARRAZH2
OHL 220 kV G_ORIAH-MI_2_220
OHL 400 kV TANTAREN-BRADU
OHL 400 kV DOMNESTI-BUC.S
OHL 400 kV DOMNESTI-BRAZI
OHL 220 kV P.D.F.A-CALAFAT
OHL 220 kV P.D.F.A-RESITA ckt.1
OHL 220 kV RESITA-TIMIS ckt.1
OHL 220 kV FUNDENI- BUC.S-B
TR 400/220 kV BRAZI
OHL 400 kV JHDJE11-JTDRMN1
OHL 400 kV JTKOSB1-JPEC 1
OHL 220 kV JBGD172-JBGD8 22 ckt.1
TR 400/220 kV JBGD8
OHL 220 kV MI_2_220_ST.ZAGORA
OHL 220 kV BUC.S-B-FUNDENI
OHL 220 kV BUC.S-B-FUNDENI
OHL 220 kV BUC.S-B-FUNDENI
OHL 220 kV P.D.F.A-CETATE1
OHL 220 kV P.D.F.A-RESITA ckt.2
OHL 220 kV RESITA-TIMIS ckt.2
OHL 220 kV BUC.S-B-FUNDENI
OHL 220 kV BUC.S-B-FUNDENI
OHL 220 kV P.D.F.A-RESITA ckt.1&2
OHL 220 kV AKASHA2-ARRAZH2
OHL 220 kV JBGD172-JBGD8 22 ckt.2
OHL 220 kV JBGD3 22-JBGD8 22
Base case
Loadings
MVA
%
414.9 103.7
311.3 109.4
277.5
299.9
242.5
359.7
254.1
254.6
328.6
321.0
354.5
259.6
335.1
338.3
367.8
369.5
132.3
103.8
100.5
143.7
102.6
104.7
107.7
102.3
114.1
128.2
127.6
126.1
117.8
121.8
Country
ROMANIA
SERBIA
ALBANIA
B&H/MONT
/ALB
BULGARIA
ROMANIA
SER/ROM
SER/ALB
475.6
366.1
139.6
108.4
SERBIA
Table 4.9.6 - Transformers overloadings for 2015–base case-wet hydrology-2015 topology scenario, single outages
Outage
Overloaded branch(es)
OHL 400 kV BLUKA 6-TS TUZL
OHL 220 kV BUC.S-B-FUNDENI
TR 400/220 kV BUC.S-BUC.S-B ckt.1
TR 400/110 kV BRASOV
TR 400/110 kV DIRSTE
TR 400/220 kV SIBIU ckt.1
TR 400/110 kV UGLJEVIK
TR 400/220 kV BRAZI
TR 400/220 kV BUC.S-BUC.S-B ckt.2
TR 400/110 kV DIRSTE
TR 400/110 kV BRASOV
TR 400/220 kV SIBIU ckt.2
TR 400/220 kV BUC.S ckt.1
TR 400/220 kV BUC.S ckt.2
TR 400/110 kV JKRAG ckt.2
TR 400/110 kV NIS ckt.2
TR 400/220 kV JBGD8 1-JBGD8 22
TR 400/110 kV JPANC ckt.2
TR 400/110 kV JTKOSB ckt.2
TR 400/220 kV BRAZI
TR 400/110 kV JKRAG ckt.1
TR 400/110 kV NIS ckt.1
OHL 220 kV JBGD3 22-JBGD8 22
TR 400/110 kV JPANC ckt.1
TR 400/110 kV JTKOSB ckt.1
Loadings
MVA
%
311.0 103.7
414.2 103.6
535.9 134.0
419.2 167.7
410.2 164.1
571.7 142.9
413.1 103.3
413.1 103.3
314.4 104.8
411.2
332.5
102.8
110.8
Country
B&H
ROMANIA
SERBIA
4.68
SVK
AUT
HUN
SLO
ROM
CRO
SRB
BIH
MNG
BUL
MKD
TUR
ALB
GRE
critical elements
400 kV line or 400/x kV transformer
220 kV line or 220/x kV transformer
Figure 4.9.6 - Geographical positions of the critical elements for 2015-base case-wet hydrology-topology 2015 scenario
4.9.4 Summary of Impacts - 2015 topology versus 2010 topology
Compared to the expected topology 2010, analyzed in previous chapter, it can be seen that the
network losses are smaller as a consequence of building the new elements for 2015 network
topology, especially in the cases of Albania, Serbia and Montenegro. Overall reduction of losses is
around 35 MW.
Comparing to the same generation pattern on the 2010 network topology, it is noticed that there are
no significant changes in the total number of branches and the ranges of internal lines loading.
Being compared to the same generation pattern on the 2010 network topology, it is noticed that
there are significant voltage profile improvements, especially in the power systems of Albania and
Serbia.
Being compared to the same generation and demand scenario on the 2010 network topology, it is
concluded that there are several new contingency cases in the power system of Albania and one new
contingency case in the power systems of Romania and Bosnia and Herzegovina. At the same time,
several contingency cases do not appear any more. These contingency cases are related to the power
systems of Albania, Serbia and Romania, although it is noticed that the influence of the new
investments is less significant in the power system of Romania than it is in Albania and southern
Serbia.
Planned new investments in 2015 make overall network performance better, especially in the region
where new planed investments are to be realized (Albania, southern Serbia and Montenegro), but do
4.69
not solve all contingences which are happening for analyzed generation and demand scenario.
Further internal network reinforcements, especially in the power systems of Romania, Albania and
Serbia, will be necessary in order to allow such generation dispatch with significant level of
network operation security.
4.70
5 LOAD
FLOW AND
SENSITIVITY CASES
CONTINGENCY
ANALYSIS
–
5.1
Introduction
In this chapter load-flow and security (n-1) analysis for sensitivity cases defined in Chapter 2 is
described. The analyzed network models are:
Special
conditions
Import/Export
Year
Topology
2010
2010
2010
2015
2010
2010
2015
2015
2010
High load
2015
The load-flow analysis includes line loading and voltage profile analysis, analysis of losses and also
analysis of power flows through interconnection lines.
The system reliability and adequacy is checked using “n-1” contingency criterion. List of
contingencies includes:
• all interconnection lines;
• all 400 and 220 kV lines in analyzed region, except lines which outage cause “island”
operation (in case of parallel and double circuit lines, outage of one line is considered);
• all transformers 400/x kV in analyzed region (in case of parallel transformers, outage of one
transformer is considered).
Current thermal limits are used as rated limits of lines and transformers, as described in Chapter 2.
Voltage limits are defined in Chapter 2, also.
Every branch with current above its thermal limit is treated as overloaded. States with overloaded
branches and/or voltages below or above defined voltage limits are treated as "insecure".
5.1 Scenario 2010 – average hydrology – import/export - 2010 topology
This part of the Study presents results of static load flow and voltage profile analyses that are
conducted for complete network topology and (n-1) contingencies in the scenario which is denoted
as Scenario 2010 – sensitivity case – average hydrology – import 1500 MW.
Concerning the import/export case, the simulated regime means the following:
▪
▪
▪
▪
Import 750 MW from UCTE
Import 500 MW from Turkey
Export 500 MW to Greece
Import 750 MW from Ukraine
5.1.1 Line loadings
Area totals and power exchanges for the 2010-sensitivity case-average hydrology-import 1500 MW
scenario are shown in Figure 5.1.1 and Table 5.1.1. Power flows along regional interconnection
lines and system balances are shown in Figure 5.1.2. Power flows along interconnection lines are
also given in Table 5.1.2, while Figure 5.1.3 shows histogram of tie lines loadings.
5.2
SVK
74
4 54
4
AUT
6
4 50
UKR
13
HUN
16
1
9
55
35
6
164
SLO
4
77
ITA
CRO
ROM
140
16
0
4
21
29
SRB
135
BIH
37
1
30
4 08
MNT
70
2
67
91
1
36
BUL
0
TUR
77
ALB
381
3
22
MKD
230
5
28
GRE
2 50
Figure 5.1.1 - Area exchanges in analyzed electric power systems for 2010-sensitivity case-average hydrology-import
1500 MW scenario
Table 5.1.1 - Area totals in analyzed electric power systems for 2010-sensitivity case-average hydrology-import 1500
MW scenario
Country
Albania
Generation
(MW)
898.2
Load
(MW)
1288.5
Bus Shunt
(Mvar)
0
Line Shunt
(Mvar)
0
Losses
(MW)
49.7
Net Interchange
(MW)
-440.0
Bulgaria
6827.0
5967.5
0
14.2
131.2
714.0
Bosnia and Herzegovina
2398.3
1965.0
0
0
54.3
379.0
Croatia
1705.2
3143.3
0
0
44.9
-1483.0
Macedonia
965.5
1178.3
0
0
20.1
-233.0
Romania
6877.4
6733.1
0
80.0
169.1
-104.8
Serbia and UNMIK
6591.7
6901.7
0
13.4
198.6
-521.9
Montenegro
542.1
670.2
0.5
1.5
16.9
-147.0
26805.4
27847.6
0.5
109.1
684.8
-1836.7
TOTAL - SE EUROPE
By comparing the average hydrology situation in 2010 and balanced SE Europe power system to
the average hydrology situation and 1500 MW of power import, significant increase of power flows
is noticed along Slovenian-Croatian, Ukrainian-Romanian and Bosnian-Montenegrin interfaces.
However, interconnection lines are not jeopardised since they are loaded far below their thermal
ratings. Power flows through Bosnian-Croatian, Serbian-Bosnian, Serbian-Croatian, SerbianMontenegrin and Grecian-Albanian interfaces are decreased at the same time.
5.3
Power losses, compared to the situation of balanced SE Europe power system, are increased in the
power systems of Bulgaria (7.9 %) and Montenegro (9 %). In other power systems these losses are
decreased, with the most significant drop in Romania (-16 %) and Serbia and UNMIK (-10.1 %).
Regional power losses are decreased (-7.1 %) when the situation includes additional power import.
Figure 5.1.3 shows that the tie lines in the region are mostly loaded less than 25% of their thermal
limits for the analyzed import/export scenario in year 2010. Among total number of forty nine 400
kV and 220 kV interconnection lines in the region only seven are loaded between 25% and 50% of
their thermal ratings. Only one line (OHL 400 kV Sofia – Nis between Bulgaria and Serbia) is
loaded more than 50% of its thermal rating, which is set at lower value (692.8 MVA) on the
Bulgarian side compared to the line rating on the Serbian side (1330.2 MVA).
Table 5.1.3 lists all network elements loaded over 80% of their thermal limits. As it can be seen
from this output list, most of the elements loaded over 80% are transformers in some substations
and internal 110 kV and 220 kV lines. Thus, certain internal network reinforcements are necessary
to sustain given load-demand level and generation pattern including import of 1500 MW from
analyzed directions.
Figure 5.1.4 shows histogram of 400 kV and 220 kV regional internal lines and 400/x kV and 220/x
kV transformers loadings. Some 46% of observed branches are loaded below 25% of their thermal
ratings, 36% are loaded between 25% and 50%, 16% are loaded between 50% and 75% and only
1% of observed branches are loaded between 75% and 100% of their thermal ratings. Two branches
(transformers 220/110 kV Fierze in Albania, 102% - 106% Sn) are overloaded when all branches are
in operation for the analyzed scenario.
By comparing the average hydrology situation in 2010 and balanced SE Europe power system to
the average hydrology situation and 1500 MW of power import it is noticed that power system of
Bosnia and Herzegovina is slightly relieved, while some highly loaded branches appear in the
power system of Bulgaria. Distribution of internal branches loading stays almost the same.
5.4
LEVIC 1400.0
CENTREL
200
148
251
113
GABICK1400.0
450.0 MW
UCTE
288
38.6
405.0
OKAINA1 402.6
233.5
224.4
UMUKACH
412.0
457
39.
8
MSAFA 4
226.3
JSOMB31
398.4
139
38.6
409.0
73.1 ARAD
26.9
407.0
ROM
-104.9 MW
JSUBO31 400.6
404.9
JHDJE11
159
54.1
160
53.1
P.D.FIE
TANTAREN
407.7
121
17.9
JHPIVA21
73.9 235.2
6.1
104
13.4
JHPERU21
103 228.3
16.4
CRG
-147.0 MW
405
149
3
45 6.1
.9
46.0
1.3
131
108
SK 1 219.1
BITOLA 2
C_MOGILA
SK 4
401.3
414.1
409.6
AZEMLA1
284
148
KARDIA K
287 414.7
102
QES/H
66.4
18.5
AHS_FLWR415.8
K
FILIPPOI
411.0
KV: 110 , 220 , 400
230
54.4
K-KEXRU
417.2
4HAMITAX
120
120 414.2
90.9
231 80.3
4BABAESK
96.8
414.4
500.0 MW
-500.1 MW
KARACQOU
250 416.3
50.0
414.6
2
31 1.3
6
1
9. 49
3
0
15 06
1
1
10. 2
43.
.5
67 .8
44
LAGAD K 412.1
397.4
MI_3_4_1 419.2
BLAGOEV412.9
409.4
0
12 .2
4
1
10. 6
65.
ALB
-440.0 MW
412.4
2
22 .3
42
STIP 1
DUBROVO
384.3
4
22 2
.
38
403.3
66.3
42.2
145
27.5
51.1
90.5
7.6 2
10.
7.6 2
10.
393.2
AVDEJA2
226.4AFIERZ2226.7
MKD
130
145
714.0 MW
JVRAN31
0.000
-233.0 MW
AKASHA1
BUL
JLESK21
382.6
JPRIZ22
219.0 JTKOSA2
221.4JTKOSB1395.7
3
53 5.
.1 7
394.7
DOBRUD4 420.7
415.6
410 SOFIA_W4
128
410.7
217
JPODG211
113
JPODG121227.1
51.2
43.6
218
78.7
AEC_400
JNIS2 1
392.5
120
80.3
73.6
17.7
-668.9 MW
SRB
-521.9 MW
7.6
20.7
RP TREB
232.3
JVARDI22
14.1 230.9
61.3
7.6
20.7
SA 20
232.7
14.0
59.1
46.2
7.0
VISEGRA
233.0
199
45.6
SCG
233.9
AVDEJA1
GIS REGIONAL MODEL - AVERAGE HYDRO
2010 SENSITIVITY CASE - IMPORT 1500 MW
416.2
UGLJEVIK 406.6
RP TREB
402.9
SHAW POWER
TECHNOLOGIES
INC. R
ISACCEA 414.5
140
JSMIT21
89.2
145
2.6
409.6
61.2 MO-4
14.0
235.1
BIH
379.0 MW
406.1
73.0
33.3
3
11
.8
23
3
11 8
.
23
MO-4
228.8
TE TUZL
3
40. 6
60.
60.7
2.5
81.4
50.3
226.5
ROSIORI 410.4
197
93.9
ERNESTIN
5
10 8
.
35
81.3
14.2
9
58. 9
17.
.2
14 .2
17
PRIJED2
GRADACAC
87.5
NADAB
39.4
407.8
114
25.5
40.3
8.7
14.1
8.2
227.0
87.4
20.0
114
25.5
DAKOVO
227.4MEDURIC
40.5
16.3
CRO
-1483.0 MW
410.6
374
50.7
4
16 5
10
407.5
HE ZAKUC
234.6
370
22.4
31
40. 5
0
404.4
MRACLIN
KONJSKO
10.9
29.3
67.7
48.6
TUMBRI
410.3
57 121
.1
1
27 5
.
42
1
27 5
.
42
10.5
26.3
407.5
PEHLIN 233.9
11.1
7.7
UKR
1200.0 MW
163
128
.1
.1
51 .8 51 .8
69
69
ZERJAVIN
407.3
ZERJAVIN 230.5
.0 5
95 9.
2
MELINA
228.8
1
16 27
.7
1
16 27
.7
LCIRKO2
403.3
127
3
12 1.1
31 7
.1
10
36 .5
.3
35.9
UMUKACH2
28.1
51.3
38.6
51.3
38.6
MPECS 4
LKRSKO1
40.2
2.1
230.8
35.8
21.3
MBEKO 4
408.6
410.2
103
38.6
30.1
39.1
SLO
58.6
22.6
30.1
35.4
MHEVI 4
215.0 MW
87
403.0 11 .2
1
LDIVAC2
MSAJO 4
410.0
LMARIB1 403.9
87.0
87.9
231.4
102
72.0
45.6 228.6
19.9
96.1
34.1
IPDRV121
102
93.9
46.0
7.2
MTLOK 2
230.5
HUN
271
62.8
401.2
79.4
12.2
350
UZUKRA01
304
772.5
-1250.0 MW
271
62.8
IRDPV111
MKISV 2
226.3
408.2
MGYOR 2
80.6
0.8
106
25.
7
213
24.
7
LPODLO2229.7
MSAJI 408.2
4
346
800
452
0
22.
106
25.
7
ONEUSI2
232.8
LDIVAC1
MGOD
OWIEN 2
106
56.2
106
56.2
217
40.6
OOBERS2238.7
199
102
971.8 MW
MAISA 7
716.0
MGYOR 4
0
25 .2
287
91
75.9
403.4
OWIEN 1
GRE
TUR
BUS -VOLTAGE(KV)
BRANCH -MW/MVAR
EQUIPMENT -MW/MVAR
Figure 5.1.2 - Power flows along interconnection lines in the region for 2010-sensitivity case-average hydrology-import 1500 MW scenario
5.5
Table 5.1.2 - Power flows along regional interconnection lines for 2010-sensitivity case-average hydrology-import 1500
MW scenario
Power Flow
Interconnection line
OHL 400 kV
OHL 220 kV
OHL 220 kV
OHL 400 kV
OHL 400 kV
OHL 400 kV
OHL 400 kV
OHL 400 kV
OHL 220 kV
OHL 220 kV
OHL 220 kV
OHL 220 kV
OHL 220 kV
OHL 220 kV
OHL 220 kV
OHL 220 kV
OHL 220 kV
OHL 400 kV
OHL 400 kV
OHL 400 kV
OHL 400 kV
OHL 400 kV
OHL 400 kV
OHL 2x400 kV ckt.1
OHL 2x400 kV ckt.2
OHL 400 kV
OHL 2x400 kV ckt.1
OHL 2x400 kV ckt.2
OHL 2x400 kV ckt.1
OHL 2x400 kV ckt.2
OHL 2x400 kV ckt.1
OHL 2x400 kV ckt.2
OHL 400 kV
OHL 400 kV
OHL 220 kV
OHL 220 kV
OHL 400 kV
OHL 400 kV
OHL 400 kV
OHL 2x220 kV ckt.1
OHL 2x220 kV ckt.2
OHL 400 kV
OHL 400 kV OHL 400 kV
OHL 400 kV
OHL 400 kV
OHL 400 kV
OHL 220 kV
OHL 220 kV
Zemlak (ALB)
Fierze (ALB)
V.Dejes (ALB)
V.Dejes (ALB)
Ugljevik (B&H)
Mostar (B&H)
Ugljevik (B&H)
Trebinje (B&H)
Trebinje (B&H)
Prijedor (B&H)
Prijedor (B&H)
Gradacac (B&H)
Tuzla (B&H)
Mostar (B&H)
Visegrad (B&H)
Sarajevo 20 (B&H)
Trebinje (B&H)
Blagoevgrad (BUL)
M.East 3 (BUL)
M.East 3 (BUL)
M.East 3 (BUL)
C.Mogila (BUL)
Dobrudja (BUL)
Kozloduy (BUL)
Kozloduy (BUL)
Sofia West (BUL)
Zerjavinec (CRO)
Zerjavinec (CRO)
Ernestinovo (CRO)
Ernestinovo (CRO)
Tumbri (CRO)
Tumbri (CRO)
Melina (CRO)
Ernestinovo (CRO)
Zerjavinec (CRO)
Pehlin (CRO)
Dubrovo (MCD)
Bitola (MCD)
Skopje (MCD)
Skopje (MCD)
Skopje (MCD)
Arad (ROM)
Nadab (ROM)
Rosiori (ROM)
Portile De Fier (ROM)
Subotica (SER)
Ribarevine (MON)
Pljevlja (MON)
Pljevlja (MON)
Kardia (GRE)
Prizren (SER)
Podgorica (MON)
Podgorica (MON)
Ernestinovo (CRO)
Konjsko (CRO)
S. Mitrovica (SER)
Podgorica (MON)
Plat (CRO)
Mraclin (CRO)
Medjuric (CRO)
Djakovo (CRO)
Djakovo (CRO)
Zakucac (CRO)
Vardiste (SER)
Piva (MON)
Perucica (MON)
Thessaloniki (GRE)
Filippi (GRE)
Babaeski (TUR)
Hamitabat (TUR)
Stip (MCD)
Isaccea (ROM)
Tantarena (ROM)
Tantarena (ROM)
Nis (SER)
Heviz (HUN)
Heviz (HUN)
Pecs (HUN)
Pecs (HUN)
Krsko (SLO)
Krsko (SLO)
Divaca (SLO)
S.Mitrovica (SER)
Cirkovce (SLO)
Divaca (SLO)
Thessaloniki (GRE)
Florina (GRE)
Kosovo B (UNMIK)
Kosovo A (UNMIK)
Kosovo A (UNMIK)
Sandorfalva (HUN)
Bekescaba (HUN)
Mukacevo (UKR)
Djerdap (SER)
Sandorfalva (HUN)
Kosovo B (UNMIK)
Bajina Basta (SER)
Pozega (SER)
MW
-283.7
36.1
-46.0
-144.6
40.3
81.4
-120.9
218.2
-104.8
38397
40.5
58.9
104.6
61.2
-14.0
-73.6
103.7
67.7
314.9
-119.9
-149.5
223.5
198.6
-113.5
-113.5
409.7
-51.1
-51.1
-126.8
-126.8
-270.6
-270.6
-87.0
-139.2
-95.0
-10.5
-66.3
-10.1
51.2
7.6
7.6
73.1
87.5
-452.1
159.5
-163.4
-38.0
-73.1
30.9
Mvar
-148.0
45.9
-1.3
-27.5
-60.6
-50.3
17.9
78.7
40.1
-17.2
-16.3
17.9
35.8
-14.0
59.1
-17.7
13.4
-48.6
-40.0
4.2
9.3
-38.2
-45.6
-23.8
-23.8
128.1
-69.8
-69.8
-31.1
-31.1
42.5
42.5
73.5
38.6
29.5
26.3
-42.2
-65.6
43.6
-20.7
-20.7
-26.9
-39.4
22.0
53.1
-127.7
12.2
5.8
31.4
% of thermal
rating
24
21
16
11
6
7
10
19
36
7
13
21
36
19
20
19
33
12
44
11
10
32
15
9
9
60
6
6
10
10
24
24
12
12
32
11
6
5
8
7
7
14
8
38
12
17
5
24
17
5.6
45
40
Frequency
35
30
25
20
15
10
5
0
x<25
25<x<50
50<x<75
75<x<100
x>100
Bin
Figure 5.1.3 - Histogram of interconnection lines loadings for 2010-sensitivity case-average hydrology-import
1500 MW scenario ("Frequency" denotes number of lines and "Bin" denotes loading range in % of thermal limit)
Table 5.1.3 - Network elements loaded over 80% of thermal limits for 2010-sensitivity case-average hydrology-import
1500 MW scenario
AREA
ALB
BUL
ALB
ROM
ELEMENT
Lines
OHL 110 kV AFIERZ5-AFARRZ5
OHL 220 kV M.EAST-ST.ZAGORA
Transformers
TR 220/110 kV AFIER 2-AFIER 5 ckt.1
TR 220/110 kV AFIER 2-AFIER 5 ckt.2
TR 220/110 kV AFIER 2-AFIER 5 ckt.3
TR 220/110 kV FUNDENI-FUNDE2B
LOADING
MVA
RATING
MVA
PERCENT
61.7
188.8
68.0
228.6
90.7
82.6
117.1
96.0
91.6
173.9
120.0
90.0
90.0
200.0
97.6
106.6
101.8
86.9
400
350
Frequency
300
250
200
150
100
50
0
x<25
25<x<50
50<x<75
75<x<100
x>100
Bin
Figure 5.1.4 - Histogram of 400 kV and 220 kV regional lines loadings for 2010-sensitivity case-average hydrologyimport 1500 MW scenario ("Frequency" denotes number of lines and "Bin" denotes loading range in % of thermal
limit)
5.7
5.1.2 Voltage Profile in the Region
Voltage profile in the region within this scenario which is defined by given generation pattern and
power import is seen as satisfactory despite several appearances of certain bus voltage deviations.
The deviations are shown in Table 5.1.4, which includes only 400 kV and 220 kV network buses.
Table 5.1.4 - Bus voltage deviations for 2010-sensitivity case-average hydrology-import 1500 MW scenario, complete
network
Country
Node
ALBANIA
BOSNIA AND HERZEGOVINA
BULGARIA
CROATIA
MACEDONIA
MONTENEGRO
ROMANIA
SERBIA AND UNMIK
Voltages
pu
1.052
1.053
1.052
1.104
1.101
1.102
1.103
-
400 kV VARNA4
400 kV MARITSA EAST2
400 kV DOBRUD4
220 kV SESTRIMO
220 kV BPC_220
220 kV AEC_220
220 kV TECVARNA
-
kV
420.8
421.0
420.7
242.8
242.2
242.4
242.7
-
Bus voltage magnitudes which are found below permitted limits (90% Vnominal in 220 kV network
and 95% Vnominal in 400 kV network) are not detected in the analyzed scenario. Bus voltage
magnitudes that are found above permitted limits (110% Vnominal in 110 kV and 220 kV networks
and 105% Vnominal in 400 kV network) are detected only in Bulgaria, where tree 400 kV buses and
four 220 kV buses have the magnitudes slightly above permitted limits. Figure 5.1.5 shows
histogram of voltages in monitored 400 kV and 220 kV substations.
120
Frequency
100
80
60
40
20
x>1.1
1.05<x<1.1
1.025<x<1.05
1<x<1.025
0.975<x<1
0.95<x<0.975
0.9<x<0.95
x<0.9
0
Bin
Figure 5.1.5 - Histogram of voltages in monitored substations for 2010-sensitivity case-average hydrology-import
1500 MW scenario ("Frequency" denotes number of busses and "Bin" denotes voltage range in p.u.)
5.8
It should be emphasized that these results represent only a situation when additional devices
(transformer automatic tap changers, switchable shunts, etc.) are not used for voltage regulation.
Impacts of such devices, which exist in many points of the SEE regional transmission network,
need more comprehensive and thorough analysis.
5.1.3 Security (n-1) analysis
Results of security (n-1) analysis for the 2010-sensitivity case-average hydrology-import 1500 MW
scenario are presented in Tables 5.1.5 - 5.1.6.
Insecure system situations for given generation pattern and power import are detected in the power
systems of Albania, Bulgaria, Romania and Serbia.
Loss of 220 kV line between the Rrashbul and Tirana substations can cause overloading of 220 kV
line between the Elbassan and Fier substations in Albania. The opposite case is also found critical.
Loss of 400 kV line in Bulgaria, M.East 2-Bourgas can cause overloading of 400 kV line PlovdivM. East 3. Loss of 220 kV line M. East 2-G.Oryahovitsa can cause overloading of 220 kV line M.
East 2-St. Zagora.
Single outages of 400/110 kV transformers in the stations Brasov and Dirste in Romania are also
found critical, since in the model the second transformer 400/110 kV in the Brasov substation is
permanently switched out of operation.
Loss of OHL 220 kV in the Belgrade area can cause overloading of the parallel line. Loss of 220
kV line between pumped storage power plant Bajina Basta and SS Pozega is also found critical.
Loss of one 400/110 kV transformer in the Nis substation is critical due to possible overloading of
the other parallel one but this problem could be solved by dispatcher intervention.
Figure 5.1.6 shows geographical positions of critical elements in the analyzed scenario. A green
colour reveals 220 kV elements (line 220 kV or transformer 220/x kV), while a red one reveals 400
kV elements (line 400 kV or transformer 400/x kV).
According to the obtained and presented results, it may be concluded that certain reinforcements in
the internal networks of Romania, Bulgaria, Albania and Serbia are necessary shall this generation
pattern and 1500 MW of power import be made more secure. None of the identified congestions is
located at the border lines.
By comparing the average hydrology situation in 2010 and balanced SE Europe power system to
the average hydrology situation and 1500 MW of power import, it may be noticed that some critical
contingencies in the Romanian power system disappear, especially those connected with Mintia
substation, while some new contingencies appear in the power system of Bulgaria (lines around
Maritsa East substation). This is due to different dispatching conditions of DEVA 1 power plant in
Romania (disconnected in import/export scenario, dispatched with 850 MW in the base case) and
power import through Turkish-Bulgarian interface that goes through Maritsa East 3 substation.
5.9
Table 5.1.5 - Lines overloadings for 2010-sensitivity case-average hydrology-import 1500 MW scenario, single outages
Outage
Overloaded line(s)
OHL 220 kV AKASHA2-ARRAZH2
OHL 220 kV AELBS12-AFIER 2
OHL 400 kV BURGAS-MI_2_400
OHL 220 kV G_ORIAH-MI_2_220
OHL 220 kV JBGD172-JBGD8 22 ckt.1
OHL 220 kV JBBAST2-JPOZEG2
OHL 220 kV AELBS12-AFIER 2
OHL 220 kV AKASHA2-ARRAZH2
OHL 400 kV PLOVDIV4-MI_400
OHL 220 kV MI_2_220-ST.ZAGORA
OHL 220 kV JBGD172-JBGD8 22 ckt.2
OHL 220 kV JPOZEG2-JVARDI22
Loadings
MVA %
254.8 116.3
263.3 100.6
762.3 107.4
255.0 104.2
438.3 122.9
294.0 100.0
Country
ALBANIA
BULGARIA
SERBIA
Table 5.1.6 - Transformers overloadings for 2010-sensitivity case-average hydrology-import 1500 MW scenario, single
outages
Outage
Overloaded branch(es)
TR 400/220 kV BUC.S ckt.1
400/110 kV BRASOV
400/110 kV DIRSTE
400/110 kV NIS ckt.1
TR 400/220 kV BUC.S ckt.2
400/110 kV DIRSTE
400/110 kV BRASOV
400/110 kV NIS ckt.2
Loadings
MVA %
407.4 101.9
343.9 137.5
340.6 136.2
304.8 101.6
Country
ROMANIA
SERBIA
SVK
AUT
HUN
SLO
ROM
CRO
SRB
BIH
MNG
BUL
MKD
TUR
ALB
GRE
critical elements
400 kV line or 400/x kV transformer
220 kV line or 220/x kV transformer
Figure 5.1.6 - Geographical positions of the critical elements for 2010-sensitivity case-average hydrology-import 1500
MW scenario
5.10
5.2 Scenario 2010 – average hydrology – high load
This part of the Study presents the results of static load flow and voltage profile analyses that are
conducted for complete network topology and (n-1) contingencies in the scenario which is denoted
as GTmax run for year 2010 - average hydrology – high load and with new generation facilities
implemented. High load conditions denotes high load growth rate scenario. This scenario (higher
growth rate) is analyzed in order to evaluate network performance for "heavy" load conditions.
5.2.1 Lines loadings
Figure 5.2.1 shows power exchanges between areas for 2010-average hydrology high load scenario.
Power flows along interconnection lines in the region together with balances of the systems are
shown in Figure 5.2.2. Area totals are shown in Table 5.2.1 and comparison of the average hydro
regimes with normal load projection and high load projection is shown in
Table 5.2.2. As it can be seen, increase of load level for around 3.61% on regional level, causes
network losses to rise for up to 177 MW or 24%. This is consequence of higher network load and
somewhat lower voltage levels.
Figure 5.2.3 shows histogram of tie lines loadings. It can be concluded that the most of the tie lines
are loaded less than 25% of their thermal limits.
UKR
450
SVK
7
7
45
3
17
AUT
HUN
12
8
49
31
72
SLO
4
26
ITA
CRO
ROM
239
2
56
11
306
SRB
267
BIH
5
4
15
6
31
306
MNG
BUL
10
145
1
5
10
30
ALB
170
85
MKD
10
TUR
38
0
GRE
Figure 5.2.1 - Area exchanges in analyzed electric power systems for 2010-average hydrology high load scenario –
2010 topology
5.11
LEVIC 1400.0
CENTREL
200
140
251
116
GABICK1400.0
450.0 MW
101
23.6
405.5
OKAINA1 402.4
234.2
82.9
49.6
UKR
450.0 MW
224.4
UMUKACH
412.0
6.7
55.
8
JSOMB31
393.1
ROM
401.8
JHDJE11
115
26.2
115
25.2
P.D.FIE
TANTAREN
403.2
213
32.1
413.7
UGLJEVIK 403.8
JHPIVA21
160 234.9
12.8
27.8
19.3
JHPERU21
27.7 228.5
27.4
46.0
97.5
46.2
JPODG211
139
JPODG121227.5
CRG
-163.0 MW
858.0 MW
SK 4
407.3
BITOLA 2
416.3
412.1
378
158
KARDIA K
383 416.6
130
QES/H
31.1
23.4
AHS_FLWR417.5
AZEMLA1
K
MI_3_4_1 419.7
413.1
FILIPPOI
415.7
2
19 6.8
7
9.7
22.0
K-KEXRU
417.9
4HAMITAX
4.0
4.0 415.4
76.8
9.7 65.5
4BABAESK
71.1
415.6
0.0 MW
-0.1 MW
KARACQOU
250 419.0
50.0
5
11 .7
.6
8
5. .5
91
0.7
4
31.
.5
22 .2
46
LAGAD K 414.1
396.2
413.1
BLAGOEV413.8
412.2
31.1
38.7
0.7 1
54.
ALB
-484.0 MW
C_MOGILA
.3
85 .7
45
STIP 1
DUBROVO
381.7
.5
85 .2
47
408.2
4.0
65.5
122
6.3
11.5
28.0
SK 1 223.1
0 1
4. 5.
1
25.5
128
BUL
JLESK21
0.000
1
43 05
.0
27.2
5.4
25.8
89.1
AVDEJA2
228.1AFIERZ2227.9
388.8
MKD
KV: 110 , 220 , 400
DOBRUD4 418.2
414.9
307 SOFIA_W4
113
411.1
JVRAN31
0.000
-261.0 MW
AKASHA1
304
154
122
41.3
6
11. 4
17.
6
11. 4
17.
25.8
89.1
AEC_400
JNIS2 1
394.3
JPRIZ22
224.6 JTKOSA2
227.1JTKOSB1406.8
44 106
.9
392.9
561.3 MW
SRB
724.2 MW
11.5
28.0
RP TREB
231.9
159
12.8
27.1
3.7
SA 20
231.1
JVARDI22
54.0 229.3
66.7
312
25.7
SCG
53.7
65.2
25.8
62.6
VISEGRA
231.4
AVDEJA1
GIS REGIONAL MODEL - AVERAGE HYDRO
2010 HIGH LOAD - NEW GENERATION - TOPOLOGY 2010
309
88.4
-56.8 MW
JSUBO31 395.6
240
JSMIT21
97.6
233.2
RP TREB
401.1
SHAW POWER
TECHNOLOGIES
INC. R
ISACCEA 409.3
2
2. 2
4.
406.8
134 MO-4
15.2
234.0
BIH
141.1 MW
39.0 ARAD
59.1
399.4
2
2. 2
4.
MO-4
228.0
TE TUZL
401.7
39.0
0.1
2.2
46.4
239
55.6
406.2
2
65. 0
63.
132
14.8
309
46.2
GRADACAC
225.5
ROSIORI 403.9
2.2
46.4
225.2
.3
85 .5
40
307
3.0
5
43. 2
22.
.4
52 .5
26
PRIJED2
89.2
NADAB
81.1
401.1
19
44. 3
4
4.0
2.5
52.1
18.8
226.4
89.1
62.8
22.5
49.1
DAKOVO
226.5MEDURIC
4.0
10.9
CRO
-1095.9 MW
406.5
82.7
67.4
MSAFA 4
403.1
HE ZAKUC
232.2
10.7
29.2
.2
72 38
1
65 213
.6
403.5
MRACLIN
KONJSKO
10.8
7.6
72.5
162
ERNESTIN
65.0
5.9
31.0
13.9
0
17 1
.
22
0
17 1
.
22
TUMBRI
409.1
46.
3
7
46 0.9
70 .3
.9
.4
.4
61 .3 61 .3
72
72
ZERJAVIN
405.9
ZERJAVIN 229.6
406.1
PEHLIN 232.9
228.4
46
21 .2
.8
46
21 .2
.8
LCIRKO2
402.7
.9
17 8.1
MELINA
35.7
UMUKACH2
27.6
61.5
35.2
MPECS 4
LKRSKO1
30
23 .9
.7
35.5
20.8
MBEKO 4
403.9
409.4
84.3
45.0
230.5
61.5
35.2
SLO
43.3
27.4
26.8
54.3
MHEVI 4
215.0 MW
84
402.6 77 .8
.5
LDIVAC2
MSAJO 4
410.0
LMARIB1 403.6
84.9
53.8
231.2
26.8
50.7
170
27.3
21.4 229.0
26.9
17.9
17.9
IPDRV121
170
48.3
21.5
13.0
MTLOK 2
230.5
HUN
170
46.7
401.3
50.5
21.1
MKISV 2
226.3
350
UZUKRA01
297
772.5
-1249.9 MW
170
46.7
IRDPV111
407.6
346
804
6.9
118
LPODLO2229.7
MSAJI 408.1
4
MGYOR 2
51.0
6.1
44.
18. 3
4
156
39.
6
44.
18. 3
4
ONEUSI2
233.5
LDIVAC1
MGOD
OWIEN 2
44.3
49.7
44.3
49.7
158
43.2
OOBERS2238.7
199
94.5
UCTE
222.2 MW
MAISA 7
714.7
MGYOR 4
0
25 .6
101
93
102
403.5
OWIEN 1
GRE
TUR
BUS -VOLTAGE(KV)
BRANCH -MW/MVAR
EQUIPMENT -MW/MVAR
Figure 5.2.2 - Power flows along interconnection lines in the region with balances of the systems for 2010-average hydrology high load scenario – 2010 topology
5.12
Table 5.2.1 - Area totals in analyzed electric power systems for 2010- average hydrology high load scenario –
2010 topology
AREA
ALBANIA
BULGARIA
BIH
CROATIA
MACEDONIA
ROMANIA
SERBIA
MONTENEGRO
TOTALS
GENERATION
931.7
7282.5
2202.8
2254.6
991.1
7113.3
8103.8
542.3
29422.1
LOAD
1358
6278
2004
3295
1234
6859.4
7131
686
28845.4
LOSSES
57.7
146.5
57.8
55.4
18.1
310.6
248.6
19.2
913.9
INTERCHANGE
-484
858
141.1
-1095.9
-261
-56.7
724.2
-163
-337.3
Table 5.2.2 – Comparison of Area totals in analyzed electric power systems for 2010- average hydrology versus average
hydrology high load scenario – 2010 topology
LOAD
LOSSES
AREA
normal
normal
high load
high load
load
load
ALBANIA
1287.3
1358
5.49%
50.4
57.7
14.48%
BULGARIA
5977.3
6278
5.03%
121.6
146.5
20.48%
BIH
1971.3
2004
1.66%
58.6
57.8
-1.37%
CROATIA
3136.7
3295
5.05%
49
55.4
13.06%
MACEDONIA
1198.2
1234
2.99%
20.1
18.1
-9.95%
ROMANIA
6728.3
6859.4
1.95%
201.2
310.6
54.37%
SERBIA
6873.1
7131
3.75%
220.8
248.6
12.59%
MONTENEGRO
669.2
686
2.51%
15.5
19.2
23.87%
TOTALS
27841.4
28845.4
3.61%
737.1
913.9
23.99%
Histogram
45
40
Frequency
35
30
25
20
15
10
5
0
x<25
25<x<50
50<x<75 75<x<100
x>100
Bin
Figure 5.2.3 - Histogram of interconnection lines loadings for 2010- average hydrology high load scenario –
2010 topology ("Frequency" denotes number of lines and "Bin" denotes loading range in % of thermal limit)
Table 5.2.3 shows all network elements loaded over 80% of their thermal limits. As it can be seen
some lines 220 kV voltage level in Albania, Romania and Serbia are loaded over 80%. Also, most
of the elements loaded over 80% are transformers in some substations, again, in Albania, Romania
and Serbia. Figure 5.2.4 shows histogram of branch loadings in the system.
5.13
As for the conclusion regarding thermal loadings in this scenario it can be said that most of the
network elements are loaded up to 75% of their thermal limits, but there are some elements highly
loaded, even overloaded. It can be concluded that overall load of the system is higher than by
normal load projection. Higher load-demand level causes that increase of load of the elements
which supply large consumption areas are is higher than increase of load-demand level. This goes
especially for transformer units over which large consumption areas are connected to high voltage
network. Also, most of the elements loaded over 80% are transformers in some substations, and
loading of these elements is higher than in case of normal load projection. There are some elements
that are overloaded (220 kV lines Targu Jiu – Paroseni and Urechesti-Targu Jiu in Romania, and
220/110 kV transformers in Fier substation in Albania), and this load is higher than in case of
normal load projection. Like in previous scenarios analyzed, higher engagement of TPP Paroseni
resolves this overloads of the 220 kV lines Targu Jiu – Paroseni and Urechesti-Targu Jiu and
decreases the load of the 400/220 transformer in Urechesti substation. Also, it is expected that
transformers in Fierza substation will be replaced with more powerful transformer units.
Table 5.2.3 - Network elements loaded over 80% of their thermal limits for 2010-average hydrology high load scenario
– 2010 topology
BRANCH LOADINGS ABOVE
80.0 % OF RATING:
LOADING
ELEMENT
MVA
AREA
HL
HL
HL
HL
ROM
TR
TR
TR
TR
TR
TR
TR
TR
TR
ALB
ROM
SRB
Lines
P.D.F.A-RESITA 1
P.D.F.A-RESITA 2
TG.JIU-PAROSEN 1
URECHESI-TG.JIU 1
Transformers
220/110 kV AFIER 1
220/110 kV AFIER 2
220/110 kV AFIER 3
400/220 kV URECHE 1
220/110 kV FUNDE2 1
220/110 kV JTKOSA 2
220/110 kV JTKOSA 3
220/110 kV JZREN2 2
400/220 kV JBGD8 1
220kV
220kV
220kV
220kV
RATING
MVA
PERCENT
226.8
226.8
277.8
277.8
277.4
277.4
208.1
277.4
81.7
81.7
133.5
100.1
122.2
100.1
95.6
344.4
178.6
131.5
133.9
120.2
326.3
120
90
90
400
200
150
150
150
400
101.8
111.3
106.2
86.1
89.3
87.7
89.3
80.1
81.6
Histogram
350
300
Frequency
250
200
150
100
50
0
x<25
25<x<50 50<x<75 75<x<100
x>100
Bin
Figure 5.2.4 - Histogram of branch loadings for 2010- average hydrology high load scenario – 2010 topology
("Frequency" denotes number of lines and "Bin" denotes loading range in % of thermal limit)
5.14
5.2.2 Voltage Profile in the Region
Figure 5.2.5 shows histogram of voltages in monitored substations. Voltages in almost all
monitored substations are found within permitted limits. Voltage profile in network of Albania is
somewhat near low limits, but this can be resolved by changing of the setting of the tap changing
transformers in some substations. Also, voltage levels are somewhat lower comparing to normal
load projection scenario. Higher load of network elements, which is consequence of the higher
demand level, causes voltage drops along network elements to be higher.
Histogram
120
Frequency
100
80
60
40
20
x>1.1
1.05<x<1.1
1.025<x<1.05
1<x<1.025
0.975<x<1
0.95<x<0.975
0.9<x<0.95
x<0.9
0
Bin
Figure 5.2.5 - Histogram of voltages in monitored substations for 2010- average hydrology high load scenario –
2010 topology ("Frequency" denotes number of busses and "Bin" denotes voltage range in p.u.)
5.2.3 Security (n-1) analysis
Results of security (n-1) analysis for 2010-average hydrology high load scenario and expected
topology for 2010 are presented in Table 5.2.4. Figure 5.2.6 shows the geographical position of the
critical elements in monitored systems.
It can be concluded that all identified insecure situations are located in internal networks that belong
to monitored power systems of Albania, Romania and Serbia. In most critical case in Romanian
system, the critical elements are 220 kV lines Targu Jiu – Paroseni and Urechesti-Targu Jiu and
400/220 kV transformer in Urechesti substation, but these elements are overloaded in case of full
topology also, which is the main reason why they appear as critical by most outages analyzed. As it
has been stated before, this can be resolved by higher engagement of the TPP Paroseni. Compared
to the normal load projection (chapter 4.1) it can be concluded that the critical elements are almost
the same. The only difference is overload of the 400/110 kV transformer unit in SS Pancevo. This is
cosequence of higher load of these transformers in full network topology compared to the normal
demand scenario.
5.15
Like by normal demand scenario (chapter 4.1), some of the overloadings identified can be relieved
by certain dispatch actions (splitting busbars, changing lower voltage network topology in order to
redistribute load-demand or change of generation units engagement).
All in all, certain reinforcement of internal network is necessary in order to make this regime more
secure. Especially, taking into consideration that higher demand causes higher load of transformer
units in whole monitored region.
Table 5.2.4 - Network overloadings for 2010- average hydrology high load scenario , single outages – 2010 topology
Area
1
contingency
2
AL
BASE CASE
OHL 220kV AELBS12 -AFIER 2
1
RO
OHL 220kV P.D.F.A -RESITA
1
RO
OHL 220kV RESITA
-TIMIS
1
RO
OHL 220kV PESTIS
-MINTIA A 1
RO
OHL 220kV CLUJ FL -AL.JL
1
RO
OHL 220kV AL.JL
1
RO
OHL 400kV TANTAREN-URECHESI 1
RO
OHL 400kV TANTAREN-BRADU
1
RO
OHL 400kV TANTAREN-SIBIU
1
RO
OHL 400kV URECHESI-P.D.FIE
1
RO
OHL 400kV URECHESI-DOMNESTI 1
RO
OHL 400kV MINTIA
-ARAD
1
RO
OHL 400kV MINTIA
-SIBIU
1
RO
OHL 400kV DOMNESTI-BRAZI
1
RO
OHL 400kV SMIRDAN -GUTINAS
1
RO
OHL 400kV BRASOV
1
CS
OHL 220kV JBGD172 -JBGD8 22 1
CS
OHL 400kV JBOR 21 -JHDJE11
1
CS
OHL 400kV JHDJE11 -JTDRMN1
1
CS
OHL 400kV JNSAD31 -JSUBO31
1
RO
RO
CS
CS
TR
TR
TR
TR
400/110
400/110
400/110
400/110
-GILCEAG
-BRADU
BRASOV
1
DIRSTE
1
JNIS2 1
JPANC2 1
overloadings
/
out of limits voltages
Area
3
4
RO
HL 220kV TG.JIU-PAROSEN
AL
HL 220kV AKASHA2-ARRAZH2
RO
HL 220kV URECHESI-TG.JIU
RO
HL 220kV P.D.F.A-RESITA
RO
HL 220kV TG.JIU-PAROSEN
RO
HL 220kV URECHESI-TG.JIU
RO
HL 220kV RESITA-TIMIS
RO
HL 220kV TG.JIU-PAROSEN
RO
HL 220kV URECHESI-TG.JIU
RO
HL 220kV TG.JIU-PAROSEN
RO
HL 220kV TG.JIU-PAROSEN
RO
HL 220kV URECHESI-TG.JIU
RO
HL 220kV TG.JIU-PAROSEN
RO
HL 220kV URECHESI-TG.JIU
RO
HL 220kV URECHESI-TG.JIU
RO
HL 220kV TG.JIU-PAROSEN
RO
HL 220kV URECHESI-TG.JIU
RO
HL 220kV TG.JIU-PAROSEN
RO
HL 220kV TG.JIU-PAROSEN
RO
TR 400/220kV/kV URECHESI
RO
HL 220kV URECHESI-TG.JIU
RO
HL 220kV TG.JIU-PAROSEN
RO
HL 220kV TG.JIU-PAROSEN
RO
HL 220kV URECHESI-TG.JIU
RO
HL 220kV TG.JIU-PAROSEN
RO
HL 220kV URECHESI-TG.JIU
RO
HL 220kV TG.JIU-PAROSEN
RO
HL 220kV URECHESI-TG.JIU
RO
HL 220kV TG.JIU-PAROSEN
RO
HL 220kV URECHESI-TG.JIU
RO
HL 220kV TG.JIU-PAROSEN
CS
HL 220kV JBGD172-JBGD8 22
RO
HL 220kV URECHESI-TG.JIU
RO
HL 220kV TG.JIU-PAROSEN
RO
HL 220kV URECHESI-TG.JIU
RO
HL 220kV TG.JIU-PAROSEN
RO
HL 220kV URECHESI-TG.JIU
RO
HL 220kV TG.JIU-PAROSEN
RO
TR 400/110kV/kV DIRSTE
RO
TR 400/110kV/kV BRASOV
CS
TR 400/110kV/kV JNIS2 1
CS
TR 400/110kV/kV JPANC21
#
5
1
1
1
2
1
1
2
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
1
1
1
1
1
1
1
1
2
2
limit
/
Unom
6
208.1MVA
270MVA
277.4MVA
277.4MVA
208.1MVA
277.4MVA
277.4MVA
208.1MVA
277.4MVA
208.1MVA
208.1MVA
277.4MVA
208.1MVA
277.4MVA
277.4MVA
208.1MVA
277.4MVA
208.1MVA
208.1MVA
400MVA
277.4MVA
208.1MVA
208.1MVA
277.4MVA
208.1MVA
277.4MVA
208.1MVA
277.4MVA
208.1MVA
277.4MVA
208.1MVA
365.8MVA
277.4MVA
208.1MVA
277.4MVA
208.1MVA
277.4MVA
208.1MVA
250MVA
250MVA
300MVA
300MVA
rate
Flow
/
/
Voltage volt.dev.
7
8
285.6MVA
133.5%
289.9MVA
112.2%
301.5MVA
106.1%
331.5MVA
122.8%
301.5MVA
141.4%
295.7MVA
103.9%
348.7MVA
127.0%
295.7MVA
138.5%
296.3MVA
104.8%
290.4MVA
139.7%
271MVA
126.4%
308.8MVA
109.1%
308.8MVA
145.4%
292.5MVA
102.6%
300.1MVA
105.9%
300.1MVA
141.1%
331.1MVA
117.3%
331.1MVA
156.4%
268.7MVA
124.5%
400.6MVA
100.1%
305.6MVA
107.7%
305.6MVA
143.6%
263.1MVA
123.1%
326.2MVA
115.5%
326.2MVA
153.9%
296.2MVA
104.0%
296.2MVA
138.7%
296.9MVA
104.5%
296.9MVA
139.3%
296.3MVA
104.3%
296.3MVA
139.0%
453.5MVA
129.0%
300.9MVA
106.1%
300.9MVA
141.4%
315.8MVA
111.9%
315.8MVA
149.1%
304.1MVA
107.2%
304.1MVA
142.8%
353.9MVA
141.6%
350MVA
140.0%
323.1MVA
107.7%
318.8MVA
106.3%
5.16
SVK
AUT
HUN
SLO
ROM
CRO
SRB
BIH
MNG
BUL
MKD
TUR
ALB
GRE
critical elements
400 kV line or 400/x kV transformer
220 kV line or 220/x kV transformer
Figure 5.2.6 – geographical position of critical elements for 2010- average hydrology high load scenario –
2010 topology
5.17
5.3 Scenario 2015 – average hydrology – import/export – 2010 topology
This part of the Study presents results of static load flow and voltage profile analyses that are
conducted for complete network topology and (n-1) contingencies in the scenario which is denoted
as Scenario 2015 – sensitivity case – average hydrology – import 1500 MW – topology 2010.
Concerning the import/export case, the simulated regime means the following:
▪
▪
▪
▪
Import 750 MW from UCTE
Import 500 MW from Turkey
Export 500 MW to Greece
Import 750 MW from Ukraine
Calculations on the model characterized by the 2010 network topology, appropriate GTMax
generation dispatch, predicted load level in 2015 and power import/export as explained above,
could not lead to a convergent load flow solution due to reactive power problems. Lack of reactive
power in the analyzed situation is visible especially in the power system of Albania.
If the power plants reactive limits are ignored in the model, a convergent solution is found and the
following can be concluded:
▪
There is a permanent lack of reactive power for analyzed generation dispatch, load level
and power import/export in the power system of Albania. Generators in power plants
Balsh, Fierza and Vau Dejes exceed their Var limits in order to keep nominal voltages.
This proves the necessity to install compensation devices or new power plants in the
Albanian power system, or to construct new interconnection lines to make network
connections stronger.
▪
Generators in Bulgarian power system, which are dispatched in the analyzed scenario
and whose reactive power limits are exceeded, are operated mostly in the underexcitation area below their minimum Var limits. This means that there is a sufficient
reactive power reserve, but problems with high voltages may be expected especially in
lower bus loading conditions (nights, summer).
▪
Reactive power problems in the analyzed situation are not detected in the power systems
of Bosnia and Herzegovina, Croatia and Macedonia.
▪
In the power system of Romania, the generator Var limits are occasionally exceeded in
both directions (some generators work in the over-excitation area and some in the underexcitation one) in order to achieve scheduled (mostly nominal) voltages in the analyzed
situation. The Var limits are not significantly exceeded when observing all dispatched
generators in Romania.
▪
Certain lack of reactive power is visibly present in the power system of Serbia and
UNMIK. Ten power plants are operated slightly above their maximum Var limits.
▪
Smaller lack of reactive power also exists in the power system of Montenegro that is
affected by the poor voltage profile in Albania.
5.18
To provide deeper insight into the voltage problems which are obviously present in the analyzed
export/import scenario in 2015, but on the network topology predicted to exist in 2010, it is
necessary to conduct more comprehensive and thorough analysis. It is assumed that the utilisation
of existing devices such as transformer automatic tap changers and switchable shunts, or the
generation re-dispatch may mitigate these voltage problems without a need to construct new lines.
If new interconnection lines predicted to exist in 2015 are included in the model, Vranje (Serbia) –
Skopje 4 (Macedonia), Zemlak (Albania) – Bitola 2 (Macedonia), and V. Dejes (Albania) – Kosovo
B (UNMIK), a convergent solution is found. Power flow solution is explained in the following
Chapter.
A convergent solution for the 2010 network topology is found when only 400 kV line V.Dejes
(Albania) – Kosovo B (UNMIK) is included in the model, but without satisfactory network voltage
profile. Two 400 kV nodes in Albania (Elbassan and Tirana) and five 400 kV nodes in Romania
(Domnesti, Dirste, Brazi, Brasov, Bradu) have small voltage values (between 374 kV and 379 kV).
Two 220 kV nodes in Albania (Fier and Babic) have also small voltage values (193 kV and 196 kV
respectively). Sixty two 110 kV nodes, mostly in Albania, have low voltages, ranging between 80
kV and 99 kV.
Convergent solution on network topology in 2010 is found also if only 400 kV line Zemlak
(Albania) – Bitola (Macedonia) is included on the model, but voltage profile in the network is
unsatisfactory again. Two 400 kV nodes in Albania (Elbassan and Tirana) and five 400 kV nodes in
Romania (Domnesti, Dirste, Brazi, Brasov, Bradu) have small voltage values (between 373 kV and
379 kV). Two 220 kV nodes in Albania (Fier and Babic) have small voltage values also (192 kV
and 195 kV respectively). Seventy five 110 kV nodes, mostly in Albania, have low voltages,
ranging between 78 kV and 99 kV.
To conclude, the analyzed scenario which is characterized by large power import in 2015, but on
the 2010 network topology, can not be supported from a transmission network viewpoint due to
voltage problems. Their existence is the most obvious in the power system of Albania due to scarce
reactive power sources. To mitigate such problems it is necessary to construct at least one new 400
kV interconnection line between Albania and UNMIK or Macedonia.
5.19
5.4 Scenario 2015 – average hydrology – import/export – 2015 topology
This part of the Study presents results of static load flow and voltage profile analyses that are
conducted for complete network topology and (n-1) contingencies in the scenario which is denoted
as Scenario 2015 – sensitivity case – average hydrology – import 1500 MW – topology 2015.
5.4.1 Lines loadings
Area totals and power exchanges for the 2015-sensitivity case-average hydrology-import 1500
MW-topology 2015 scenario are shown in Figure 5.4.1. Power flows along interconnection lines in
the region together with balances of the systems are shown in Figure 5.4.2. Area totals are shown in
Table 5.4.1. Table 5.4.2 shows power flows along regional interconnection lines for 2015sensitivity case-average hydrology-import 1500 MW-topology 2015 scenario
SVK
68
5 20
3
AUT
0
4 50
UKR
61
HUN
62
4
61
17
8
1
SLO
9
82
ITA
CRO
ROM
175
0
3
36
390
SRB
245
BIH
20
88
3 36
MNT
88
2
162
28
26
4
BUL
7
28
217
ALB
352
7
21
MKD
TUR
212
65
GRE
2 50
Figure 5.4.1 - Area exchanges in analyzed electric power systems for 2010-sensitivity case-average hydrology-import
1500 MW-topology 2015 scenario
5.20
Table 5.4.1 - Area totals in analyzed electric power systems for 2015-sensitivity case-average hydrology-import 1500
MW-topology 2015 scenario
Country
Albania
Generation
(MW)
1027.5
Load
(MW)
1544.0
Bus Shunt
(Mvar)
0
Line Shunt
(Mvar)
0
Losses
(MW)
71.5
Net Interchange
(MW)
-588.0
Bulgaria
7582.2
6418.9
0
14.6
142.7
1006.0
Bosnia and Herzegovina
2394.9
2293.8
0
0
70.1
31.0
Croatia
2173.9
3660.1
0
0
58.8
-1545.0
Macedonia
1076.3
1393.7
0
0
22.5
-340.0
Romania
7187.7
7831.2
0
74.6
253.9
-972.0
Serbia and UNMIK
8029.8
7270.7
0
14.1
215.1
530.0
Montenegro
731.9
675.0
0.6
1.8
15.6
39.0
30204.2
31087.4
0.6
105.0
850.3
-1839.0
TOTAL - SE EUROPE
By comparing the average hydrology situation in 2015 and balanced SE Europe power system to
the average hydrology situation and 1500 MW of power import, a significant increase of the power
flows is noticed along the Slovenian-Croatian and Ukrainian-Romanian interfaces. However, the
interconnection lines are not jeopardized since their loading levels fall far below their thermal
ratings. In the same time, power flows through other interfaces between countries in the SE Europe
are mostly decreased.
Power losses, in comparison to the situation of balanced SE Europe power system, are increased
only in the power system of Macedonia (5%). In other power systems power losses are decreased,
the most significantly in the power systems of Romania (-26%) and Montenegro (-23%). Power
losses are decreased (-16%) in the region when analyzing the additional power import.
Figure 5.4.3 shows that all tie lines in the region are loaded less than 50% of their thermal limits for
the analyzed import/export scenario in year 2015. Among total number of fifty two 400 kV and 220
kV interconnection lines in the region only ten are loaded between 25% and 50% of their thermal
ratings. Other tie lines are loaded less than 25% of their thermal ratings.
Table 5.4.3 lists all network elements loaded over 80% of their thermal limits. As it can be seen
from this output list, most of the elements loaded over 80% are transformers in some substations
and internal 110 kV and 220 kV lines. 110 kV lines which are loaded over 80% Ithermal are not
shown in the table. Certain internal network reinforcements are necessary to sustain given loaddemand level and production pattern including import of 1500 MW from analyzed directions.
Figure 5.4.4 shows histogram of 400 kV and 220 kV regional internal lines and 400/x kV and 220/x
kV transformers loadings. Some 40% of observed branches are loaded below 25% of their thermal
ratings, 36% are loaded between 25% and 50%, 20% are loaded between 50% and 75% and 5% of
observed branches are loaded between 75% and 100% of their thermal ratings. Four branches
(transformers 220/110 kV Fierze in Albania, 102% - 106% Sn and 220/110 kV transformer in
Fundeni substation in Romania) are overloaded when all branches are in operation.
By comparing the average hydrology situation in 2015 and balanced SE Europe power system to
the average hydrology situation and 1500 MW of power import, it is noticed that power system of
Romania becomes slightly relieved while some highly loaded branches are present in the power
systems of Albania, Serbia and Bulgaria. Distribution of internal branches loading is slightly
changed because a number of branches loaded between 25% and 75% of their thermal ratings is
increased compared to the situation characterized by balanced SE Europe power system.
5.21
LEVIC 1400.0
CENTREL
200
140
251
102
GABICK1400.0
450.0 MW
244
38.8
404.4
233.3
121
25.
3
224.4
UMUKACH
412.0
510
66.
5
MSAFA 4
ROM
JHDJE11
P.D.FIE
2.1
31.2
2.1
32.3
302
107
305 SOFIA_W4
62.6
410.0
TANTAREN
402.0
204
2.8
42.0
11.2
JHPERU21
41.8 229.9
19.0
0.6
43.1
0.7
JPODG211
87.8
JPODG121229.2
CRG
39.0 MW
JPRIZ22
222.9 JTKOSA2
227.3JTKOSB1408.0
SK 4
404.5
-340.0 MW
MKD
111
119
392.6
ALB
408.7
DUBROVO
27
75 9
.7
AHS_FLWR415.3
AZEMLA1
58.6
111
KARDIA K
59.0 415.4
40.8
QES/H
K
411.2
FILIPPOI
414.6
9
28 6.7
5
214
55.6
K-KEXRU
417.2
KV: 110 , 220 , 400
4HAMITAX
128
128 414.0
84.4
214 73.9
4BABAESK
98.8
414.3
500.0 MW
-500.2 MW
KARACQOU
250 416.7
50.0
1
1. 58
9
9
15 .7
97
.4
50 .9
39
LAGAD K 412.2
406.4
MI_3_4_1 418.0
BLAGOEV412.2
408.8
184
5
82.
-588.0 MW
412.0
411.6
6
19 .0
47
STIP 1
184
102
AKASHA1
280
7
39.
C_MOGILA
7
19 0
.
36
404.8
7
12 .6
1
111
119
BITOLA 2
1006.0 MW
JVRAN31
401.8
61.8
59.9
SK 1 222.5
DOBRUD4 413.8
413.2
BUL
JLESK21
398.6
6
14 1.
.2 9
1
29 7.1
.0
AVDEJA2
227.7AFIERZ2227.9
399.4
11.1
32.8
9.7
15.2
33.3
23.9
290 0
47.
111
78.7
111
78.7
218
5.8
3
11. 3
22.
289
68.7
3
11. 3
22.
1
37 6.
.8 9
399.6
AEC_400
JNIS2 1
398.3
128
73.9
JHPIVA21
158 237.3
34.1
32.7 32.8
37.2 74.4
156
34.2
569.0 MW
SRB
530.0 MW
61.0
50.9
RP TREB
232.6
JVARDI22
28.9 231.2
41.1
217
47.0
SA 20
230.8
28.8
38.7
11.1
32.8
VISEGRA
232.5
396
3.9
SCG
230.8
AVDEJA1
GIS REGIONAL MODEL - AVERAGE HYDRO
2015 SENSITIVITY CASE - IMPORT 1500 MW - TOPOLOGY 2015
410.2
UGLJEVIK 400.3
RP TREB
404.3
SHAW POWER
TECHNOLOGIES
INC. R
391
88.3
-972.0 MW
JSUBO31 395.8
400.5
9.8
5.6
405.9
113 MO-4
10.9
233.2
BIH
31.0 MW
ISACCEA 401.3
149
JSMIT21
74.3
33.3
4.6
MO-4
18.6 ARAD
54.7
399.3
.5
28 .1
27
.5
28 .1
27
112
6.7
228
30.6
224.7
TE TUZL
401.6
18.6
4.7
28.5
76.6
221.8
JSOMB31
393.2
148
25.3
403.5
8
13. 5
63.
227
21.9
219.6
ROSIORI 400.5
28.5
76.6
ERNESTIN
1
10 3
.
45
5.9
3.2
.0
16 .2
20
PRIJED2
GRADACAC
6.1
NADAB
77.6
400.5
29
48. 1
8
5.9
11.1
16.0
11.8
218.6
6.0
58.5
50.5
53.9
DAKOVO
220.8MEDURIC
0
52. 8
22.
CRO
-1545.0 MW
404.4
321
52.6
.0
33 22
1
402.1
HE ZAKUC
231.0
318
32.6
181 181
14.1 32.7
399.8
MRACLIN
407.7
31 204
.1
37.4
4.6
5
26 5
.
11
5
26 5
.
11
TUMBRI
KONJSKO
10.9
29.2
61.1
9.8
.7
.7
77 .8 77 .8
86
86
ZERJAVIN
401.3
ZERJAVIN 225.7
402.4
PEHLIN 229.8
226.3
.8
74 6.5
MELINA
11.0
7.6
UKR
1200.0 MW
32.8
147
LCIRKO2
399.9
25
35 .5
.4
25
35 .5
.4
37
14 .5
.2
35.9
UMUKACH2
27.7
77.9
17.8
77.9
17.8
MPECS 4
LKRSKO1
25.
5
8
25 4.1
84 .5
.1
229.3
35.8
21.0
MBEKO 4
403.5
407.2
13.8
6.5
49.9
67.1
SLO
51.8
27.5
49.9
63.7
MHEVI 4
215.0 MW
10
400.8 60 7
.1
LDIVAC2
MSAJO 4
410.0
LMARIB1 400.9
107
36.9
230.3
108
3.4
40.7 228.4
21.1
75.5
13.0
IPDRV121
108
25.4
41.0
8.1
MTLOK 2
230.5
HUN
265
31.6
400.6
73.5
13.7
350
UZUKRA01
296
772.5
-1250.0 MW
265
31.6
IRDPV111
MKISV 2
226.3
407.6
MGYOR 2
74.5
1.5
99.5
47.5
LPODLO2227.8
MSAJI 408.2
4
346
805
504
9
68.
121
25.
3
214
21.
9
ONEUSI2
232.6
LDIVAC1
MGOD
OWIEN 2
121
55.0
121
55.0
OKAINA1 399.7
218
38.7
OOBERS2236.5
199
94.3
UCTE
974.2 MW
MAISA 7
714.7
MGYOR 4
0
25 .7
243
79
78.8
402.9
OWIEN 1
GRE
TUR
BUS -VOLTAGE(KV)
BRANCH -MW/MVAR
EQUIPMENT -MW/MVAR
Figure 5.4.2 - Power flows along interconnection lines in the region for 2015-sensitivity case-average hydrology-import 1500 MW-topology 2015 scenario
5.22
Table 5.4.2 - Power flows along regional interconnection lines for 2015-sensitivity case-average hydrology-import 1500
MW-topology 2015 scenario
Interconnection line
OHL 400 kV
Zemlak (ALB)
OHL 220 kV
Fierze (ALB)
OHL 220 kV
V.Dejes (ALB)
OHL 400 kV
V.Dejes (ALB)
OHL 400 kV
Ugljevik (B&H)
OHL 400 kV
Mostar (B&H)
OHL 400 kV
Ugljevik (B&H)
OHL 400 kV
Trebinje (B&H)
OHL 220 kV
Trebinje (B&H)
OHL 220 kV
Prijedor (B&H)
OHL 220 kV
Prijedor (B&H)
OHL 220 kV
Gradacac (B&H)
OHL 220 kV
Tuzla (B&H)
OHL 220 kV
Mostar (B&H)
OHL 220 kV
Visegrad (B&H)
OHL 220 kV
Sarajevo 20 (B&H)
OHL 220 kV
Trebinje (B&H)
OHL 400 kV
Blagoevgrad (BUL)
OHL 400 kV
M.East 3 (BUL)
OHL 400 kV
M.East 3 (BUL)
OHL 400 kV
M.East 3 (BUL)
OHL 400 kV
C.Mogila (BUL)
OHL 400 kV
Dobrudja (BUL)
OHL 2x400 kV ckt.1
Kozloduy (BUL)
OHL 2x400 kV ckt.2
Kozloduy (BUL)
OHL 400 kV
Sofia West (BUL)
OHL 2x400 kV ckt.1
Zerjavinec (CRO)
OHL 2x400 kV ckt.2
Zerjavinec (CRO)
OHL 2x400 kV ckt.1
Ernestinovo (CRO)
OHL 2x400 kV ckt.2
Ernestinovo (CRO)
OHL 2x400 kV ckt.1
Tumbri (CRO)
OHL 2x400 kV ckt.2
Tumbri (CRO)
OHL 400 kV
Melina (CRO)
OHL 400 kV
Ernestinovo (CRO)
OHL 220 kV
Zerjavinec (CRO)
OHL 220 kV
Pehlin (CRO)
OHL 400 kV
Dubrovo (MCD)
OHL 400 kV
Bitola (MCD)
OHL 400 kV
Skopje (MCD)
OHL 2x220 kV ckt.1
Skopje (MCD)
OHL 2x220 kV ckt.2
Skopje (MCD)
OHL 400 kV
Arad (ROM)
OHL 400 kV Nadab (ROM)
OHL 400 kV
Rosiori (ROM)
OHL 400 kV
Portile De Fier (ROM)
OHL 400 kV
Subotica (SER)
OHL 400 kV
Ribarevine (MON)
OHL 220 kV
Pljevlja (MON)
OHL 220 kV
Pljevlja (MON)
OHL 400 kV*
Skopje 4 (MCD)
OHL 400 kV*
Zemlak (ALB)
OHL 400 kV*
V.Dejes (ALB)
* new lines planned till 2015
Kardia (GRE)
Prizren (SER)
Podgorica (MON)
Podgorica (MON)
Ernestinovo (CRO)
Konjsko (CRO)
S. Mitrovica (SER)
Podgorica (MON)
Plat (CRO)
Mraclin (CRO)
Medjuric (CRO)
Djakovo (CRO)
Djakovo (CRO)
Zakucac (CRO)
Vardiste (SER)
Piva (MON)
Perucica (MON)
Thessaloniki (GRE)
Filippi (GRE)
Babaeski (TUR)
Hamitabat (TUR)
Stip (MCD)
Isaccea (ROM)
Tantarena (ROM)
Tantarena (ROM)
Nis (SER)
Heviz (HUN)
Heviz (HUN)
Pecs (HUN)
Pecs (HUN)
Krsko (SLO)
Krsko (SLO)
Divaca (SLO)
S.Mitrovica (SER)
Cirkovce (SLO)
Divaca (SLO)
Thessaloniki (GRE)
Florina (GRE)
Kosovo B (UNMIK)
Kosovo A (UNMIK)
Kosovo A (UNMIK)
Sandorfalva (HUN)
Bekescaba (HUN)
Mukacevo (UKR)
Djerdap (SER)
Sandorfalva (HUN)
Kosovo B (UNMIK)
Bajina Basta (SER)
Pozega (SER)
Vranje (SER)
Bitola (MCD)
Kosovo B (UNMIK)
Power Flow
MW
Mvar
-58.6
-110.9
17.1
29.0
-9.7
-15.2
33.3
-23.9
-13.8
-63.5
228.5
-30.6
-203.8
-2.8
-0.6
43.1
-104.8
40.0
16.0
-20.2
5.9
3.2
52.0
22.8
100.8
45.3
113.4
-10.9
-28.8
38.7
-156.5
-34.2
42.0
11.2
50.5
-53.9
290.6
-48.8
-127.3
-1.6
-158.4
1.9
197.4
-36.0
396.1
3.9
28.5
27.1
28.5
27.1
304.5
62.6
-77.7
-86.8
-77.7
-86.8
-25.5
-84.1
-25.5
-84.1
-264.6
11.5
-264.6
11.5
-106.6
22.8
-148.4
25.3
-74.8
6.5
-37.4
4.6
-61.0
-50.9
-183.9
-101.9
-216.8
-47.0
-11.1
-32.8
-11.1
-32.8
-18.6
-54.7
6.1
-77.6
-503.9
-68.9
-2.1
-32.3
-32.8
-147.1
-148.4
-17.0
-11.3
9.6
75.9
40.3
61.9
14.2
-279.2
-75.7
-289.3
-68.7
% of thermal
rating
9
14
6
3
5
17
16
7
36
8
4
20
37
36
16
41
15
10
41
12
10
28
29
6
6
44
9
9
7
7
23
23
11
12
25
12
5
17
16
11
11
5
6
43
2
11
11
9
29
6
21
22
5.23
45
40
Frequency
35
30
25
20
15
10
5
0
x<25
25<x<50
50<x<75
75<x<100
x>100
Bin
Figure 5.4.3 - Histogram of interconnection lines loadings for 2015-sensitivity case-average hydrology-import 1500
MW-topology 2015 scenario ("Frequency" denotes number of lines and "Bin" denotes loading range in % of thermal
limit)
Table 5.4.3 - Network elements loaded over 80% of thermal limits for 2015-sensitivity case-average hydrology-import
1500 MW-topology 2015 scenario
AREA
ALB
BUL
ROM
SRB
ALB
B&H
ROM
SRB
ELEMENT
Lines
OHL 220 kV AKASHA2-ARRAZH2
OHL 220 kV MI_2_220-ST ZAGORA
OHL 400 kV PLOVDIV4-MI_400
OHL 220 kV BRADU-TIRGOVI
OHL 220 kV P.D.F.A-CETATE1
OHL 220 kV BUC.S-B-FUNDENI
OHL 220 kV JBGD3 21-JOBREN2
Transformers
TR 220/110 kV AFIERZ2-AFIERZ5 ckt.1
TR 220/110 kV AFIERZ2-AFIERZ5 ckt.2
TR 220/110 kV ABURRE2-ABURRL5 ckt. 1
TR 220/110 kV ABURRE2-ABURRL5 ckt. 2
TR 220/110 kV ABURRE2-ABURRL5 ckt. 3
TR 220/110 kV AELBS12-AELBS15 ckt.1
TR 220/110 kV AELBS12-AELBS15 ckt.2
TR 220/110 kV AELBS12-AELBS15 ckt.3
TR 220/110 kV AKASHA2-AKASH25 ckt.1
TR 220/110 kV AKASHA2-AKASH25 ckt.2
TR 220/110 kV AFIER 2-AFIER 5 ckt.1
TR 220/110 kV AFIER 2-AFIER 5 ckt.2
TR 220/110 kV AFIER 2-AFIER 5 ckt.3
TR 400/110 kV UGLJEVIK
TR 400/110 kV BRASOV
TR 220/110 kV FUNDENI
TR 220/110 kV FUNDENI-FUNDE2B
TR 220/110 kV JBGD3 21-JBGD 351
TR 220/110 kV JBGD3 22-JBGD 352
TR 220/110 kV JZREN22-JZREN25
LOADING
MVA
RATING
MVA
PERCENT
235.7
191.0
605.2
246.9
204.3
266.0
277.4
270.0
228.6
692.8
302.6
208.1
320.0
301.0
87.3
83.5
87.4
81.6
98.2
83.1
92.2
50.9
50.9
49.1
49.1
49.1
74.0
74.0
79.5
82.9
82.9
130.7
107.1
102.2
256.3
202.5
181.0
214.4
169.4
126.8
121.5
60.0
60.0
40.0
40.0
40.0
90.0
90.0
90.0
100.0
100.0
120.0
90.0
90.0
300.0
250.0
200.0
200.0
200.0
150.0
150.0
84.8
84.8
81.8
81.8
81.8
82.2
82.2
88.3
82.9
82.9
108.9
119.0
113.6
85.4
81.0
90.5
107.2
84.7
84.5
81.0
5.24
350
300
Frequency
250
200
150
100
50
0
x<25
25<x<50
50<x<75
75<x<100
x>100
Bin
Figure 5.4.4 - Histogram of 400 kV and 220 kV regional branches loadings for 2015-sensitivity case-average
hydrology-import 1500 MW-topology 2015 scenario ("Frequency" denotes number of lines and "Bin" denotes loading
range in % of thermal limit)
5.4.2 Voltage Profile in the Region
x>1.1
1.05<x<1.1
1.025<x<1.05
1<x<1.025
0.975<x<1
0.95<x<0.975
0.9<x<0.95
100
90
80
70
60
50
40
30
20
10
0
x<0.9
Frequency
Voltage profile in the region within this scenario which is defined by given generation pattern and
power import is seen as satisfactory because there are no voltages outside permitted limits. Figure
5.4.5 shows histogram of voltages in monitored 400 kV and 220 kV substations.
Bin
Figure 5.4.5 - Histogram of voltages in monitored substations for 2015-sensitivity case-average hydrology-import
1500 MW-topology 2015 scenario ("Frequency" denotes number of busses and "Bin" denotes voltage range in p.u.)
5.25
5.4.3 Security (n-1) analysis
Results of security (n-1) analysis for the 2015-sensitivity case-average hydrology-import 1500
MW-topology 2015 scenario are presented in Tables 5.3.4 - 5.3.5.
Insecure system situations for given generation pattern and power import are detected in the power
systems of Albania, Bulgaria, Bosnia and Herzegovina, Croatia, Romania and Serbia. Some
possible actions for transmission system relief during certain contingency cases are previously
described while the transformer overloadings in Zerjavinec (Croatia), Nis and Pancevo (Serbia)
substations may be removed by system operator actions.
Figure 5.4.6 shows geographical positions of critical elements in the analyzed scenario. A green
colour reveals 220 kV elements (line 220 kV or transformer 220/x kV), while a red one reveals 400
kV elements (line 400 kV or transformer 400/x kV).
According to the obtained and presented results, it may be concluded that certain reinforcements in
the internal networks of Romania, Bulgaria, Albania and Serbia are necessary shall this generation
pattern and 1500 MW of power import be made more secure. None of the identified congestions is
located at the border lines.
Table 5.4.4 - Lines overloadings for 2015-sensitivity case-average hydrology-import 1500 MW scenario-topology 2015,
single outages
Outage
Overloaded line(s)
OHL 220 kV AELBS12-AFIER 2
OHL 400 kV BURGAS-MI_2_400
OHL 220 kV G_ORIAH-MI_2_220
OHL 400 kV VISEGRA-HE VG
OHL 400 kV TANTAREN-BRADU
OHL 400 kV DOMNESTI-BRAZI
OHL 220 kV P.D.F.A-CALAFAT
OHL 220 kV P.D.F.A-RESITA ckt.1
OHL 220 kV FUNDENI-BUC.S-B
TR 400/220 kV ROSIORI
OHL 220 kV AKASHA2-ARRAZH2
OHL 400 kV PLOVDIV4-MI_400
OHL 220 kV MI_2_220-ST.ZAGORA
OHL 220 kV RP KAKAN-KAKANJ5
OHL 220 kV BUC.S-B-FUNDENI
OHL 220 kV BUC.S-B-FUNDENI
OHL 220 kV P.D.F.A-CETATE1
OHL 220 kV P.D.F.A-RESITA ckt.2
OHL 220 kV BUC.S-B-FUNDENI
OHL 400 kV GADALIN-CLUJ E
OHL 220 kV BRADU-TIRGOVI
OHL 220 kV BUC.S-B-FUNDENI
OHL 220 kV JBGD3 21-JOBREN2
OHL 220 kV JBGD3 21-JOBREN2
OHL 220 kV JBGD3 22-JBGD8 22
OHL 220 kV JBGD3 21-JOBREN2
OHL 220 kV JBGD3 21-JOBREN2
OHL 220 kV JBGD3 21-JOBREN2
OHL 220 kV JBGD3 21-JOBREN2
OHL 220 kV JBGD3 21-JOBREN2
OHL 220 kV JBGD172-JBGD8 22 ckt.2
TR 400/220 kV BRAZI
TR 400/220 kV JBGD8 ckt.2
TR 400/220 kV JBGD8 ckt.1
OHL 400 kV JBGD8 1-JOBREN11
OHL 400 kV JHDJE11-JTDRMN1
OHL 400 kV JPANC21-JTDRMN1
OHL 220 kV JBBAST2-JBGD3 21
OHL 220 kV JBGD3 22-JBGD8 21
OHL 220 kV JBGD172-JBGD8 22 ckt.1
Loadings
MVA
%
351.1 136.4
742.9 104.9
255.1 105.0
335.1 102.3
312.8 103.4
319.8 103.1
257.9 125.0
273.3 100.9
342.7 110.2
242.2 105.7
285.9 105.9
355.6 118.4
307.4 104.7
329.2 112.9
373.7 107.3
468.7 167.2
321.3 109.1
302.0 103.1
304.8 103.6
300.2 102.0
464.7 132.6
Country
ALBANIA
BULGARIA
B&H
ROMANIA
SERBIA
5.26
Table 5.4.5 - Transformer overloadings for 2015-sensitivity case-average hydrology-import 1500 MW scenariotopology 2015, single outages
Outage
Overloaded branch(es)
OHL 400 kV BLUKA 6-TS TUZL
TR 400/110 kV ZERJAVIN ckt.1
OHL 400 kV BRASOV-DIRSTE
OHL 220 kV BUC.S-B-FUNDENI
TR 400/220 kV BUC.S ckt.1
TR 400/110 kV BRASOV
TR 400/110 kV DIRSTE
TR 400/110 kV NIS ckt.1
TR 400/110 kV PANCEVO ckt.1
TR 400/110 kV UGLJEVIK
TR 400/110 kV ZERJAVIN ckt.2
TR 400/110 kV BRASOV
TR 400/220 kV BRAZI
TR 400/220 kV BUC.S ckt.2
TR 400/110 kV DIRSTE
TR 400/110 kV BRASOV
TR 400/110 kV NIS ckt.2
TR 400/110 kV PANCEVO ckt.2
Loadings
MVA
%
302.7 100.9
300.8 100.3
253.8 101.5
418.6 104.6
505.1 126.3
434.5 173.8
423.8 169.5
304.8 101.6
305.8 101.9
Country
B&H
CROATIA
ROMANIA
SERBIA
SVK
AUT
HUN
SLO
ROM
CRO
SRB
BIH
MNG
BUL
MKD
TUR
ALB
GRE
critical elements
400 kV line or 400/x kV transformer
220 kV line or 220/x kV transformer
Figure 5.4.6 – Geographical positions of the critical elements for 2015-sensitivity case-average hydrology-import
1500 MW-topology 2015 scenario
5.4.4 Summary of Impacts - 2015 topology versus 2010 topology
Compared to the expected topology 2010, analyzed in previous chapter, it can be seen that planned
investments make analyzed generation, demand and power import scenario feasible. This scenario
could not be solved on 2010 topology because lack of reactive power support especially in the
power system of Albania. Construction of at least one new interconnection line from Albania
(V.Dejes or Zemlak) to UNMIK (Kosovo B) or Macedonia (Bitola) makes this scenario feasible but
voltage profile in Albania, Montenegro and Serbia is poor. If all planned new lines are in operation
voltage profile is satisfactory but some internal reinforcements will be necessary to make this
generation pattern and power import more secure.
5.27
5.5 Scenario 2015 – average hydrology – high load – 2010 topology
This part of the Study presents the results of static load flow and voltage profile analyses that are
conducted for complete network topology and (n-1) contingencies in the scenario which is denoted
as GTmax run for year 2015 - average hydrology high load and expected network topology for
2010 with new generation facilities implemented.
5.5.1 Lines loadings
Figure 5.5.1 shows power exchanges between areas for 2015-average hydrology high load scenario
and 2010 topology. Power flows along interconnection lines in the region together with balances of
the systems are shown in Figure 5.5.2. Area totals are shown in Table 5.5.1 and comparison of the
average hydro regimes with normal load projection and high load projection is shown in Table
5.5.2. Difference of load-demand level for these two regimes for 2015 is around 8.11% on regional
level. As a consequence of this high load level, network losses are increased by 262 MW or 25%.
Further differences are explained in detail in following chapter.
UKR
450
SVK
52
8
39
1
16
AUT
HUN
62
60
12
3
180
SLO
27
ITA
ROM
5
CRO
318
97
8
43
883
SRB
249
BIH
61
5
5
31
109
MNG
BUL
59
849
7
28
7
20
169
ALB
30
63
MKD
59
TUR
38
9
GRE
Figure 5.5.1 - Area exchanges in analyzed electric power systems for 2015-average hydrology high load scenario –
topology 2010
Figure 5.5.3 shows histogram of tie lines loadings. It is concluded that most of the tie lines are
loaded less than 50% of their thermal limits, but there are some that are loaded up to 75% of their
thermal limit like the 220 kV line from Pizren (Serbia-UNMIK)-Fierza (Albania).
5.28
LEVIC 1400.0
CENTREL
200
124
251
106
GABICK1400.0
450.0 MW
MSAJO 4
410.0
224.4
UMUKACH
412.0
53.
276 3
221.3
22.5
5.2
49.4
32.1
222.0MEDURIC
220.6
58.1 ARAD
130
381.3
ROM
43.1
51.8
P.D.FIE
TANTAREN
395.6
228
113
43.1
50.8
187
7.8
402.8
UGLJEVIK 396.1
SCG
JVARDI22
61.7 226.4
64.7
61.4
63.3
SA 20
225.4
CRG
587.0 MW
AEC_400
JNIS2 1
390.4
109
168
DOBRUD4 408.8
409.5
109 SOFIA_W4
109
406.0
JHPERU21
30.9 225.3
41.6
30.7
34.4
315
JPODG211
125
JPODG121223.8
2
1. 05
8
76.7
31.4
209
148
SK 1 223.2
SK 4
408.2
BITOLA 2
413.9
408.8
QES/H
76.7
0.0
KARDIA K
393 415.1
217
K
411.9
FILIPPOI
415.4
2
12 0.3
7
59.1
23.6
K-KEXRU
417.7
KV: 110 , 220 , 400
4HAMITAX
26.2
26.2 414.9
64.8
59.0 53.5
4BABAESK
72.2
415.2
0.0 MW
-0.1 MW
KARACQOU
250 418.2
50.0
3
26 2 . 9
.3
.9
32 6.1
7
.6
92 .4
27
LAGAD K 412.9
0
92. 1
67.
AZEMLA1
386
232
MI_3_4_1 417.4
BLAGOEV409.6
409.7
AHS_FLWR415.9
385.6
408.5
.2 5
26 6.
2
1
92. 7
88.
ALB
-883.0 MW
C_MOGILA
.2
63 .4
47
STIP 1
DUBROVO
366.3
.1
63 .5
44
407.8
76.8
60.3
209
148
717
113
3
69. 8
26.
3
69. 8
26.
AVDEJA2
218.1AFIERZ2218.4
378.4
MKD
208
174
899.0 MW
JVRAN31
0.000
-617.0 MW
AKASHA1
BUL
JLESK21
0.000
JPRIZ22
224.7 JTKOSA2
230.2JTKOSB1419.8
15 210
.8
387.2
709
71.3
313
103
68.6
34.0
RP TREB
229.2
JHPIVA21
264 234.3
45.3
2639.2 MW
SRB
2052.2 MW
68.6
34.0
VISEGRA
228.4
AVDEJA1
GIS REGIONAL MODEL - AVERAGE HYDRO
2015 HIGH LOAD - NEW GENERATION - TOPOLOGY 2010
421
123
-1039.9 MW
JSUBO31 384.7
JHDJE11
229.7
RP TREB
393.6
SHAW POWER
TECHNOLOGIES
INC. R
ISACCEA 387.7
395.6
77.5
26.3
223.6
TE TUZL
391.4
58.4
77.6
319
JSMIT21
119
210
131
396.9
191 MO-4
4.1
230.0
BIH
113.2 MW
ROSIORI 382.8
9
22 3
.
73
9
22 3
.
73
MO-4
3.9
NADAB
170
384.5
180
177
JSOMB31
381.7
317
88.6
401.6
157
103
185
11.0
494
5.6
GRADACAC
219.6
2
10 8
.
45
489
2.9
8
52. 2
23.
.8
49 .1
39
PRIJED2
4.2
155
12
57. 2
1
DAKOVO
22.5
13.0
CRO
-1447.8 MW
393.0
24.2
46.0
MSAFA 4
395.8
HE ZAKUC
224.1
24.1
73.1
UKR
450.0 MW
92.3
64.1
MRACLIN
KONJSKO
10.7
28.9
0
18 9
15
41 188
.6
399.2
156
52.8
ERNESTIN
100
47.7
9.9
15.0
1
16 4
.
12
1
16 4
.
12
TUMBRI
406.7
108
1
10 08
10 8
8
.0
.0
46 .8 46 .8
89
89
ZERJAVIN
401.3
ZERJAVIN 226.1
399.9
PEHLIN 228.0
226.6
1
62 08
.6
1
62 08
.6
LCIRKO2
399.7
8
7. .4
10
MELINA
35.7
UMUKACH2
26.9
10.8
7.3
MBEKO 4
392.1
407.1
MPECS 4
LKRSKO1
9.
4. 9
9
35.5
20.1
46.1
15.9
SLO
82
400.1 7. .7
8
LDIVAC2
228.8
46.1
15.9
215.0 MW
52.6
27.7
9.4
86.8
MHEVI 4
7.8
0.6
9.3
83.4
21.2 228.8
27.3
LMARIB1 401.7
82.8
15.8
230.0
21.3
13.5
MTLOK 2
230.4
HUN
161
12.1
IPDRV121
175
49.8
50.1
21.6
MKISV 2
226.2
-1249.8 MW
161
12.1
400.7
175
29.1
MGYOR 2
50.6
6.6
44.
6.7 9
44.
6.7 9
153
48.
6
IRDPV111
406.4
9
50. 5
31
LPODLO2228.5
MSAJI 408.1
4
427
61.9
44.9
37.8
44.9
37.8
234.1
ONEUSI2
233.4
LDIVAC1
MGOD
OWIEN 2
350
UZUKRA01
282
772.5
26.2
53.5
OKAINA1 401.0
156
52.3
OOBERS2238.7
346
813
228
113
89.3
21.5
405.2
199
77.9
UCTE
221.2 MW
MAISA 7
712.0
MGYOR 4
0
25 .7
89.2
83
104
403.1
OWIEN 1
GRE
TUR
BUS -VOLTAGE(KV)
BRANCH -MW/MVAR
EQUIPMENT -MW/MVAR
Figure 5.5.2 - Power flows along interconnection lines in the region with balances of the systems for 2015-average hydrology high load scenario – 2010 topology
5.29
Table 5.5.1 - Area totals in analyzed electric power systems for 2015-average hydrology high load scenario –
2010 topology
AREA
GENERATION
LOAD
LOSSES
INTERCHANGE
ALBANIA
896.1
1680
99
-883
BULGARIA
8157.8
7083
175.8
899
BIH
2479.4
2274
92.2
113.2
CROATIA
2591
3959
79.8
-1447.8
MACEDONIA
895
1489
23
-617
ROMANIA
7684.5
8269.8
454.5
-1039.8
SERBIA
10024.2
7635
337
2052.2
MONTENEGRO
1324.6
702
35.6
587
TOTALS
34052.6
33091.8
1296.9
-336.2
Table 5.5.2 - Comparison of Area totals in analyzed electric power systems for 2015- average hydrology versus average
hydrology high load scenario – 2010 topology
LOAD
LOSSES
AREA
normal
normal
high load
high load
load
load
ALBANIA
1531
1680
9.73%
81.2
99
21.92%
BULGARIA
6483
7083
9.25%
150.7
175.8
16.66%
BIH
2279
2274
-0.22%
78.7
92.2
17.15%
CROATIA
3657
3959
8.26%
63.4
79.8
25.87%
MACEDONIA
1407
1489
5.83%
20
23
15%
ROMANIA
7317.4
8269.8
13.02%
347.4
454.5
30.83%
SERBIA
7263
7635
5.12%
268.7
337
25.42%
MONTENEGRO
671
702
4.62%
25
35.6
42.4%
TOTALS
30608.4
33091.8
8.11%
1035.1
1296.9
25.29%
Histogram
35
30
Frequency
25
20
15
10
5
0
x<25
25<x<50
50<x<75 75<x<100
x>100
Bin
Figure 5.5.3 - Histogram of interconnection lines loadings for 2015- average hydrology high load scenario – topology
2010 ("Frequency" denotes number of lines and "Bin" denotes loading range in % of thermal limit)
Following Table 5.5.3 lists all network elements loaded over 80% of their thermal limits. As it can
be seen large number of lines 220 kV voltage level in Albania, Romania and Serbia are loaded over
80%. It can be concluded that overall load level of the transmission network increased, especially in
parts of the network that supply major consumption areas (Tirana in Albania, Bucharest and
Timisoara in Romania, Belgrade in Serbia and Pristina in Serbia-UNMIK). Most of the elements
loaded over 80% are 220 kV lines in Romania, but also transformers in some substations that
supply the major consumption areas.
5.30
There are some elements that are overloaded (220 kV lines Targu Jiu – Paroseni and Urechesti –
Targu Jiu in Romania and 220/110 kV transformers in Fierza and Elbasan 1 substation in Albania,
and 400/220 kV and 220/110 kV transformers in Kosovo B and Kosovo A substations in SerbiaUNMIK. This leads to conclusion that transmission network is not able to sustain this load-demand
level and this production pattern.
Table 5.5.3 - Network elements loaded over 80% of their thermal limits for 2015-average hydrology high load scenario
– 2010 topology
BRANCH LOADINGS ABOVE
AREA
ALB
ROM
SRB
ALB
BIH
CRO
ROM
SRB
80.0 % OF RATING:
LOADING
MVA
Lines
HL 220kV AKASHA2-ARRAZH2 1
287.7
HL 220kV BRADU-TIRGOVI 1
285.4
HL 220kV BUC.S-B-FUNDENI 1
309.4
HL 220kV FILESTI-BARBOSI 1
248.3
HL 220kV L.SARAT-FILESTI 1
241
HL 220kV P.D.F.A-CETATE1 1
207.2
HL 220kV P.D.F.A-RESITA 1
266.9
HL 220kV P.D.F.A-RESITA 2
266.9
HL 220kV P.D.F.II-CETATE1 1
271.1
HL 220kV PAROSEN-BARU M 1
227.6
HL 220kV RESITA-TIMIS 1
229.6
HL 220kV RESITA-TIMIS 2
229.6
HL 220kV TG.JIU-PAROSEN 1
320.7
HL 220kV URECHESI-TG.JIU 1
320.4
HL 400kV GADALIN-CLUJ E 1
208.4
HL 220kV JBGD3 21-JOBREN2 1
291
Transformers
TR 220/110 kV ABURRE 1
49.5
TR 220/110 kV ABURRE 2
49.5
TR 220/110 kV ABURRE 3
49.5
TR 220/110 kV AELBS1 1
87.7
TR 220/110 kV AELBS1 2
87.7
TR 220/110 kV AELBS1 3
94.2
TR 220/110 kV AFIER 1
149
TR 220/110 kV AFIER 2
122.1
TR 220/110 kV AFIER 3
116.5
TR 220/110 kV AFIERZ 1
58
TR 220/110 kV AFIERZ 2
58
TR 220/110 kV AKASHA 1
93.4
TR 220/110 kV AKASHA 2
93.4
TR 220/110 kV ARRAZH 1
87.7
TR 220/110 kV ARRAZH 2
87.7
TR 220/110 kV ATIRAN 2
97.5
TR 220/110 kV ATIRAN 3
102.2
TR 400/110 kV UGLJEV 1
258.8
TR 220/110 kV TESISA 1
178.8
TR 220/110 kV BARBOS 1
167.1
TR 220/110 kV FUNDE2 1
240.8
TR 220/110 kV FUNDEN 1
203.9
TR 220/110 kV TIMIS 1
171.7
TR 400/110 kV BRASOV 1
231.5
TR 400/110 kV CLUJ E 1
208.4
TR 400/110 kV DIRSTE 1
221
TR 400/220 kV BUC.S 1
378.4
TR 400/220 kV BUC.S 2
378.4
TR 400/220 kV IERNUT 1
408.1
TR 400/220 kV URECHE 1
486.6
TR 220/110 kV JBGD3 1
174.9
TR 220/110 kV JBGD3 2
137.1
TR 220/110 kV JPRIS4 1
137
TR 220/110 kV JPRIS4 2
137
TR 220/110 kV JTKOSA 2
151.3
TR 220/110 kV JTKOSA 3
154
TR 220/110 kV JZREN2 2
133.8
TR 400/110 kV JJAGO4 A
245.8
TR 400/220 kV JBGD8 1
332.2
TR 400/220 kV JTKOSB 1
411.6
ELEMENT
RATING
MVA
PERCENT
270
302.6
320
277.4
277.4
208.1
277.4
277.4
277.4
277.4
277.4
277.4
208.1
277.4
238.3
301
106.6
94.3
96.7
89.5
86.9
99.6
96.2
96.2
97.7
82
82.8
82.8
154.1
115.5
87.5
96.7
60
60
60
90
90
90
120
90
90
60
60
100
100
100
100
120
120
300
200
200
200
200
200
250
250
250
400
400
400
400
200
150
150
150
150
150
150
300
400
400
82.5
82.5
82.5
97.5
97.5
104.7
124.1
135.7
129.5
96.7
96.7
93.4
93.4
87.7
87.7
81.3
85.2
86.3
89.4
83.6
120.4
101.9
85.9
92.6
83.4
88.4
94.6
94.6
102
121.6
87.5
91.4
91.3
91.3
100.9
102.7
89.2
81.9
83
102.9
5.31
Histogram
300
Frequency
250
200
150
100
50
0
x<25
25<x<50 50<x<75 75<x<100
x>100
Bin
Figure 5.5.4 - Histogram of branch loadings for 2015-average hydrology high load scenario – 2010 topology
("Frequency" denotes number of lines and "Bin" denotes loading range in % of thermal limit)
5.5.2 Voltage Profile in the Region
Figure 5.5.5 shows histogram of voltages in monitored substations. Overall voltage profile is not
adequate. In most of the network, especially in major consumption areas of Albania and Romania,
voltages in some monitored substations are below limits. Using of voltage control devices (tap
changing transformers, shunt devices) in Albania and Romania is not analyzed, but especially
Albania, these devices have very limited abilities for improving voltage profile. All in all it can be
concluded that conditions in some parts of the transmission network in this regime are dangerously
close to voltage collapse.
Histogram
80
70
Frequency
60
50
40
30
20
10
x>1.1
1.05<x<1.1
1.025<x<1.05
1<x<1.025
0.975<x<1
0.95<x<0.975
0.9<x<0.95
x<0.9
0
Bin
Figure 5.5.5 - Histogram of voltages in monitored substations for 2015-dry hydrology scenario – 2010 topology
("Frequency" denotes number of busses and "Bin" denotes voltage range in p.u.)
5.32
5.5.3 Security (n-1) analysis
Results of security (n-1) analysis for 2015-average hydrology high load scenario and expected
topology for 2010 are presented in Table 5.5.4. Figure 5.5.6 shows the geographical position of the
critical elements in monitored systems. As it can be seen, by lot of investigated contingencies,
insecure states are identified. Almost all elements that are loaded over 80% in base case (Table
5.5.3), represent critical elements in n-1 analyses. Furthermore, for lot of contingency cases (losing
of major interconnection and internal lines) results could not be presented because of mathematical
instability of the model. This indicates that all these cases are not feasible in reality, and that these
contingencies can lead to voltage collapse and partial black outs in some parts of the monitored
network.
Table 5.5.4 - Network overloadings for 2015-dry hydrology scenario , single outages – 2010 topology
Area
1
contingency
2
BASE CASE
AL
AL
IN
IN
RO
RO
OHL
OHL
OHL
OHL
OHL
OHL
220kV
220kV
400kV
400kV
220kV
220kV
AVDEJA2 -ATIRAN2
ATIRAN2 -AKASHA2
KONJSKO -MO-4
ISACCEA -DOBRUD4
STUPARE -BRADU
URECHESI-SARDANE
RO
OHL 220kV URECHESI-TG.JIU
1
RO
OHL 220kV P.D.F.A -CALAFAT
1
RO
OHL 220kV P.D.F.A -RESITA
1
RO
OHL 220kV RESITA
1
RO
RO
OHL 220kV CRAIOV A-TR. MAG 1
OHL 220kV CRAIOV B-ISALNI A 1
RO
OHL 220kV TG.JIU
-PAROSEN
RO
OHL 220kV PESTIS
-MINTIA A 1
RO
OHL 220kV MINTIA B-AL.JL
1
RO
OHL 220kV FUNDENI -BUC.S-B
1
RO
OHL 220kV STEJARU -GHEORGH
1
RO
RO
OHL 220kV GHEORGH -FINTINE
OHL 220kV CLUJ FL -MARISEL
1
1
RO
OHL 220kV CLUJ FL -AL.JL
1
RO
OHL 220kV AL.JL
1
RO
RO
RO
OHL 220kV BRAZI A-TELEAJEN 1
OHL 220kV TELEAJEN-STILPU
1
OHL 400kV TANTAREN-URECHESI 1
RO
OHL 400kV TANTAREN-SLATINA
-TIMIS
-GILCEAG
1
1
1
1
1
1
1
1
Area
3
AL
RO
RO
RO
RO
RO
CS
CS
CS
AL
AL
HR
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
overloadings
/
out of limits voltages
4
HL 220kV AKASHA2-ARRAZH2
TR 400/220kV/kV URECHESI
HL 220kV LOTRU-SIBIU
HL 220kV LOTRU-SIBIU
HL 220kV URECHESI-TG.JIU
HL 220kV TG.JIU-PAROSEN
TR 400/220kV/kV JTKOSB1
TR 400/220kV/kV JTKOSB1
TR 400/220kV/kV JTKOSB1
HL 220kV AKASHA2-ARRAZH2
HL 220kV AKASHA2-ARRAZH2
HL 220kV XRA_ZA21-HE ZAKUC
TR 400/220kV/kV URECHESI
HL 220kV AREF-RIURENI
TR 400/220kV/kV URECHESI
HL 400kV MINTIA-SIBIU
HL 220kV P.D.F.A-RESITA
HL 220kV P.D.F.A-RESITA
HL 220kV PESTIS-MINTIA A
HL 220kV P.D.F.A-CETATE1
TR 400/220kV/kV URECHESI
HL 220kV URECHESI-TG.JIU
HL 220kV P.D.F.A-RESITA
HL 220kV TG.JIU-PAROSEN
TR 400/220kV/kV URECHESI
HL 220kV URECHESI-TG.JIU
HL 220kV RESITA-TIMIS
TR 400/220kV/kV URECHESI
TR 400/220kV/kV URECHESI
HL 400kV MINTIA-SIBIU
HL 220kV P.D.F.A-RESITA
HL 220kV P.D.F.A-RESITA
HL 220kV PESTIS-MINTIA A
TR 400/220kV/kV URECHESI
HL 220kV URECHESI-TG.JIU
HL 220kV TG.JIU-PAROSEN
HL 400kV GADALIN-CLUJ E
HL 220kV URECHESI-TG.JIU
HL 220kV TG.JIU-PAROSEN
HL 220kV BRADU-TIRGOVI
HL 220kV BUC.S-B-FUNDENI
TR 400/220kV/kV URECHESI
TR 400/220kV/kV IERNUT
TR 400/220kV/kV IERNUT
TR 400/220kV/kV URECHESI
TR 400/220kV/kV URECHESI
HL 220kV URECHESI-TG.JIU
HL 220kV TG.JIU-PAROSEN
TR 400/220kV/kV URECHESI
HL 220kV URECHESI-TG.JIU
HL 220kV TG.JIU-PAROSEN
HL 220kV BUC.S-B-FUNDENI
HL 220kV BUC.S-B-FUNDENI
TR 400/220kV/kV URECHESI
TR 400/220kV/kV URECHESI
HL 220kV BRADU-TIRGOVI
HL 220kV URECHESI-TG.JIU
HL 220kV TG.JIU-PAROSEN
#
5
1
1
1
2
1
1
1
2
3
1
1
1
1
1
1
1
1
2
1
1
1
1
2
1
1
1
2
1
1
1
1
2
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
limit
/
Unom
6
270MVA
400MVA
277.4MVA
277.4MVA
277.4MVA
208.1MVA
400MVA
400MVA
400MVA
270MVA
270MVA
297MVA
400MVA
277.4MVA
400MVA
381.1MVA
277.4MVA
277.4MVA
277.4MVA
208.1MVA
400MVA
277.4MVA
277.4MVA
208.1MVA
400MVA
277.4MVA
277.4MVA
400MVA
400MVA
381.1MVA
277.4MVA
277.4MVA
277.4MVA
400MVA
277.4MVA
208.1MVA
238.3MVA
277.4MVA
208.1MVA
302.6MVA
320MVA
400MVA
400MVA
400MVA
400MVA
400MVA
277.4MVA
208.1MVA
400MVA
277.4MVA
208.1MVA
320MVA
320MVA
400MVA
400MVA
302.6MVA
277.4MVA
208.1MVA
Flow
/
Voltage
7
263MVA
483.6MVA
275.6MVA
275.6MVA
307.7MVA
298.9MVA
432MVA
451.4MVA
451.4MVA
275.2MVA
274.6MVA
343.9MVA
506.7MVA
252.6MVA
455.2MVA
387.2MVA
269.1MVA
269.1MVA
256.1MVA
259.6MVA
504MVA
325.7MVA
355.4MVA
315.2MVA
494.2MVA
317.8MVA
351.8MVA
464.8MVA
499.6MVA
386.7MVA
269.1MVA
269.1MVA
257.2MVA
507.3MVA
326MVA
313.5MVA
216.3MVA
296.3MVA
288.2MVA
273.6MVA
389.3MVA
495.9MVA
487.2MVA
412.5MVA
497MVA
471.7MVA
294.1MVA
286.3MVA
514.1MVA
333MVA
320.8MVA
307.4MVA
307.8MVA
467.2MVA
559.7MVA
272.9MVA
292.5MVA
284.5MVA
rate
/
volt.dev.
8
106.6%
120.9%
102.1%
102.1%
115.5%
154.1%
108.0%
112.9%
112.9%
123.9%
111.1%
115.1%
126.7%
106.6%
113.8%
116.9%
108.8%
108.8%
105.5%
129.9%
126.0%
123.6%
145.4%
165.0%
123.6%
119.9%
136.8%
116.2%
124.9%
117.0%
109.0%
109.0%
106.3%
126.8%
124.6%
166.4%
101.8%
111.1%
148.3%
108.2%
134.8%
124.0%
121.8%
103.1%
124.3%
117.9%
110.1%
146.9%
128.5%
127.8%
170.6%
105.4%
105.6%
116.8%
139.9%
104.1%
112.9%
150.6%
5.33
Area
1
contingency
2
RO
OHL 400kV URECHESI-P.D.FIE
RO
OHL 400kV URECHESI-DOMNESTI 1
RO
OHL 400kV MINTIA
-ARAD
1
RO
OHL 400kV MINTIA
-SIBIU
1
RO
RO
OHL 400kV P.D.FIE -SLATINA
OHL 400kV DOMNESTI-BUC.S
1
1
RO
OHL 400kV DOMNESTI-BRAZI
1
RO
RO
RO
RO
OHL
OHL
OHL
OHL
-L.SARAT
-CERNAV
-CERNAV
-ROMAN
1
1
2
1
RO
OHL 400kV BRASOV
-BRADU
1
RO
OHL 400kV IERNUT
-GADALIN
1
CS
CS
CS
CS
CS
OHL
OHL
OHL
OHL
OHL
CS
OHL 220kV JLESK22 -JNIS 22
1
CS
CS
CS
CS
OHL
OHL
OHL
OHL
1
1
1
1
CS
OHL 400kV JBGD8 1 -JOBREN11 1
CS
CS
OHL 400kV JBGD8 1 -JBGD201
OHL 400kV JBOR 21 -JHDJE11
A
1
CS
OHL 400kV JNIS2 1 -JJAGO41
A
CS
OHL 400kV JNSAD31 -JSUBO31
1
CS
OHL 400kV JOBREN12-JTKOLB1
A
CS
OHL 400kV JPANC21 -JTDRMN1
1
CS
OHL 400kV JRPMLA1 -JSMIT21
1
BG
CS
CS
CS
CS
TR
TR
TR
TR
TR
CS
TR 400/110 JPEC
BG
TR 400/110 PLOVDIV 1
MK
TR 400/110 SK 1
RO
RO
TR 400/110 SMIRDAN
TR 400/220 BRAZI
400kV
400kV
400kV
400kV
220kV
220kV
220kV
220kV
220kV
220kV
220kV
220kV
220kV
GR.IAL
GR.IAL
GR.IAL
BACAU
JBBAST2 -JBGD3 21
JBGD172 -JBGD8 22
JBGD3 22-JBGD8 21
JBGD3 22-JBGD8 22
JHIP 2 -JPANC22
JNSAD32
JNSAD32
JOBREN2
JTKOSA2
400/110
400/110
400/110
400/110
400/110
-JOBREN2
-JZREN22
-JVALJ32
-JTKOSB2
CAREVEC 1
JBGD20 A
JKRAG2 1
JNIS2 1
JPANC2 1
A
1
1
1
1
1
1
1
2
1
Area
3
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
RO
CS
CS
CS
CS
CS
CS
CS
CS
CS
CS
CS
CS
CS
CS
CS
RO
RO
RO
RO
CS
CS
CS
RO
RO
RO
RO
RO
RO
RO
RO
CS
RO
RO
BG
CS
CS
CS
CS
CS
CS
CS
BG
CS
CS
CS
RO
RO
RO
RO
RO
overloadings
/
out of limits voltages
4
TR 400/220kV/kV URECHESI
HL 220kV URECHESI-TG.JIU
HL 220kV TG.JIU-PAROSEN
TR 400/220kV/kV URECHESI
HL 220kV BRADU-TIRGOVI
HL 220kV URECHESI-TG.JIU
HL 220kV TG.JIU-PAROSEN
HL 220kV URECHESI-TG.JIU
HL 220kV TG.JIU-PAROSEN
TR 400/220kV/kV URECHESI
HL 220kV URECHESI-TG.JIU
HL 220kV P.D.F.A-RESITA
HL 220kV P.D.F.A-RESITA
HL 220kV TG.JIU-PAROSEN
TR 400/220kV/kV URECHESI
HL 220kV BUC.S-B-FUNDENI
TR 400/220kV/kV URECHESI
TR 400/220kV/kV BUC.S
TR 400/220kV/kV BUC.S
HL 220kV URECHESI-TG.JIU
HL 220kV TG.JIU-PAROSEN
HL 220kV BUC.S-B-FUNDENI
HL 220kV BUC.S-B-FUNDENI
HL 400kV GR.IAL-CERNAV
HL 400kV GR.IAL-CERNAV
TR 400/220kV/kV URECHESI
TR 400/220kV/kV URECHESI
HL 220kV BRADU-TIRGOVI
HL 220kV URECHESI-TG.JIU
HL 220kV TG.JIU-PAROSEN
HL 400kV MINTIA-SIBIU
TR 400/220kV/kV IERNUT
HL 220kV JBGD3 21-JOBREN2
HL 220kV JBGD172-JBGD8 22
HL 220kV JBGD3 21-JOBREN2
TR 400/220kV/kV JBGD8 1
HL 220kV JBGD3 21-JOBREN2
TR 400/220kV/kV JTKOSB1
TR 400/220kV/kV JTKOSB1
TR 400/220kV/kV JTKOSB1
HL 220kV JBGD3 21-JOBREN2
HL 220kV JBGD3 21-JOBREN2
HL 220kV JBGD3 21-JOBREN2
HL 220kV JTKOSA2-JTKOSB2
HL 220kV JBGD3 21-JOBREN2
HL 220kV JBGD3 22-JBGD8 22
HL 220kV JBGD3 21-JOBREN2
TR 400/220kV/kV URECHESI
TR 400/220kV/kV URECHESI
HL 220kV URECHESI-TG.JIU
HL 220kV TG.JIU-PAROSEN
TR 400/220kV/kV JTKOSB1
TR 400/220kV/kV JTKOSB1
TR 400/220kV/kV JTKOSB1
TR 400/220kV/kV URECHESI
HL 220kV URECHESI-TG.JIU
HL 220kV P.D.F.A-RESITA
HL 220kV P.D.F.A-RESITA
HL 220kV TG.JIU-PAROSEN
TR 400/220kV/kV URECHESI
HL 220kV URECHESI-TG.JIU
HL 220kV TG.JIU-PAROSEN
HL 220kV JBGD3 21-JOBREN2
TR 400/220kV/kV URECHESI
HL 220kV URECHESI-TG.JIU
TR 400/110kV/kV CAREVEC
HL 220kV JBGD3 21-JOBREN2
TR 400/110kV/kV JKRAG21
TR 400/110kV/kV JNIS2 1
TR 400/110kV/kV JPANC21
TR 400/220kV/kV JTKOSB1
TR 400/220kV/kV JTKOSB1
TR 400/220kV/kV JTKOSB1
TR 400/110kV/kV PLOVDIV4
TR 400/220kV/kV JTKOSB1
TR 400/220kV/kV JTKOSB1
TR 400/220kV/kV JTKOSB1
HL 220kV L.SARAT-FILESTI
TR 400/220kV/kV BUC.S
TR 400/220kV/kV BUC.S
HL 220kV BRADU-TIRGOVI
HL 220kV FUNDENI-BUC.S-B
#
5
1
1
1
1
1
1
1
1
1
1
1
1
2
1
1
1
1
1
2
1
1
1
1
2
1
1
1
1
1
1
1
1
1
2
1
1
1
1
2
3
1
1
1
2
1
2
1
1
1
1
1
1
2
3
1
1
1
2
1
1
1
1
1
1
1
2
1
2
2
2
1
2
3
2
1
2
3
1
1
2
1
1
limit
/
Unom
6
400MVA
277.4MVA
208.1MVA
400MVA
302.6MVA
277.4MVA
208.1MVA
277.4MVA
208.1MVA
400MVA
277.4MVA
277.4MVA
277.4MVA
208.1MVA
400MVA
320MVA
400MVA
400MVA
400MVA
277.4MVA
208.1MVA
320MVA
320MVA
791.2MVA
791.2MVA
400MVA
400MVA
302.6MVA
277.4MVA
208.1MVA
381.1MVA
400MVA
301MVA
365.8MVA
301MVA
400MVA
301MVA
400MVA
400MVA
400MVA
301MVA
301MVA
301MVA
365.8MVA
301MVA
365.8MVA
301MVA
400MVA
400MVA
277.4MVA
208.1MVA
400MVA
400MVA
400MVA
400MVA
277.4MVA
277.4MVA
277.4MVA
208.1MVA
400MVA
277.4MVA
208.1MVA
301MVA
400MVA
277.4MVA
275MVA
301MVA
300MVA
300MVA
300MVA
400MVA
400MVA
400MVA
275MVA
400MVA
400MVA
400MVA
277.4MVA
400MVA
400MVA
302.6MVA
320MVA
Flow
/
Voltage
7
451.1MVA
290.7MVA
282.7MVA
551.6MVA
274.9MVA
329.2MVA
318.2MVA
295.6MVA
283.8MVA
527.9MVA
355.6MVA
262.9MVA
262.9MVA
343.2MVA
518.9MVA
327.2MVA
505.8MVA
436MVA
436MVA
323.6MVA
312.2MVA
374.2MVA
304.5MVA
775.9MVA
773.4MVA
493.8MVA
498.6MVA
304.9MVA
319.2MVA
309MVA
345.1MVA
413.3MVA
314.9MVA
488.5MVA
308.5MVA
425.8MVA
296.7MVA
443.8MVA
463.8MVA
463.8MVA
303.6MVA
303.6MVA
291MVA
453.9MVA
490.7MVA
337MVA
298.4MVA
494.5MVA
496.6MVA
319.6MVA
309.6MVA
444.6MVA
464.6MVA
464.6MVA
511.8MVA
331.7MVA
266.3MVA
266.3MVA
320.3MVA
501MVA
323.4MVA
313MVA
313MVA
497MVA
318.6MVA
281.3MVA
289.5MVA
314.1MVA
353.3MVA
327.4MVA
472.2MVA
493.4MVA
493.4MVA
284.4MVA
448.2MVA
468.3MVA
468.3MVA
302.2MVA
447MVA
447MVA
302.8MVA
290.1MVA
rate
/
volt.dev.
8
112.8%
108.8%
145.2%
137.9%
106.9%
125.9%
168.0%
113.4%
151.7%
132.0%
135.8%
103.9%
103.9%
181.2%
129.7%
110.4%
126.4%
109.0%
109.0%
123.6%
165.0%
131.0%
103.3%
102.0%
101.6%
123.5%
124.7%
114.7%
121.2%
161.7%
102.9%
103.3%
110.0%
144.0%
107.6%
106.5%
103.4%
111.0%
115.9%
115.9%
105.2%
106.0%
100.8%
119.0%
187.1%
108.2%
104.1%
123.6%
124.1%
121.0%
161.5%
111.1%
116.1%
116.1%
128.0%
126.7%
106.1%
106.1%
169.1%
125.3%
122.9%
164.0%
110.1%
124.2%
120.6%
102.3%
100.6%
104.7%
117.8%
109.1%
118.1%
123.4%
123.4%
103.4%
112.0%
117.1%
117.1%
113.5%
111.8%
111.8%
123.3%
104.0%
5.34
Area
1
contingency
2
RO
TR 400/220 BUC.S
CS
TR 400/220 JBGD8
1
CS
TR 400/220 JBGD8
2
CS
TR 400/220 JNIS2
1
CS
TR 400/220 JTKOSB 1
RO
TR 400/220 MINTIA
1
RO
TR 400/220 SIBIU
1
RO
TR 400/220 SLATINA
1
Area
3
RO
RO
CS
CS
CS
CS
CS
CS
CS
CS
RO
RO
RO
RO
RO
RO
1
overloadings
/
out of limits voltages
4
HL 220kV BUC.S-B-FUNDENI
TR 400/220kV/kV BUC.S
HL 220kV JBGD3 21-JOBREN2
HL 220kV JBGD3 22-JBGD8 22
HL 220kV JBGD3 21-JOBREN2
TR 400/220kV/kV JTKOSB1
TR 400/220kV/kV JTKOSB1
TR 400/220kV/kV JTKOSB1
TR 400/220kV/kV JTKOSB1
TR 400/220kV/kV JTKOSB1
TR 400/220kV/kV URECHESI
HL 220kV URECHESI-TG.JIU
HL 220kV TG.JIU-PAROSEN
TR 400/220kV/kV SIBIU
TR 400/220kV/kV URECHESI
TR 400/220kV/kV SLATINA
#
5
1
2
1
2
1
1
2
3
2
3
1
1
1
2
1
2
limit
/
Unom
6
320MVA
400MVA
301MVA
365.8MVA
301MVA
400MVA
400MVA
400MVA
400MVA
400MVA
400MVA
277.4MVA
208.1MVA
400MVA
400MVA
400MVA
Flow
/
Voltage
7
387.2MVA
577.8MVA
340.4MVA
398.3MVA
315.3MVA
447.5MVA
467.6MVA
467.6MVA
610MVA
610MVA
501.5MVA
323MVA
311.7MVA
490.9MVA
518.8MVA
460.5MVA
rate
/
volt.dev.
8
138.6%
144.4%
120.3%
118.3%
110.4%
111.9%
116.9%
116.9%
152.5%
152.5%
125.4%
122.7%
163.7%
122.7%
129.7%
115.1%
SVK
AUT
HUN
SLO
ROM
CRO
SRB
BIH
MNG
BUL
MKD
TUR
ALB
GRE
critical elements
400 kV line or 400/x kV transformer
220 kV line or 220/x kV transformer
Figure 5.5.6 – geographical position of critical elements for 2015-average hydrology high load scenario –
topology 2010
Some of these problems identified are connected to the problem of new generation units
implemented in the model. For some, there is great insecurity how these units will be connected to
the transmission network, especially in the case of new generation capacities in Kosovo B region
(Serbia-UNIMK). In this regime new 3000 MW generation capacity is analyzed. Problem of
feasible connection of such a large generation capacity is not in the scope of this study and separate
study, that will analyze this problem alone is necessary to give right answer to this question.
All in all, it can be concluded that this regime is not feasible without major reinforcement of
transmission network.
5.35
5.6 Scenario 2015 – average hydrology – high load – topology 2015
This part of the Study presents the results of static load flow and voltage profile analyses that are
conducted for complete network topology and (n-1) contingencies in the scenario which is denoted
as GTmax run for year 2015 - average hydrology high load scenario and expected network topology
for 2015 and new planned generation capacities implemented.
5.6.1 Lines loadings
Figure 5.6.1 shows power exchanges between areas for 2015-average hydrology high load scenario.
Power flows along interconnection lines in the region together with balances of the systems are
shown in Figure 5.6.2. Area totals are shown in Table 5.6.1 and comparison of the average hydro
regimes with normal load projection and high load projection is shown in Table 5.6.2. Difference of
load-demand level for these two regimes is around 8.11% on regional level. As a consequence of
this high load level, network losses are increased by 251 MW or 25%. Figure 5.6.3 shows
histogram of tie lines loadings. It is concluded that most of the tie lines are loaded less than 50% of
their thermal limits.
UKR
450
SVK
61
9
38
3
17
AUT
HUN
89
49
15
1
175
SLO
4
26
ITA
CRO
ROM
305
10
30
55
835
SRB
208
BIH
0
71
95
32
MNG
BUL
691
91
1
68
56
ALB
23
7
170
MKD
32
7
56
TUR
57
GRE
Figure 5.6.1 - Area exchanges in analyzed electric power systems for 2015-average hydrology high load scenario –
2015 topology
As it can be seen, new elements that are expected to be build till 2015 cause totally different
distribution of power flows in the southern part of the region (Albania, FYR of Macedonia, Serbia
and Montenegro). Compared to the expected topology 2010, analyzed in previous chapter, it can be
seen that the network losses are decreased as a consequence of building of new elements for 2015
network topology, especially in the cases of Albania, Serbia and Montenegro. Overall reduction of
loses is around 39 MW.
5.36
LEVIC 1400.0
CENTREL
200
125
251
107
GABICK1400.0
450.0 MW
MSAJO 4
410.0
406.9
220.7
224.4
UMUKACH
412.0
62.
264 4
18.0
NADAB
163
385.4
392.0
71.0
70.9
70.7 ARAD
123
382.4
ROM
ISACCEA 388.6
-1040.2 MW
JSUBO31 385.4
396.2
JHDJE11
P.D.FIE
55.4
74.2
55.4
75.2
95.1
125
95.5 SOFIA_W4
60.8
407.1
TANTAREN
209
112
396.9
157
3.0
403.6
UGLJEVIK 396.7
SCG
50.6
29.4
JHPERU21
51.0 228.1
36.2
386
61.2
388
JPODG211
77.7
JPODG121227.2
JPRIZ22
228.7 JTKOSA2
231.8JTKOSB1420.0
-617.0 MW
MKD
184
141
386.0
412.1
2
7. .5
53
STIP 1
408.3
DUBROVO
AZEMLA1
56.2
146
KARDIA K
56.8 416.9
77.2
QES/H
9.9
5.9
23
11 6
4
AHS_FLWR416.0
K
412.4
KV: 110 , 220 , 400
415.5
6
12 0.1
7
56.4
23.7
K-KEXRU
417.8
4HAMITAX
25.0
25.0 414.9
65.5
56.4 54.2
4BABAESK
72.3
415.3
0.0 MW
0.0 MW
KARACQOU
250 419.1
50.0
FILIPPOI
3
25 1.4
.4
.4
31 7.0
7
160
107
.5
93 .4
29
LAGAD K 413.4
404.2
MI_3_4_1 417.5
BLAGOEV410.3
408.9
ALB
-883.0 MW
409.3
.0 8
25 5.
2
237
0
76.
2
7. 6
.
39
406.1
160
127
AKASHA1
184
141
BITOLA 2
C_MOGILA
1
27 84
.7
SK 4
406.9
899.0 MW
JVRAN31
401.5
183
68.0
SK 1 223.8
DOBRUD4 409.2
410.0
BUL
JLESK21
397.2
9.8
67.4
1
17 11
.5
AVDEJA2
222.7AFIERZ2223.7
395.0
77.9
38.8
17.6
36.9
109
39.7
570
142
185
107
185
107
725
134
9
78. 6
32.
566
131
9
78. 6
32.
15 112
.3
395.7
AEC_400
JNIS2 1
396.5
12
57. 6
1
CRG
587.0 MW
717
89.1
RP TREB
230.9
JHPIVA21
258 235.7
52.3
77.9
38.8
SA 20
226.0
2638.9 MW
SRB
2051.9 MW
93.2
62.3
JVARDI22
50.8 227.4
59.8
50.7
58.1
17.8
28.4
VISEGRA
229.3
95.1 94.9
46.2 81.9
229.8
AVDEJA1
GIS REGIONAL MODEL - AVERAGE HYDRO
2015 HIGH LOAD - NEW GENERATION - TOPOLOGY 2015
ROSIORI 383.9
306
JSMIT21
114
108
13.4
398.4
195 MO-4
5.2
230.5
BIH
112.9 MW
18.3
148
0
21 6
.
70
0
21 6
.
70
MO-4
223.7
TE TUZL
RP TREB
398.3
SHAW POWER
TECHNOLOGIES
INC. R
UKR
450.0 MW
411
123
114
1
11 07
10 4
7
JSOMB31
382.4
304
81.8
401.9
168
100
190
11.2
513
3.1
GRADACAC
219.7
5
10 7
.
45
PRIJED2
507
1.9
6
55. 9
22.
.2
55 .4
39
19.5
13.2
CRO
-1448.1 MW
393.7
15.0
45.3
MSAFA 4
397.0
HE ZAKUC
224.5
14.9
73.9
5
17 7
15
38 157
.7
19.6
5.4
54.7
32.8
221.3
104
47.2
DAKOVO
222.1MEDURIC
55.3
27.4
399.2
MRACLIN
167
50.7
ERNESTIN
1
8. .4
10
12.1
15.2
226.6
.2
.2
38 .5 38 .5
90
90
ZERJAVIN
401.4
ZERJAVIN 226.1
1
16 8
.
11
1
16 8
.
11
TUMBRI
KONJSKO
10.7
28.9
67.6 67.6
14.1 34.9
LCIRKO2
1
61 14
.1
1
61 14
.1
399.8
400.0
PEHLIN 228.0
10.8
7.3
MBEKO 4
392.7
407.2
MPECS 4
LKRSKO1
MELINA
35.7
UMUKACH2
27.0
176
176
SLO
12
5. .1
2
35.5
20.2
38.4
15.5
38.4
15.5
215.0 MW
91
400.1 7. .5
6
LDIVAC2
228.9
MHEVI 4
8.1
0.6
10.2
86.6
22.6 228.8
27.0
LMARIB1 401.8
91.6
15.9
230.0
10.1
83.3
178
49.1
22.7
13.1
MTLOK 2
230.4
HUN
161
12.7
IPDRV121
178
28.4
51.9
21.2
MKISV 2
226.2
-1249.9 MW
161
12.7
400.7
MGYOR 2
52.4
6.3
41.
6.5 7
41.
6.5 7
152
48.
7
IRDPV111
406.4
2
60. 4
30
LPODLO2228.5
MSAJI 408.1
4
417
57.5
41.7
37.7
41.7
37.7
234.1
ONEUSI2
233.4
LDIVAC1
MGOD
OWIEN 2
350
UZUKRA01
282
772.5
25.0
54.2
OKAINA1 401.0
155
52.2
OOBERS2238.7
346
813
209
112
98.4
22.2
405.2
199
78.7
UCTE
222.4 MW
MAISA 7
712.1
MGYOR 4
0
25 .2
98.3
84
103
403.1
OWIEN 1
GRE
TUR
BUS -VOLTAGE(KV)
BRANCH -MW/MVAR
EQUIPMENT -MW/MVAR
Figure 5.6.2 - Power flows along interconnection lines in the region with balances of the systems for 2015-dry hydrology scenario
5.37
Table 5.6.1 - Area totals in analyzed electric power systems for 2015-average hydrology high load scenario –
2015 topology
AREA
GENERATION
LOAD
LOSSES
INTERCHANGE
ALBANIA
897.7
1684
96.7
-883
BULGARIA
8157.4
7083
175.4
899
BIH
2479.7
2272
94.7
112.9
CROATIA
2592.1
3959
81.1
-1448.1
MACEDONIA
895.7
1489
23.7
-617
ROMANIA
7676.4
8269.8
446.7
-1040.1
SERBIA
10023.9
7660
311.9
2051.9
MONTENEGRO
1322.6
708
27.6
587
TOTALS
34045.5
33124.8
1257.8
-337.4
Table 5.6.2 - Comparison of Area totals in analyzed electric power systems for 2015- average hydrology versus average
hydrology high load scenario –2015 topology
LOAD
LOSSES
AREA
normal
normal
high load
high load
load
load
ALBANIA
1541
1684
9.28%
73.1
96.7
32.28%
BULGARIA
6483
7083
9.25%
150.9
175.4
16.24%
BIH
2279
2272
-0.31%
79.2
94.7
19.57%
CROATIA
3657
3959
8.26%
64
81.1
26.72%
MACEDONIA
1407
1489
5.83%
21.4
23.7
10.75%
ROMANIA
7317.4
8269.8
13.02%
343.1
446.7
30.2%
SERBIA
7279
7660
5.23%
254.4
311.9
22.6%
MONTENEGRO
676
708
4.73%
20.3
27.6
35.96%
TOTALS
30639.4
33124.8
8.11%
1006.4
1257.8
24.98%
Histogram
40
35
Frequency
30
25
20
15
10
5
0
x<25
25<x<50
50<x<75 75<x<100
x>100
Bin
Figure 5.6.3 - Histogram of interconnection lines loadings for 2015-average hydrology high load scenario –
2015 topology ("Frequency" denotes number of lines and "Bin" denotes loading range in % of thermal limit)
Following Table 5.6.3 shows all network elements loaded over 80% of their thermal limits.
Histogram of branch loadings in the system is shown in Figure 5.6.4. As it can be seen large
number of lines 220 kV voltage level in Albania, Romania and Serbia are loaded over 80% for the
reasons described in previous chapter in detail. It can be concluded that new expected network
topology for 2015 compared to 2010 topology has small influence in overall load level of the
transmission network. Almost the same elements are found critical as for topology 2010, especially
in parts of the network that supply major consumption areas (Tirana in Albania, Bucharest and
Timisoara in Romania, Belgrade in Serbia and Pristina in Serbia-UNMIK).
5.38
Table 5.6.3 - Network elements loaded over 80% of their thermal limits for 2015-average hydrology high load scenario
– 2015 topology
BRANCH LOADINGS ABOVE
AREA
ALB
ROM
SRB
ALB
BIH
CRO
ROM
SRB
MNG
80.0 % OF RATING:
LOADING
MVA
Lines
HL 220kV AKASHA2-ARRAZH2 1
290.4
HL 220kV BRADU-TIRGOVI 1
283.9
HL 220kV BUC.S-B-FUNDENI 1
307.5
HL 220kV FILESTI-BARBOSI 1
245.8
HL 220kV L.SARAT-FILESTI 1
239
HL 220kV LOTRU-SIBIU 1
281.2
HL 220kV LOTRU-SIBIU 2
281.2
HL 220kV P.D.F.A-CETATE1 1
206.3
HL 220kV P.D.F.A-RESITA 1
262.1
HL 220kV P.D.F.A-RESITA 2
262.1
HL 220kV P.D.F.II-CETATE1 1
269.9
HL 220kV RESITA-TIMIS 1
225
HL 220kV RESITA-TIMIS 2
225
HL 220kV TG.JIU-PAROSEN 1
313.8
HL 220kV URECHESI-TG.JIU 1
313.5
HL 220kV JBGD3 21-JOBREN2 1
292
Transformers
TR 220/110 kV ABURRE 1
52.2
TR 220/110 kV ABURRE 2
52.2
TR 220/110 kV ABURRE 3
52.2
TR 220/110 kV AELBS1 1
87.8
TR 220/110 kV AELBS1 2
87.8
TR 220/110 kV AELBS1 3
94.4
TR 220/110 kV AFIER 1
151.7
TR 220/110 kV AFIER 2
124.4
TR 220/110 kV AFIER 3
118.7
TR 220/110 kV AFIERZ 1
59.9
TR 220/110 kV AFIERZ 2
59.9
TR 220/110 kV AKASHA 1
94.7
TR 220/110 kV AKASHA 2
94.7
TR 220/110 kV ARRAZH 1
87.5
TR 220/110 kV ARRAZH 2
87.5
TR 220/110 kV ATIRAN 2
96.4
TR 220/110 kV ATIRAN 3
100.6
TR 220/110 kV MO-4 3
124.4
TR 400/110 kV UGLJEV 1
256.1
TR 220/110 kV TESISA 1
179.4
TR 220/110 kV BARBOS 1
166.6
TR 220/110 kV FUNDE2 1
239.8
TR 220/110 kV FUNDEN 1
203
TR 220/110 kV TIMIS 1
170.4
TR 400/110 kV BRASOV 1
230.2
TR 400/110 kV CLUJ E 1
206.4
TR 400/110 kV DIRSTE 1
219.6
TR 400/220 kV BUC.S 1
376.5
TR 400/220 kV BUC.S 2
376.5
TR 400/220 kV IERNUT 1
404.8
TR 400/220 kV URECHE 1
479.7
TR 220/110 kV JBGD3 1
174.8
TR 220/110 kV JBGD3 2
137.5
TR 220/110 kV JTKOSA 2
133.5
TR 220/110 kV JTKOSA 3
135.9
TR 220/110 kV JZREN2 2
133.8
TR 400/220 kV JBGD8 1
330.3
TR 400/220 kV JTKOSB 1
351.9
TR 400/220 kV JTKOSB 2
367.7
TR 400/220 kV JTKOSB 3
367.7
TR 220/110 kV JTPLJE 1
102.4
ELEMENT
RATING
MVA
PERCENT
270
302.6
320
277.4
277.4
277.4
277.4
208.1
277.4
277.4
277.4
277.4
277.4
208.1
277.4
301
107.6
93.8
96.1
88.6
86.2
101.4
101.4
99.1
94.5
94.5
97.3
81.1
81.1
150.8
113
97
60
60
60
90
90
90
120
90
90
60
60
100
100
100
100
120
120
150
300
200
200
200
200
200
250
250
250
400
400
400
400
200
150
150
150
150
400
400
400
400
125
87
87
87
97.6
97.6
104.8
126.4
138.2
131.9
99.8
99.8
94.7
94.7
87.5
87.5
80.3
83.9
83
85.4
89.7
83.3
119.9
101.5
85.2
92.1
82.6
87.8
94.1
94.1
101.2
119.9
87.4
91.7
89
90.6
89.2
82.6
88
91.9
91.9
82
Histogram
300
Frequency
250
200
150
100
50
0
x<25
25<x<50 50<x<75 75<x<100
x>100
Bin
Figure 5.6.4 - Histogram of branch loadings for 2015-average hydrology scenario – 2015 topology ("Frequency"
denotes number of lines and "Bin" denotes loading range in % of thermal limit)
5.39
This leads to conclusion that planned network reinforcements compared to network topology 2010,
reduce loading of some elements in southern part of Serbia, and that in order for transmission
network to sustain this load-demand level and this production pattern, additional network
reinforcements are necessary, especially in increasing transformer capacity in substations that
supply major consumption areas in Albania, Romania and Serbia.
5.6.2 Voltage Profile in the Region
Figure 5.6.5 shows histogram of voltages in monitored substations. Like for the investigated
topology 2010, overall voltage profile is not adequate, but comparing to 2010 topology voltage
profile is much better in monitored parts of Albanian and Serbian network. In some parts of the
network, especially in major consumption areas of Romania, voltages in some monitored
substations are below allowed limits. Using of voltage control devices (tap changing transformers,
shunt devices) can improve voltage profile in this region, but it can be concluded that conditions in
some parts of the transmission network of Romania are dangerously close to voltage collapse. Since
this is local problem, detailed analyses is necessary to analyze this situation.
Histogram
80
70
Frequency
60
50
40
30
20
10
x>1.1
1.05<x<1.1
1.025<x<1.05
1<x<1.025
0.975<x<1
0.95<x<0.975
0.9<x<0.95
x<0.9
0
Bin
Figure 5.6.5 - Histogram of voltages in monitored substations for 2015-average hydrology high load scenario –
2015 topology ("Frequency" denotes number of busses and "Bin" denotes voltage range in p.u.)
5.6.3 Security (n-1) analysis
Results of security (n-1) analysis for 2015-average hydrology high load scenario and expected
network topology for 2015 are presented in Table 5.6.4. Figure 5.6.6 shows the geographical
position of the critical elements in monitored systems.
5.40
Like for expected topology 2010 (previous chapter), it can be concluded that all identified insecure
situations are located in internal networks that belong to monitored power systems of Albania,
Romania and Serbia. Also, the planned network reinforcements till 2015 resolve some of the
noticed critical contingencies, especially in southern part of Serbia. Compared to the similar
scenario with normal load projection, it can be concluded that higher load-demand projected and
new production capacities modeled, increase the overload level of some critical elements. The rest
of the conclusions are the same as in case of the analyzed normal load projection and expected
topology 2015, and that is that certain level of network reinforcement is necessary to make this
regime more secure.
SVK
AUT
HUN
SLO
ROM
CRO
SRB
BIH
MNG
BUL
MKD
TUR
ALB
GRE
critical elements
400 kV line or 400/x kV transformer
220 kV line or 220/x kV transformer
Figure 5.6.6 – geographical position of critical elements for 2015-average hydrology high load scenario –
2015 topology
5.41
Table 5.6.4 - Network overloadings for 2015-average hydrology high load scenario , single outages – 2015 topology
Area
1
contingency
2
BASE CASE
AL
OHL 220kV AELBS12 -AFIER 2
1
RO
OHL 220kV P.D.F.A -RESITA
1
RO
OHL 220kV RESITA
-TIMIS
1
RO
OHL 220kV PESTIS
-MINTIA A 1
RO
OHL 220kV CLUJ FL -AL.JL
1
RO
OHL 220kV AL.JL
1
RO
OHL 400kV TANTAREN-URECHESI 1
RO
OHL 400kV TANTAREN-BRADU
1
RO
OHL 400kV TANTAREN-SIBIU
1
RO
OHL 400kV URECHESI-P.D.FIE
1
RO
OHL 400kV URECHESI-DOMNESTI 1
RO
OHL 400kV MINTIA
-ARAD
1
RO
OHL 400kV MINTIA
-SIBIU
1
RO
OHL 400kV DOMNESTI-BRAZI
1
RO
OHL 400kV SMIRDAN -GUTINAS
1
RO
OHL 400kV BRASOV
-BRADU
1
CS
CS
RO
RO
CS
CS
RO
RO
OHL 220kV JBGD172
OHL 400kV JBGD8 1
TR 400/110 BRASOV
TR 400/110 DIRSTE
TR 400/110 JNIS2
TR 400/110 JPANC2
TR 400/220 BUC.S
TR 400/220 IERNUT
-JBGD8 22 1
-JOBREN11 1
1
1
1
1
1
1
-GILCEAG
overloadings
/
Area
out of limits voltages
3
4
RO
HL 220kV URECHESI-TG.JIU
RO
HL 220kV TG.JIU-PAROSEN
AL
HL 220kV AKASHA2-ARRAZH2
RO
HL 220kV URECHESI-TG.JIU
RO
HL 220kV P.D.F.A-RESITA
RO
HL 220kV TG.JIU-PAROSEN
RO
HL 220kV URECHESI-TG.JIU
RO
HL 220kV RESITA-TIMIS
RO
HL 220kV TG.JIU-PAROSEN
RO
HL 220kV URECHESI-TG.JIU
RO
HL 220kV TG.JIU-PAROSEN
RO
HL 220kV TG.JIU-PAROSEN
RO
HL 220kV URECHESI-TG.JIU
RO
HL 220kV TG.JIU-PAROSEN
RO
HL 220kV URECHESI-TG.JIU
RO
HL 220kV URECHESI-TG.JIU
RO
HL 220kV TG.JIU-PAROSEN
RO
HL 220kV URECHESI-TG.JIU
RO
HL 220kV TG.JIU-PAROSEN
RO
HL 220kV TG.JIU-PAROSEN
RO
TR 400/220kV/kV URECHESI
RO
HL 220kV URECHESI-TG.JIU
RO
HL 220kV TG.JIU-PAROSEN
RO
HL 220kV TG.JIU-PAROSEN
RO
HL 220kV URECHESI-TG.JIU
RO
HL 220kV TG.JIU-PAROSEN
RO
HL 220kV URECHESI-TG.JIU
RO
HL 220kV TG.JIU-PAROSEN
RO
HL 220kV URECHESI-TG.JIU
RO
HL 220kV TG.JIU-PAROSEN
RO
HL 220kV URECHESI-TG.JIU
RO
HL 220kV TG.JIU-PAROSEN
CS
HL 220kV JBGD172-JBGD8 22
CS
HL 220kV JBGD3 21-JOBREN2
RO
TR 400/110kV/kV DIRSTE
RO
TR 400/110kV/kV BRASOV
CS
TR 400/110kV/kV JNIS2 1
CS
TR 400/110kV/kV JPANC21
RO
TR 400/220kV/kV BUC.S
RO
HL 220kV STEJARU-GHEORGH
#
5
1
1
1
1
2
1
1
2
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
1
1
1
2
2
2
1
limit
/
Unom
6
277.4MVA
208.1MVA
270MVA
277.4MVA
277.4MVA
208.1MVA
277.4MVA
277.4MVA
208.1MVA
277.4MVA
208.1MVA
208.1MVA
277.4MVA
208.1MVA
277.4MVA
277.4MVA
208.1MVA
277.4MVA
208.1MVA
208.1MVA
400MVA
277.4MVA
208.1MVA
208.1MVA
277.4MVA
208.1MVA
277.4MVA
208.1MVA
277.4MVA
208.1MVA
277.4MVA
208.1MVA
365.8MVA
301MVA
250MVA
250MVA
300MVA
300MVA
400MVA
208.1MVA
Flow
rate
/
/
Voltage volt.dev.
7
8
285.6MVA
100.1%
285.6MVA
133.5%
289.9MVA
112.2%
301.5MVA
106.1%
331.5MVA
122.8%
301.5MVA
141.4%
295.7MVA
103.9%
348.7MVA
127.0%
295.7MVA
138.5%
296.3MVA
104.8%
290.4MVA
139.7%
271MVA
126.4%
308.8MVA
109.1%
308.8MVA
145.4%
292.5MVA
102.6%
300.1MVA
105.9%
300.1MVA
141.1%
331.1MVA
117.3%
331.1MVA
156.4%
268.7MVA
124.5%
400.6MVA
100.1%
305.6MVA
107.7%
305.6MVA
143.6%
263.1MVA
123.1%
326.2MVA
115.5%
326.2MVA
153.9%
296.2MVA
104.0%
296.2MVA
138.7%
296.9MVA
104.5%
296.9MVA
139.3%
296.3MVA
104.3%
296.3MVA
139.0%
453.5MVA
129.0%
293.8MVA
102.6%
353.9MVA
141.6%
350MVA
140.0%
323.1MVA
107.7%
318.8MVA
106.3%
405.6MVA
101.4%
196.5MVA
106.0%
5.6.4 Summary of Impacts - 2015 topology versus 2010 topology
Compared to the expected topology 2010, analyzed in previous chapter, it can be seen that the
network losses are smaller as a consequence of building of new elements for 2015 network
topology, especially in the cases of Albania, Serbia and Montenegro. Overall reduction of loses is
around 39 MW.
Also, planned network reinforcements compared to network topology 2010, reduce load of some
elements in southern part of Serbia. Compared to the 2010 topology, voltage profile is somewhat
better, especially in southern and central part of Serbia, as well as in Albania. This is direct
consequence of the building of the new 400 kV line Nis-Leskovac-Vranje-Skopje and substation
400/110 kV Vranje. However, voltage profile is still not satisfying.
Realization of the planed investments till 2015 has impact on secure operation of the network.
Some of the insecure states identified with 2010 topology are relieved and do not exist with
expected 2015 network topology.
Main conclusion can be that even with 2015 topology, this load-generation regime is not feasible
enough and additional network reinforcements are necessary, especially in the major consumption
area.
5.42
6 ANALYSES SUMMARY AND RECOMMENDATIONS
6.1
In this chapter, comparison of the all analyzed scenarios is presented. Main issues are impact of
different generation patterns and planed network investments on system losses and system security.
The authors tried to give main indicators how to resolve some insecure states identified, based on
prerequisites and assumptions presented in Chapter 2 and only results of the analyses of scenarios.
Deeper impacts of the proposed remedies are not analyzed whether they are or not in compliance
with long term investments plans of the countries in the region.
6.1 Reference cases
Reference cases consist of nine cases. Three different hydrology scenarios are analyzed (average,
dry and wet) for years 2010 and 2015. Also, for each of these generation-load patterns in year 2015
two cases are analyzed (first for 2010 topology, i.e. without any new line or transformer planned for
commissioning between 2010 and 2015 and second for expected topology in year 2015).
6.1.1 Analyses comparison
Main focus of this subchapter is to analyze influence of different generation patterns and planed
network investments on overall network performance and system losses.
Table 6.1.1 shows overview of losses on regional level. It can be seen that new elements planned
for commissioning between 2010 and 2015 influence reduction of losses in analyzed region, no
matter of hydrology. This reduction of losses is between 30 and 40 MW.
Table 6.1.1: Peak Hour - Regional Losses (MW)
Topology
2010
2015
Dry
756.9
-
Year 2010
Average
737.1
-
Wet
741.5
-
Dry
951.6
909.6
Year 2015
Average
1,035.1
1,006.4
Wet
976.6
941.8
These changes are consequence of better voltage profile in analyzed region, especially in areas
directly influenced by the new elements. For example, new interconnection line Nis-LeskovacVranje-Skopje greatly improves voltage profile in southern part of Serbia and new interconnection
lines Podgorica – V.Dejes – Kashar – Elbasan and Kosovo B – V.Dejes – Kashar – Elbasan greatly
improve voltage profile in Albania. Higher voltages and reduction of load of some elements that are
highly influenced by the new elements, reduce power losses mainly in Albania, Southern Serbia,
and in the region as whole.
Table 6.1.2 shows overview of area summaries for analyzed reference cases. Comparing cases with
different hydrology (for the same year and topology) it can be concluded following:
ƒ Since analyzed region is balanced, deficit of power produced by hydro power plants is
compensated by greater power produced by thermal power plants. Different hydrology
causes different generation and interchange patterns.
ƒ Because of different generation and interchange patterns, power flows are different and this
cause different level of losses, but it is not possible to establish clear connection between
these factors.
ƒ Level of intersystem interchanges is much higher than in present or history. It has to be
stated that networks of single systems are developed mainly to cover their demand with their
own production capacities, so this kind of power plant engagement and level of interchanges
cause power flows and network status that are different from ones recorded. This goes
especially Albanian and Serbian system.
6.2
Table 6.1.2: Overview of area summaries for analyzed reference cases (in MW)
Year
2010
Hydrology
Topology
Average
2010
Dry
2010
Wet
2010
2010
Average
2015
2010
2015
Dry
2015
2010
Wet
2015
Data type
Albania
B&H
Bulgaria
Croatia
Macedonia
Montenegro
Romania
Serbia
Generation
Load
Losses
Interchange
Generation
Load
Losses
Interchange
Gen
Load
Losses
Interchange
Generation
Load
Losses
Interchange
Generation
Load
Losses
Interchange
Generation
Load
Losses
Interchange
Generation
Load
Losses
Interchange
Generation
Load
Losses
Interchange
Generation
Load
Losses
Interchange
896.7
1287.3
50.4
-441.0
953.3
1290.5
46.8
-384.0
898.6
1287.3
50.3
-439.0
1116.1
1531.0
81.2
-496.0
1118.1
1541.0
73.1
-496.0
894.7
1521.9
92.7
-720.0
893.6
1542.0
71.6
-720.0
1124.5
1528.9
85.8
-490.1
1111.0
1528.9
72.1
-490.0
2266.1
1971.3
58.6
236.2
2842.4
1989.0
60.3
793.0
2037.1
1961.1
57.0
19.0
2316.6
2279.0
78.7
-41.1
2317.2
2279.0
79.2
-41.0
3140.9
2304.0
78.2
758.8
3141.2
2305.0
77.1
759.0
2154.2
2278.6
74.8
-199.3
2155.1
2278.6
75.6
-199.0
6900.4
5977.3
121.6
787.0
6709.6
5970.0
133.6
606.0
6940.4
5966.2
120.9
839.0
7332.7
6483.0
150.7
699.0
7332.9
6483.0
150.9
699.0
7363.9
6450.0
147.9
766.0
7363.9
6450.0
147.8
766.0
7553.2
6446.1
137.4
955.0
7553.3
6446.1
137.5
955.0
1502.9
3136.7
49.0
-1682.8
2294.6
3147.0
41.6
-894
1735.4
3136.7
50.7
-1452.0
2316.3
3657.0
63.4
-1404.1
2317.0
3657.0
64.0
-1404.0
2636.7
3665.0
58.2
-1086.4
2637.7
3665.0
58.7
-1085.9
2799.9
3660.4
60.9
-921.3
2800.9
3660.4
61.6
-921.1
950.3
1198.2
20.1
-268.0
1021.1
1206.0
18.1
-203.0
1081.8
1194.0
19.8
-132.0
1087.0
1407.0
20.0
-340.0
1088.4
1407.0
21.4
-340.0
985.1
1410.0
19.1
-444.0
985.2
1409.0
20.2
-444.0
869.5
1407.6
18.9
557.0
870.8
1407.6
20.2
-557.0
539.8
669.2
15.5
-147.0
467.8
671.0
16.9
-220.0
586.3
669.2
15.9
-101.0
735.0
671.0
25.0
39.0
735.2
676.0
20.3
39.0
586.3
672.0
22.1
-107.7
586.3
678.0
16.4
-108.0
778.7
669.8
21.6
85.0
774.1
669.8
16.9
85.0
7939.9
6728.3
201.2
930.4
6742.0
6703.8
238.0
-199.9
7342.4
6423.7
206.6
632.1
7712.8
7317.4
347.4
47.9
7708.6
7317.4
343.1
48.1
7724.6
7798.4
295.5
-369.2
7722.9
7798.4
293.5
-368.9
7953.9
7704.0
298.9
-123.3
7950.3
7704.0
294.8
-123.0
7355.9
6873.1
220.8
248.4
7311.7
6944.0
201.6
166.0
7406.2
6875.1
220.4
297.1
8687.6
7263.0
268.7
1156.0
8689.4
7279.0
254.4
1156.0
8397.9
7296.0
237.9
864.0
8396.3
7308.0
224.3
864.0
8413.1
7209.2
278.4
912.7
8398.5
7209.2
263.2
913.0
Total
(SE Europe)
28351.9
27841.4
737.1
-336.7
28342.5
27921.3
756.9
-335.9
28028.1
27513.3
741.5
-336.7
31304.1
30608.4
1035.1
-339.3
31306.8
30639.4
1006.4
-338.9
31730.1
31117.3
951.6
-338.5
31727.1
31155.4
909.6
-337.8
31646.9
30904.6
976.6
-338.3
31613.9
30904.6
941.8
-337.2
6.3
Table 6.1.3: Overview of high loaded elements in analyzed reference cases
AREA
ELEMENT
ALB OHL 220
BUL OHL 220
OHL 220
OHL 220
OHL 220
OHL 220
OHL 220
ROM
OHL 220
OHL 220
OHL 220
OHL 220
OHL 220
SRB OHL 220
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
AKASHA2-ARRAZH2
MI_2_220-ST ZAGORA
MINTIA-SIBIU
P.D.F.II-CETATE1
P.D.F.A-CETATE1
P.D.F.A-RESITA ckt.1
P.D.F.A-RESITA ckt.2
LOTRU-SIBIU ckt.1
LOTRU-SIBIU ckt.2
URECHESI-TG.JIU
TG.JIU-PAROSEN
BUC.S-B-FUNDENI
JBGD3 21-JOBREN2
LOAD LEVEL
HYDROLOGY average
TOPOLOGY
RATING LOAD
%
MVA
MVA
2010
dry
LOAD
MVA
wet
%
LOAD
MVA
Lines
270
228.6
381.1
277.4
208.1
277.4
277.4
277.4
277.4
277.4
208.1
320
301
%
average
2010
2015
LOAD
LOAD
%
%
MVA
MVA
250
92.6
238
88
268
205
236
236
96.5
98.6
85.2
85.2
268
205
232
232
96.4
98.5
83.7
83.7
278
278
100
134
273
273
98.3
131
261
86.7
262
87
51.9
51.9
78
78
83.8
84.1
84.1
117 93.7 107
89
117 97.1 138
95.7 106 87.5 97.2 95.5 106 113
91.4 102 83.5 92.8 91.1 101 108
86.5
86.5
86.6
86.6
93.1
84.1
84.1
115
126
120
50.3
50.3
74.3
74.3
79.8
82.1
82.1
135
111
106
83.8
83.8
82.5
82.5
88.7
82.1
82.1
113
123
117
255
395
84.9
98.9
193
163
96.7
81.4
320
80.1
167
126
124
131
133
83.4
83.7
82.4
87.1
88.7
326
326
81.4
81.4
276
276
99.5
99.5
182
87.3
2015
dry
2010
LOAD
%
MVA
wet
2015
LOAD
%
MVA
2010
LOAD
%
MVA
2015
LOAD
%
MVA
259
191
324
269
206
239
239
304
304
280
280
285
313
96
83.5
85.1
97
99
86.2
86.2
109
109
101
135
89.2
104
237
190
313
268
206
236
236
303
303
275
275
285
313
87.8
83.3
82.2
96.8
98.8
84.9
84.9
109
109
99.3
132
88.9
104
91.1
91.1
90.5
90.5
97.2
87.2
50.2
50.2
74.8
74.8
80.4
82.3
82.3
135
110
105
83.7
83.7
83.2
83.2
89.3
82.3
82.3
112
123
117
275
102
243
89.9
261
294
81.6
97.5
261
295
81.4
97.9
98.3
98.3
94.1
94.1
101
90.8
90.8
122
134
128
84.7
84.7
81.9
80.8
52.1
52.1
75.4
75.4
81
83.5
83.5
137
112
107
86.9
86.9
83.8
83.8
90
83.5
83.5
114
125
119
54.6
54.6
81.4
81.4
87.5
87.2
253
391
59
59
84.7
84.7
91
90.8
90.8
147
120
115
84.7
84.7
98.2
84.2 242
97.7
241
80.2
193
163
96.5
81.3
200
169
99.9
84.2
200
168
99.8
84.1
270
414
340
340
209
176
90
104
85
85
104
88
268
410
339
339
208
176
89.2
103
84.8
84.8
104
87.8
325
81.3
324
81.1
171
130
85.6
86.5
171
130
85.5
86.6
83.7
98.3
89.6
81
88.9
90.5
80.8
84.4
84.4
81.1
81.1
84.5
98.6
89.7
81.3
87.9
89.5
335
197
134
122
133
136
323
338
338
122
122
338
197
135
122
132
134
325
325
81.1
81.1
247
82.5
Transformers
ALB
B&H
ROM
SRB
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
220/110
220/110
220/110
220/110
220/110
220/110
220/110
220/110
220/110
220/110
220/110
220/110
220/110
400/110
400/220
400/220
400/220
220/110
220/110
400/220
400/220
400/220
220/110
220/110
220/110
220/110
220/110
400/220
400/220
400/220
220/110
220/110
400/110
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
AFIERZ2-AFIERZ5 ckt.1
AFIERZ2-AFIERZ5 ckt.2
AELBS12-AELBS15 ckt.1
AELBS12-AELBS15 ckt.2
AELBS12-AELBS15 ckt.3
AKASHA2-AKASH25 ckt.1
AKASHA2-AKASH25 ckt.2
AFIER 2-AFIER 5 ckt.1
AFIER 2-AFIER 5 ckt.2
AFIER 2-AFIER 5 ckt.3
ARRAZH 1
ARRAZH 2
ATIRAN 3
UGLJEVIK
URECHESI
BUC.S-BUC.S-B ckt.1
BUC.S-BUC.S-B ckt.2
FUNDE2 1
FUNDEN 1
MINTIA-MINTIA B
IERNUT 1
JBGD8 1
JBGD3 1
JBGD3 2
JZREN2 2
JTKOSA 2
JTKOSA 3
JTKOSB 1
JTKOSB 2
JTKOSB 3
JPRIS4 1
JPRIS4 2
JJAGO4
60
60
90
90
90
100
100
120
90
90
100
100
120
300
400
400
400
200
200
400
400
400
200
150
150
150
150
400
400
400
150
150
300
255
84.9
173
86.6
386
96.4
168
84.1
169
84.4
167
126
124
83.3
83.9
82.3
141 118
116 129
110 123
80.3 80.3
80.3 80.3
6.4
Table 6.1.3 shows summary for elements loaded over 80% of their thermal limits. In network
planning practice, this is used as one of the main indicators which parts of the transmission network
represent weak spots and what network reinforcements are necessary in sense of generation and
supply, without taking into consideration the security aspects. Analysis of these results leads to
conclusion that transmission network is capable of transferring power from generation facilities to
major consumption areas, and network reinforcement is not necessary (without taking n-1 criterion
into consideration). But, in some substations, there is not enough transformer capacity to supply
demand from transmission network, since the load of these elements do not depend on generation
pattern and level of system interchange, but only on load level in these substations. This is problem
of local supply, and this should be studied as program of local not regional system development.
Increase of transformer installed capacity is needed in substations 220/110 kV Fier, Elbasan, Fierza
and Kashar in Albania, 400/110 kV Ugljevik in Bosnia and Herzegovina.
Also, it should be noted that level of reactive power consumption in load flow model of Albania is
very high comparing to the other systems and not realistic. This is one of the reasons why voltage
profile in Albanian network is very close to voltage collapse in some investigated scenarios.
Because of the problem with unrealistic reactive power demand new reactive power demand
forecast is needed.
Influence of voltage control equipment was not tested on regional level, and therefore it is not
possible to give answer weather there is possibility to better voltage profile in certain regimes. Also,
no light load regimes were analyzed, so there is no possibility to analyze effect of new reactive
power control devices.
6.1.2 Overview on possible solutions for system relief
In this subchapter, security aspects of network performance in analyzed scenarios are presented. As
it has been concluded in Chapter 4, in all investigated scenarios, some elements are highly loaded
even in case of full topology. Most of the elements loaded over 80% are transformers in some
substations and also there are some overloaded elements (Table 6.1.3).
Based on full topology and n-1 contingency analyses, some internal network reinforcements are
necessary to sustain analyzed load-demand level and production pattern. In other words, optimum
GTMax dispatch, according to analyzed generation and load scenarios, is not possible to achieve
with desired level of network security. Internal network reinforcements are necessary in order to
achieve such optimal dispatch.
Most of the identified critical network elements in Romania can be relieved by dispatching actions
(change of network topology or production units engagement). Re-dispatching actions, mostly in
the power system of Romania (power plants DEVA 1, LOTRU CIUNGET, PORTILE 1,
PAROSENI and ROVINARI) will be necessary to keep desired network security level according to
the (n-1) criterion. Also, installing a new transformer unit 400/110 kV in Brasov is planned and
helps resolving some critical states in Romanian network. Upgrading the Timisoara substation to
400 kV level and supplying this area from 400 kV network instead from 220 kV would resolve
some problems in Romanian network.
Reinforcement of local network in vicinity of Belgrade is necessary, but as it has been stated before,
this is local development. Problem of supplying Belgrade area must be part of more complex and
thorough analyses.
6.5
Central part of the 400 kV network in Serbia is heavy loaded due to large energy transits from
Romania, Bulgaria and south of Serbia (Kosovo and Metohija) to Croatia and Hungary.
Strengthening of the 400 kV corridor from Romania through Serbia to Croatia (East-West), by
building 400 kV lines Romania (e.g. new proposed substation Timisoara)-(Vrsac)-Drmno and
Belgrade-Obrenovac can relieve this problem.
For all of this, more thorough analyses are needed and only adequate feasibility studies can give the
proper answer to the question about the level and type of network reinforcements.
Final conclusions concerning year 2010 are:
- expected network topology in year 2010 is sufficient for all investigated generation and
load pattern for year 2010, except in South Serbia and Belgrade area;
- building of the 400 kV corridor Nis-Leskovac-Vranje-Skoplje before 2010 resolves all
insecure states identified in network of southern Serbia for investigated regimes in
2010;
- network reinforcement in Belgrade area is necessary;
- level of reactive power consumption in load flow model of Albania is very high
comparing to the other systems and not realistic and this causes some overloads due to
low voltage profile in Albanian network;
- high loads and overloads of elements in Romania can be relieved by operational
methods.
Expected network topology for 2015 generally improves network performance, especially in the
cases of Albanian and Serbian network. But, additional network reinforcements are necessary,
especially in increasing transformer capacities in mentioned substations, over which large
consumption areas are supplied. In case of Romania, it is hard to say what reinforcements are
necessary, since most of the identified insecure states can be relieved with production redispatch
and network topology changes.
New proposed lines (Visegrad – Pljevlja, Tumbri – Banja Luka and Pecs – Sombor) do not resolve
any of the identified insecure states, since these lines are electrically far from these critical network
elements, and therefore do not have significant influence. But, they do increase transfer capabilities
of interconnected network.
Final conclusions concerning year 2015 are:
- new lines commissioned to come into operation between 2010 and 2015 greatly reduce
losses in analyzed region, regardless of hydrology;
- expected network topology in year 2015 is sufficient for all investigated generation and
load pattern for year 2010, except in Belgrade area;
- building of the 400 kV corridor Nis-Leskovac-Vranje-Skoplje resolves all insecure
states identified in network of southern Serbia for investigated regimes in 2015;
- network reinforcement in east-west corridor in Serbia is necessary (this includes
Belgrade area also);
- level of reactive power consumption in load flow model of Albania is very high
comparing to the other systems and not realistic and this causes some overloads due to
low voltage profile in Albanian network;
- high loads and overloads of elements in Romania can be relieved by operational
methods;
- proposed interconnection lines candidates do not resolve any of the identified
problems.
6.6
6.2 Sensitivity cases
Sensitivity cases consist of six scenarios. In all of these cases average hydrology is used. Two
different scenarios are analyzed (first 1500 MW of import plus 500 MW of transit and second high
load growth) for years 2010 and 2015. Also, for each of these scenarios in year 2015 two cases are
analyzed (first for 2010 topology, i.e. without any new line or transformer planned for
commissioning between 2010 and 2015 and second for expected topology in year 2015).
6.2.1 Analyses comparison
Main focus of this subchapter is to analyze influence of high load growth rate and import/export
regime on system losses and overall network performance.
Table 6.2.1 shows overview of losses on regional level. It can be seen that new elements planned
for commissioning between 2010 and 2015 influence reduction of losses in analyzed region, no
matter of hydrology. This reduction of losses is between 30 and 40 MW.
Table 6.2.1: Peak Hour - Regional Losses (MW) under Average Hydrologic Conditions
Case
Reference Case
Imports/Exports
High Load Forecast
*
topogy
Year 2010
2010*
737.1
684.8
913.9
Year 2015
*
2010
1,035.1
NA
1,296.9
2015*
1,006.4
850.3
1,257.8
These changes are consequence of better voltage profile in analyzed region, especially in areas
directly influenced by the new elements. For example, new interconnection line Nis-LeskovacVranje-Skopje greatly improves voltage profile in southern part of Serbia and new interconnection
lines Podgorica – V.Dejes – Kashar – Elbasan and Kosovo B – V.Dejes – Kashar – Elbasan greatly
improve voltage profile in Albania. Higher voltages and reduction of load of some elements that are
highly influenced by the new elements, reduce power losses mainly in Albania, Southern Serbia,
and in the region as whole.
Table 6.2.2 shows overview of area summaries for analyzed reference cases and Table 6.2.3 shows
overview of high loaded elements in analyzed sensitivity cases.
Import/export sensitivity scenarios
Concerning the import/export case, the simulated regime means the following:
▪ Import 750 MW from UCTE
▪ Import 500 MW from Turkey
▪ Export 500 MW to Greece
▪ Import 750 MW from Ukraine
Large imports from analyzed directions cause significant increase of power flows along SlovenianCroatian, Ukrainian-Romanian and Bosnian-Montenegrin interfaces in 2010, but interconnection
lines are not jeopardised since they are loaded far below their thermal ratings.
6.7
Power losses on network topology in 2010, compared to the situation of balanced SE Europe power
system, are increased in the power systems of Bulgaria (7.9 %) and Montenegro (9 %). In other
power systems these losses are decreased, with the most significant drop in Romania (-16 %) and
Serbia and UNMIK (-10.1 %). Regional power losses are decreased (-7.1 %) when the situation
includes additional power import. It can be concluded that for this exchange scenario, power flows
and voltage profile are such that transmission network is better utilized, and as a consequence losses
are smaller.
By comparing the average hydrology situation in 2010 and balanced SE Europe power system to
the average hydrology situation and 1500 MW of power import, it may be noticed that some critical
contingencies in the Romanian power system disappear, especially those connected with Mintia
substation, while some new contingencies appear in the power system of Bulgaria (lines around
Maritsa East substation). This is due to different dispatching conditions of DEVA 1 power plant in
Romania (disconnected in import/export scenario, dispatched with 850 MW in the base case) and
power import through Turkish-Bulgarian interface that goes through Maritsa East 3 substation.
The analyzed scenario which is characterized by large power import in 2015, but on the 2010
network topology, can not be supported from a transmission network viewpoint due to voltage
problems. Their existence is the most obvious in the power system of Albania due to scarce reactive
power sources. To mitigate such problems it is necessary to construct at least one new 400 kV
interconnection line between Albania and UNMIK or Macedonia.
Large imports from analyzed directions in 2015 cause significant increase of the power flows along
the Slovenian-Croatian and Ukrainian-Romanian interfaces. However, the interconnection lines are
not jeopardized since their loading levels fall far below their thermal ratings.
Power losses, in comparison to the situation of balanced SE Europe power system, are increased
only in the power system of Macedonia (5%). In other power systems power losses are decreased,
the most significantly in the power systems of Romania (-26%) and Montenegro (-23%). Power
losses are decreased (-16%) in the region when analyzing the additional power import.
Certain reinforcements in the internal networks of Romania, Bulgaria, Albania and Serbia till 2015
are necessary shall analyzed generation pattern and 1500 MW of power import be made more
secure. None of the identified congestions is located at the border lines.
High load growth rate
Pessimistic (high) load growth rate is more close to load forecast made by electric power utilities in
analyzed region.
In case of high load growth rate power flows through lines and transformers are greater. This cause
greater level of losses, increased number of elements loaded over 80 % and overloaded elements
also and greatly decreased level of system security (Table 6.2.3).
In case of year 2015 and 2010 topology it is not feasible to realize this generation-load pattern due
to network constraints in Albania, but in Serbia and Romania also.
As it has been concluded in Chapter 5, in all investigated scenarios, some elements are highly
loaded even in case of full topology. Most of the elements loaded over 80% are transformers in
some substations and also there are some overloaded elements (Table 6.2.3).
6.8
Table 6.2.2: Overview of area summaries for analyzed sensitivity cases (in MW)
Year
Case
Topology
Data type
Generation
Load
2010
Losses
Interchange
Generation
Import/export
Load
2010
2010
Average hydrology
Losses
Interchange
Generation
High load
Load
2010
Average hydrology
Losses
Interchange
Generation
Load
2010
Losses
Reference case
Interchange
Average hydrology
Generation
Load
2015
Losses
Interchange
Generation
Load
2010
Losses
Import/export
Interchange
2015
Average hydrology
Generation
Load
2015
Losses
Interchange
Generation
Load
2010
Losses
High load
Interchange
Average hydrology
Generation
Load
2015
Losses
Interchange
* convergent load flow solution was not found
Reference case
Average hydrology
Albania
B&H
Bulgaria
Croatia
Macedonia
Montenegro
Romania
Serbia
896.7
1287.3
50.4
-441.0
898.2
1288.5
49.7
-440.0
931.7
1358.0
57.7
-484.0
1116.1
1531.0
81.2
-496.0
1118.1
1541.0
73.1
-496.0
*
*
*
*
1027.5
1544.0
71.5
-588.0
896.1
1680.0
99.0
-883.0
897.7
1684.0
96.7
-883.0
2266.1
1971.3
58.6
236.2
2398.3
1965.0
54.3
379.0
2202.8
2004
57.8
141.1
2316.6
2279.0
78.7
-41.1
2317.2
2279.0
79.2
-41.0
*
*
*
*
2394.9
2293.8
70.1
31.0
2479.4
2274.0
92.2
113.2
2479.7
2272.0
94.7
112.9
6900.4
5977.3
121.6
787.0
6827.0
5967.5
131.2
714.0
7282.5
6278.0
146.5
858.0
7332.7
6483.0
150.7
699.0
7332.9
6483.0
150.9
699.0
*
*
*
*
7582.2
6418.9
142.7
1006.0
8157.8
7083.0
175.8
899.0
8157.4
7083.0
175.4
899.0
1502.9
3136.7
49.0
-1682.8
1705.2
3143.3
44.9
-1483.0
2254.6
3295.0
55.4
-1095.9
2316.3
3657.0
63.4
-1404.1
2317.0
3657.0
64.0
-1404.0
*
*
*
*
2173.9
3660.1
58.8
-1545.0
2591.0
3959.0
79.8
-1447.8
2592.1
3959.0
81.1
-1448.1
950.3
1198.2
20.1
-268.0
965.5
1178.3
20.1
-233.0
991.1
1234.0
18.1
-261.0
1087.0
1407.0
20.0
-340.0
1088.4
1407.0
21.4
-340.0
*
*
*
*
1076.3
1393.7
22.5
-340.0
895.0
1489.0
23.0
-617.0
895.7
1489.0
23.7
-617.0
539.8
669.2
15.5
-147.0
542.1
670.2
16.9
-147.0
542.3
686.0
19.2
-163
735.0
671.0
25.0
39.0
735.2
676.0
20.3
39.0
*
*
*
*
731.9
675.0
15.6
39.0
1324.6
702.0
35.6
587.0
1322.6
708.0
27.6
587.0
7939.9
6728.3
201.2
930.4
6877.4
6733.1
169.1
-104.8
7113.3
6859.4
310.6
-56.7
7712.8
7317.4
347.4
47.9
7708.6
7317.4
343.1
48.1
*
*
*
*
7187.7
7831.2
253.9
-972.0
7684.5
8269.8
454.5
-1039.8
7676.4
8269.8
446.7
-1040.1
7355.9
6873.1
220.8
248.4
6591.7
6901.7
198.6
-521.9
8103.8
7131.0
248.6
724.2
8687.6
7263.0
268.7
1156.0
8689.4
7279.0
254.4
1156.0
*
*
*
*
8029.8
7270.7
215.1
530.0
10024.2
7635.0
337.0
2052.2
10023.9
7660.0
311.9
2051.9
Total
(SE Europe)
28351.9
27841.4
737.1
-336.7
26805.4
27847.6
684.8
-1836.7
29422.1
28845.4
913.9
-337.3
31304.1
30608.4
1035.1
-339.3
31306.8
30639.4
1006.4
-338.9
*
*
*
*
30204.2
31087.4
850.3
-1839.0
34052.6
33091.8
1296.9
-336.2
34045.5
33124.8
1257.8
-337.4
6.9
Table 6.2.3: Overview of high loaded elements in analyzed sensitivity cases
year
regime
topology
AREA
ALB
BUL
ROM
SRB
ELEMENT
HL
HL
HL
HL
HL
HL
HL
HL
HL
HL
HL
HL
HL
HL
HL
HL
HL
HL
HL
220kV AKASHA2-ARRAZH2 1
220 kV M.EAST-ST.ZAGORA
220kV BRADU-TIRGOVI 1
220kV BUC.S-B-FUNDENI 1
220kV FILESTI-BARBOSI 1
220kV L.SARAT-FILESTI 1
220kV LOTRU-SIBIU 1
220kV LOTRU-SIBIU 2
220kV P.D.F.A-CETATE1 1
220kV P.D.F.A-RESITA 1
220kV P.D.F.A-RESITA 2
220kV P.D.F.II-CETATE1 1
220kV PAROSEN-BARU M 1
220kV RESITA-TIMIS 1
220kV RESITA-TIMIS 2
220kV TG.JIU-PAROSEN 1
220kV URECHESI-TG.JIU 1
220 kV MINTIA-SIBIU
220kV JBGD3 21-JOBREN2 1
average
RATING
MVA
LOAD
MVA
%
270
228.6
302.6
320
277.4
277.4
277.4
277.4
208.1
277.4
277.4
277.4
277.4
277.4
277.4
208.1
277.4
381.1
301
2010
imp/exp
LOAD
MVA
high load
LOAD
%
MVA
Lines
%
average
2010
2015
LOAD
LOAD
%
%
MVA
MVA
250
189
92.6
238
88
82.6
227
227
278
278
81.7
81.7
134
100
2015
imp/exp
2010*
2015
LOAD
LOAD
%
%
MVA
MVA
high load
2010
2015
LOAD
LOAD
%
%
MVA
MVA
236
191
247
266
87.3
83.5
81.6
83.1
288
107
290
108
285
309
248
241
94.3
96.7
89.5
86.9
204
98.2
207
267
267
271
228
230
230
321
320
99.6
96.2
96.2
97.7
82
82.8
82.8
154
116
284
308
246
239
281
281
206
262
262
270
93.8
96.1
88.6
86.2
101
101
99.1
94.5
94.5
97.3
225
225
314
314
81.1
81.1
151
113
205
236
98.6
85.2
205
232
98.5
83.7
236
85.2
232
83.7
278
278
268
261
134
100
96.5
86.7
273
273
268
262
131
98.3
96.4
87
277
92.2
291
96.7
292
97
81.8
81.8
81.8
82.2
82.2
88.3
109
119
114
84.8
84.8
82.9
82.9
49.5
49.5
49.5
87.7
87.7
94.2
149
122
117
58
58
93.4
93.4
87.7
87.7
97.5
102
82.5
82.5
82.5
97.5
97.5
105
124
136
130
96.7
96.7
93.4
93.4
87.7
87.7
81.3
85.2
259
179
167
241
204
172
232
208
221
378
378
408
487
86.3
89.4
83.6
120
102
85.9
92.6
83.4
88.4
94.6
94.6
102
122
52.2
52.2
52.2
87.8
87.8
94.4
152
124
119
59.9
59.9
94.7
94.7
87.5
87.5
96.4
101
124
256
179
167
240
203
170
230
206
220
377
377
405
480
87
87
87
97.6
97.6
105
126
138
132
99.8
99.8
94.7
94.7
87.5
87.5
80.3
83.9
83
85.4
89.7
83.3
120
102
85.2
92.1
82.6
87.8
94.1
94.1
101
120
175
137
137
137
151
154
134
246
332
412
430
430
87.5
91.4
91.3
91.3
101
103
89.2
81.9
83
103
108
108
175
138
87.4
91.7
134
136
134
89
90.6
89.2
330
352
368
368
102
82.6
88
91.9
91.9
82
Transformers
ALB
BIH
CRO
ROM
SRB
MNG
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
220/110
220/110
220/110
220/110
220/110
220/110
220/110
220/110
220/110
220/110
220/110
220/110
220/110
220/110
220/110
220/110
220/110
220/110
400/110
220/110
220/110
220/110
220/110
220/110
400/110
400/110
400/110
400/220
400/220
400/220
400/220
400/220
220/110
220/110
220/110
220/110
220/110
220/110
220/110
400/110
400/220
400/220
400/220
400/220
220/110
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
kV
ABURRE 1
ABURRE 2
ABURRE 3
AELBS1 1
AELBS1 2
AELBS1 3
AFIER 1
AFIER 2
AFIER 3
AFIERZ 1
AFIERZ 2
AKASHA 1
AKASHA 2
ARRAZH 1
ARRAZH 2
ATIRAN 2
ATIRAN 3
MO-4 3
UGLJEV 1
TESISA 1
BARBOS 1
FUNDE2 1
FUNDEN 1
TIMIS 1
BRASOV 1
CLUJ E 1
DIRSTE 1
BUC.S 1
BUC.S 2
IERNUT 1
URECHE 1
MINTIA
JBGD3 1
JBGD3 2
JPRIS4 1
JPRIS4 2
JTKOSA 2
JTKOSA 3
JZREN2 2
JJAGO4 A
JBGD8 1
JTKOSB 1
JTKOSB 2
JTKOSB 3
JTPLJE 1
60
60
60
90
90
90
120
90
90
60
60
100
100
100
100
120
120
150
300
200
200
200
200
200
250
250
250
400
400
400
400
400
200
150
150
150
150
150
150
300
400
400
400
400
125
117 97.6 122
117 93.7 96
107 100
95.7 106 91.6 102 95.6
91.4 102
255
84.9
174
173
86.6
78
78
83.8
138
113
108
51.9
51.9
84.1
84.1
86.6
86.6
93.1
115
126
120
86.5
86.5
84.1
84.1
74.3
74.3
79.8
135
111
106
50.3
50.3
82.1
82.1
82.5
82.5
88.7
113
123
117
83.8
83.8
82.1
82.1
255
84.9
253
84.2
256
85.4
193
163
96.7
81.4
193
163
96.5
81.3
214
181
107
90.5
203
81
86.9
179
344
386
102
111
106
49.1
49.1
49.1
74
74
79.5
131
107
102
50.9
50.9
82.9
82.9
89.3
86.1
320
395
80.1
98.9
391
97.7
167
126
83.4
83.7
167
126
83.3
83.9
169
127
84.7
84.5
131
133
124
87.1
88.7
82.4
124
82.3
122
81
326
326
81.4
81.4
96.4
132
134
120
87.7
89.3
80.1
326
81.6
*convergent load flow solution was not found
6.10
As it has been said above, in some substations there is not enough transformer capacity to supply
demand from transmission network, since the load of these elements do not depend on generation
pattern and level of system interchange, but only on load level in these substations, particularly in
this case.
This is problem of local supply. It should be pointed out that load patterns (in all reference and
sensitivity cases) have the greatest influence to identified problems.
6.2.2 Overview on possible solutions for system relief
Security aspects of network performance in analyzed scenarios are presented in this subchapter.
It has to be said, that new generation facilities in UNMIK area, are too large to be investigated as
they were. Over 3000 MW of new production were connected to only one substation, Kosovo B. It
is almost certain, that installing such large capacities demands different and more detailed solution,
as well as large changes in internal network of UNMIK area, in order to supply this energy to
consumers. Also, large changes in internal network of Serbia are necessary too, since installing of
these 3000 MW causes great changes in energy flows in region, and therefore different supply
routes in internal network. That is why, more detailed analysis is necessary in order to draw
conclusions that will give right indices about eventual congestions and necessary network
reinforcements. One of possible solutions is building of 400 kV ring (Kosovo B – Pristina 4 –
Kosovo C) that will connect new production facilities with existing substations Kosovo B and new
substation 400 kV Pristina 4. The other is to connect the all new plants to existing 400 kV
substation Kosovo B radial through new 400 kV lines (as it was analyzed in this study). This
solution is better from network point of view, but from generation point of view is less reliable and
secure.
More thorough analyses is needed and only adequate feasibility studies can give the proper answer
to the question about the level and type of network reinforcements, especially in the case of the new
generation facilities proposed.
Concerning internal network reinforcements, same conclusions can be made as in previous
chapter.
Additional conclusions concerning year 2010 are:
- level of power losses for import/export scenario are significantly lower as consequence
of better utilization of the transmission network. It should be pointed out that this
scenario is more close to recorded system behavior, since Greece is one of the major
regional importers
- level of load of high loaded and overloaded elements is greater, especially in case of
high load scenario;
- building of the 400 kV corridor Nis-Leskovac-Vranje-Skoplje before 2010 resolves has
much greater significance.
Additional conclusions concerning year 2015 are:
- expected network topology in year 2015 is not sufficient in case of high load scenario;
- operational methods in network of Romania can relieve some of identified problems,
but in case of high load scenario network reinforcement should be analyzed also;
- new production units in UNMIK area require detailed analysis related to connection of
these units to power system network.
6.11
6.3 List of priorities
Based on main conclusions above lists of priorities are given in Table 6.3.1 and Table 6.3.2. It
should be pointed out that these elements are not ranked according to some criteria, but presented
according commissioning date and approved development plans from each responsible power
company in the region. Realization of all of these projects is necessary for secure and reliable
operation of regional transmission network (as described in previous subchapters).
Table 6.3.1: List of priorities until year 2010
Interconnection line
Ugljevik - S. Mitrovica
Kashar - Podgorica
C. Mogila - Stip
Florina - Bitola
Maritsa Istok - Filipi
Ernestinovo - Pecs (double)
(Filipi) - Kehros - Babaeski
Bekescaba - Nadab (Oradea)
Interconnected
countries
Year of
commissioning
BA - SER
AL - MN
BG - MK
GR - MK
BG - GR
HR - HU
GR - TR
HU - RO
2005/06
2006/07
2006/07
2006/07
2007/08
2007/08
2007/08
2008
Table 6.3.2: List of priorities between 2010 and 2015
Interconnection line
Zemlak - Bitola
Kashar (V. Dejes) - Kosovo B
(Nis) – (Leskovac) – Vranje - Skopje
Interconnected
countries
Year of
commissioning
AL - MK
AL – CS
CS – MK
2010/15
2010/15
2010/15
It also should be pointed out that local networks also need to be reinforced to satisfy security and
reliability criteria, but this is not problem of transmission network then local supply problem. In
case of new production units, especially in UNMIK area, detailed analysis related to connection of
these units to power system network is required.
In Table 6.3.3 and Table 6.3.4 are presented some recommendations that would make operation of
regional transmission network more reliable and secure.
Table 6.3.3 – Recommendation for new transformers in new substations
Length
Country
Line
Voltage
(km)
Romania
Arad-Timisoara
400 kV
55
Obrenovac-Belgrade ?-Pancevo
400 kV
80
Serbia
Drmno-Vrsac
400 kV
50
TPP Kosovo NEW-Pristina
400 kV
20
UNMIK
TPP Kosovo NEW-Kosovo B
400 kV
20
Interconnection
Timisoara (ROM)-Vrsac (SCG)
400 kV
80
6.12
Table 6.3.4 – Recommendation for new transformers in new substations
Country
Romania
Serbia
UNMIK
Name of
substation
Timisoara
Vrsac
Beograd ?
Pristina
Voltage
levels
kV/kV
400/110
400/110
400/110
400/110
New
transformers
MVA
2x300
2x300
2x300
2x300
It should be pointed out that these recommendations are not according to the long term development
plans of power utilities in the region, but just ideas of authors that can resolve some of the identified
problems in regional network. They should be taken into consideration in future studies and
analyses. Without detailed analyses it is not possible to rank these recommendations.
6.13
7. LITERATURE
[1] Study, Development of the interconnection of the electric power systems of
SECI member countries for better integration with European systems: Project
of regional transmission network planning, Construction of the regional model,
2002, EKC,
[2] Study, Development of the interconnection of the electric power systems of
SECI member countries for better integration with European systems: Project
of regional transmission network planning, effects of the construction of new
proposed interconnection lines in SECI countries, 2002, EKC, EIHP, NEK,
ZEKC
[3] Project group on development of interconnection of electric power systems of
Black sea region, Regional Transmission Planning Project Regional model
construction for 2010. year, EKC
[4] Study, Audit of overcurrent protection relay settings on north-south
transmission corridor, HTSO-Hellenic Transmission System Operator, EKC,
2004
[5] Report of UCTE Executive Team "North-South Re-Synchronization", Loadflow analyses, Technical committee UCTE, EKC, MVM, ZEKC, 2004
[6] Report of UCTE Executive Team "North-South Re-Synchronization", Loadflow analyses, Technical committee UCTE, EKC, MVM, ZEKC, 2004
[7] SECI project group on development of interconnection of electric power
systems of SECI countries for better integration to the European system,
Regional Transmission Planning Project, Regional model construction for
2005. year(updating and expanding existing Regional Transmission Network
Model for the year 2005 ) and 2010. year, USAID, SECI
[8] Study, Technical and economical aspects of connection electric power systems
between Serbia and Macedonia with new transmission line 400 kV Niš(Leskovac-Vranje)-Skopje, Feasibility study, EPS-Serbia, EKC, 2003
[9] Study for new 400 kV interconnection lines between FYROM-Serbia and
Albania-Montenegro, Transient Security Assessment, Total Transmission
Capacities (TTC’s), European Commission, TREN Energy Directorate
General, EKC, HTSO,EPS,ESM,KESH,EPCG, 2003
[10] The study of long-term development of 400 kV, 220 kV and 110 kV
transmission networks in the area of the Republic of Serbia for the period until
2020, 1997, EINT
[11] Analysis of technical possibilities for the reconnection of the south-eastern
with the western part of UCPTE Interconnection, Institute Nikola Tesla,
Beograd, 1996
[12] Stability of the Synchronously Interconnected Operation of the Electricity
Networks of UCTE/CENTREL, Bulgaria and Romania, Final Report 2000
[13] Research Project, Evaluation Of Transfer Capabilities In Balkan
Interconnected Network, Technical Report, Joint Greek-Yugoslav Research,
General Secretariat for Research and Technology, Greece, EKC, HTSOGREECE, NTUA-GREECE, 2004
[14] Study for new 400 kV interconnection lines between FYROM-Serbia and
Albania-Montenegro Energy Directorate General, HSSO, EKC 2003
[15] Report of UCTE Executive Team "North-South Re-Synchronization": Loadflow analyses, Technical committee UCTE, MVM, ZEKC 2003
[16] Bosnia and Herzegovina and Republic of Macedonia Study on Gas and
Electricity Interconnection in the Trans European Networks, 1997
[17] Feasibility Study, 400 kV Transmission line Elbasan-Podgorica, FICHTNER,
July 2001
[18] Study, Upgrading of Transmission Lines 220 kV, ALSTOM 2001
[19] Feasibility Study, Rehabilitation of transmission network, DECON,BEA, EINT
2002
[20] Reconnection of the UCPTE Network and Parallel operation of the Bulgarian
and Romanian networks with UCPTE - SUDEL ad hoc Group 1996
[21] Technical feasibility study of interconnection of the electric power systems of
Bulgaria (NEK) and Romania (RENEL) with the interconnected power system
of Greece (PPC), power systems under EKC coordination and Albania
(KESH), for parallel and synchronous operation in compliance with UCPTE
regulations and standards - PPC - RENEL - NEK - EKC - Nikola Tesla
Institute 1995
[22] Technical feasibility study of upgrading at 400 kV of the existing 150 kV line
Bitola - Amyndaio, PPC, NEK, EKC, 1998
[23] Technical feasibility study of a new interconnection tie-line at 400 kV between
Greece and Bulgaria, PPC, NEK, EKC, 1998
[24] Technical feasibility study of Interconnection of the Balkan Countries
(Albania, Bulgaria and Romania) to UCPTE - PHARE A 1994 and PHARE B
1996
[25] Project of the construction of 400 kV OHL Niš-Skopje, Elektroistok 2001
[26] 400 kV Interconnection Macedonia – Bulgaria, 400 kV Overhead Transmission
Line and 400/110 kV Stip Substation, Environmental Impact Assessment
(EIA)
[27] 400 kV Interconnection Macedonia – Bulgaria, 400 kV Overhead Transmission
Line and 400/110 kV Stip Substation, Techno-Economical Aspects
[28] Elaboration “Perspectives of one part of 220 kV network, Phase I, EINT
Beograd, 2003.
[29] Annual report of EKC for 1999, 2000, 2001, 2002 and 2003
[30] Annual report of the Electric Power Utility of Serbia, 2000, 2001, 2002, 2003
[31] Annual report of the Electric Power Utility of FYR of Macedonia, 2001, 2002,
2003,
[32] Annual report of the Transelectrica, 2001, 2002, 2003,
[33] Annual report of the NEK-Bulgaria, 2001, 2002, 2003,
Download