ZESCO POWER SYSTEM STABILITY STUDIES

FOR THE YEAR 2011

Master of Science thesis in the Programme of Electric Power Engineering

Chrispin Singoyi
Division of Electric Power Engineering
Department of Environment and Energy
CHALMERS UNIVERSITY OF TECHNOLOGY
Gteborg, Sweden, November 2011
i

ZESCO POWER SYSTEM STABILITY STUDIES FOR THE YEAR 2011

Chrispin Singoyi
In partial fulfillment for the award of Master of Science degree in Electric Power
Engineering, in the Department of Environment and Energy, Division of Electric Power
Engineering, Chalmers University of Technology, Goteborg, Sweden

Chrispin Singoyi, November 2011.

Supervisor:

Kennedy Mwanza
ZESCO ltd
Engineering Development Directorate
Lusaka-Zambia

Examiner/Supervisor: Dr. Tuan A. Le

Division of Electric Power Engineering
Chalmers University of Technology
SE-412 96 Gteborg
Sweden
Telephone + 46 31 772 3832

ii

Abstract
The purpose of this work is to look at the short term stability studies on 330 kV, 220 kV
and 88 kV transmission system of the whole ZESCO Power System in terms of voltage
and rotor angle stability after being exposed to severe system disturbances. The work also
considers the network expansions up to 2011. The initial conditions are considered
critically for better analysis. As one of the initial conditions to do the stability studies, all
system variables are within acceptable limits and the stability analysis is done at both off
peak and peak loads for static simulations and at peak loads only for transient simulations.
The assumptions considered, on the other hand on this study, are that the Zambian system
is connected via two 330 kV circuits to the Zimbabwe system to the south and to the 220
kV Congo DR system to the North but no power exchange between these two systems is
considered. Studies in steady-state and transient states have been carried out with an
expected peak load demand in 2011 amounting to 1670 MW (the losses on the
transmission system are not included). An additional study in stead state has been carried
out with an expected off-peak load demand in 2011 amounting to 1418 MW (the losses on
the transmission system are not included). In any case the generation is to match the load
demand values plus power system loses.
The methodology taken was to carry out Simulations in stead state to calculate the power
flows over the transmission system, the voltage profile and the active and reactive
generation of the generators. N-1 Contingency situations were examined. Whatever the
situation, the Zambian system must be able to satisfy the load demand and operate within
the acceptable limits. Simulations were also performed in transient state to verify that the
System was able to recover un acceptable situation without loss of synchronism of any
generator or sustained oscillations.The most severe faults being three phase faults near
generating stations and major substations on various system components and analyse the
results for a specific short period (10 seconds). The faults are cleared within 100 ms. Also
considered in the simulations is the taking out of some major loads and the analysis of the
effect.
The major findings are that in the stead studies the system is able to withstand any N-1
contingency without exceeding limits. And in the transient state studies the overall
behaviour of the system is not beyond limits. Oscillations are dumped within the expected
period of 4 to 10 seconds. All generators operate in synchronism.

Key words
Rotor angle stability, transient stability, inter-area oscillations, spinning or primary reserve.

iii

Acknowledgement.
Firstly I would like to give thanks to Dr. Tuan A. Le and the course coordinator, Ola
Carlson to give me the opportunity to do the thesis on our ZESCO power system.
My second gratitude goes to my classmate, Mr. Gustavo who guided me through the
process of learning the Power Factory soft ware by Dig Silent that I used on my stability
studies. There were no conditions attached, he was really helpful with his great knowledge
in Power Factory software.
Many thanks go to all my lecturers and classmates who shared the knowledge with me
during my Masters studies at Chalmers University of Technology. Class mates like Mr.
Thinley, Kalid T., Selamu T. to mention but a few, gave me un conditional guidance in
most of my Study work.
Not forgetting my workmates, from the department of Engineering Development, who
guided me in the simulations of various stability scenarios, Mr. Kennedy Mwanza, as my
supervisor who has great knowledge as well in Power Factory software, Mr. Julius
Kampamba, Mr. C. Nthala, Mr. G. Musonda and Mr. A. Sikombe. They gave me a
conducive working atmosphere. I worked freely with them.
The accomplishment of this project wouldnt have been made possible without the data
that I collected from various power stations and major substations. Therefore my sincere
gratitude goes to the ZESCO Engineers and Technicians who helped me gather all the
necessary data I needed.

iv

To My Mother Silika Nakamba Singoyi

Contents
LIST OF FIGURES ............................................................................................................ viii
LIST OF TABLES ...............................................................................................................xii
Chapter 1: Introduction .......................................................................................................... 1
Chapter 2: Overview of the Zambian Power network ............................................................ 3
2.1

ZESCO System ............................................................................................................ 3

2.1.1 Kafue Gorge Power Station .....................................................................................3

2.1.2

Kariba North Bank Power Station (KNBPS) ........................................................3

2.1.3 Victoria Falls Power Station ....................................................................................3

2.1.4 Small Hydropower Stations .....................................................................................4
2.1.5 Diesel Power Stations ................................................................................................4
2.1.6 Transmission System ...............................................................................................4
2.1.7 Distribution and Supply ...........................................................................................4
2.2 Lunsemfwa Hydro Power Company .............................................................................. 4
2.3 Copper belt Energy Corporation .................................................................................... 4
2.4 System Operation ........................................................................................................... 4
2.5 Regional Interconnections .............................................................................................. 5
Chapter 3: Basic concepts on Power System Stability. .......................................................... 7
3.1 Classification of Power System Stability. ........................................................................ 7
3.2 An Elementary View of Transient Stability ..................................................................... 9
3.2.1 Loss of one Parallel Line ...........................................................................................9
3.2.2 Equal Area Criterion ................................................................................................11
3.2.3 Fault at Either end of the Line .................................................................................13
3.2.4 Fault in the Middle of the Line ................................................................................14
3.2.5 Auto-reclosing on to Either Transient or Permanent Fault......................................17
Chapter 4: Power System Modeling..................................................................................... 19
4.1 Voltage Regulations ..................................................................................................... 19
4.2 Load Modeling ............................................................................................................... 20
4.2.1 General Description .................................................................................................20
4.2.2 Load Flow Analysis Modeling. ...............................................................................21
4.2.3 Transient Stability Simulations Modeling. ..............................................................22
Chapter Five: Simulations ................................................................................................... 25
5.1 Planning Criteria ............................................................................................................ 25
5.2 Methodology ................................................................................................................ 26
5.3 Analysis in Stead State ................................................................................................... 26
vi

5.3.1 Presentation .............................................................................................................26

5.4

Load flow Calculations at Peak Period ...................................................................... 27

5.4.1 Analysis in Normal Conditions .............................................................................27

5.5 Analysis in N-1 Conditions at Peak Load ...................................................................... 31
5.5.1 Tripping of one of the two 330 kV circuits between Kariba North and
Leopards Hill ....................................................................................................................31
5.5.2 Tripping of one of the three 330 kV Circuits between Leopards Hill and
Kabwe ...............................................................................................................................33
5.5.3 Tripping of one of the two 330 kV Circuits between Kabwe and Luano ................34
5.5.4 Tripping of one of the two 330 kV circuits between Kabwe and Kitwe .................34
5.5.5 Tripping of Kabwe-Pensulo Line ...........................................................................34
5.6 Analysis at off peak for both normal and N-1 contingencies. ........................................ 34
5.7 Analysis in transient state ............................................................................................... 34
5.7.1 Transient State Presentation ....................................................................................35
5.7.2 Hypotheses...............................................................................................................35
5.7.3. Simulated Faults .....................................................................................................36
Chapter 6: Conclusions and Recommendations ................................................................... 54
6.1 Conclusions .................................................................................................................... 54
6.2 Recommendations .......................................................................................................... 55
6.3 Future Work ................................................................................................................... 55
REFERENCES ..................................................................................................................... 56
Appendix 1: System upgrades .............................................................................................. 58
1. ZESCO System Upgrades ................................................................................................ 58
1.2 Completed generation and transmission system works. .............................................58
Appendix2: Identification of relevant sub-systems of the network .................................... 59
Appendix 3: ZESCO protection philosophy ........................................................................ 60
3. ZESCO transmission protection philosophy ................................................................... 60
3.1 Old Protection Philosophy ............................................................................................. 60
Backup Protection: Electro-mechanical Over current and Earth Fault relay with
IDMT Characteristics. ......................................................................................................60
3.2 New Protection System Philosophy ............................................................................... 60
Appendix4: Previous major faults ....................................................................................... 62
Appendix6: Unit Commitment ............................................................................................. 65
Appendix 7: Reactive power compensation and SVC Models ............................................ 66
Appendix8: Loading on the lines at peak load ..................................................................... 67
Appendix 9: Load flow calculations on N-1 ........................................................................ 73
vii

Appendix 10. Detailed parameters of the AVRs .................................................................. 87

Appendix 11. Detailed parameters of speed Governors. ...................................................... 88
Appendix 12. Transient Simulations and results .................................................................. 89

LIST OF FIGURES
Figure 1: The Zambian Power System ........................................................................................ 6
Figure 2: Classification of power system stability. ..................................................................... 8
Figure 3: Generator connected to infinite bus bar through two parallel lines. ............................ 9
Figure 4: Electrical representation. ........................................................................................... 10
Figure 5: Equal area criterion for the loss of one parallel line .................................................. 11
Figure 6: Short circuit at the end of the line .............................................................................. 13
Figure 7: Power angle relationship after a fault at either end of the parallel line ..................... 14
Figure 8: Fault in the middle of the line. ................................................................................... 15
Figure 9: Circuit model for the fault in the middle of the line .................................................. 15
Figure 10: Star delta transformation.......................................................................................... 15
Figure 11: Final circuit representation where only
is considered as the transfer
reactance .................................................................................................................................... 16
Figure 12: Equal Area criterion for the fault in the middle of the line...................................... 16
Figure 13: Critical clearing angle .............................................................................................. 17
Figure 14: Auto reclosing to either transient or persistent fault ................................................ 17
Figure 15: AVR Model IEEET1A ........................................................................................... 19
Figure 16: AVR Model SCRX/SEXS ...................................................................................... 19
Figure 17: Speed Governor Model IEEEG3 ............................................................................ 20
Figure 18: Power Factory general load model .......................................................................... 21
Figure 19: Load flow calculation window indicating the voltage dependence terms for load
flow calculations in Power Factory software. ........................................................................... 22
Figure 20: Diagram indicating the mixture of static and dynamic load model used for
stability studies. ......................................................................................................................... 22
Figure 21: Model used to approximate the behavior of the linear dynamic load. ..................... 23
Figure 22: Model used to approximate the behavior of the non-linear dynamic load. ............. 23
Figure 23: Low/High voltage approximations used in the non-linear dynamic load model. .... 24
Figure 24: Calculated load flow at peak period ....................................................................... 28
Figure 25: Copper belt load ...................................................................................................... 29
Figure 26: voltage profile ......................................................................................................... 30
Figure 27: Voltage profile ......................................................................................................... 31
Figure 28. Voltage profile with Kariba N-LHill line out .......................................................... 33
viii

Figure 29: Three phase to ground fault on KFG-LHill line ...................................................... 37

Figure 30: From top to bottom: Excitation voltage, Excitation Current, Machine reactive
power, Voltage dip and Rotor angle variations before, during and after the fault close to
KFG HPP................................................................................................................................... 37
Figure 31: Power swings on the four lines from Main Generating stations to Lhill
Substation .................................................................................................................................. 38
Figure 32: Frequency deviation on major Hydro Power Plants ................................................ 38
Figure 33: Voltage dip on major Load centers on the Copper belt after a fault close to
KFG-HPP .................................................................................................................................. 39
Figure 34: Voltage and rotor angle oscillations for a 300ms fault on KFG-Lhill line .............. 39
Figure 35: Auto reclosing of the KFG-Lhill line on transient fault .......................................... 40
Figure 36: From top to bottom, power swings on main generating stations, Voltage dips on
main Load bus bars and Frequency deviations after auto reclosing the KFG-LHill line. ......... 41
Figure 37: Auto reclosing on persistent fault and locking out ................................................. 42
Figure 38. Auto reclosing on persistent fault ............................................................................ 42
Figure 39: Auto reclosing on persistent fault- power swing on the copper belt lines, voltage
profile and machine frequency on major generating stations. .................................................. 43
Figure 40: Excitation voltage and current and reactive power on all machines on major
generating stations ..................................................................................................................... 44
Figure 41: Power swing on the four transmission lines to LHill substation from Kariba and
Kafue Gorge HPP. ..................................................................................................................... 44
Figure 42. Effect of high speed auto reclosing (Rotor angle, Voltage profile and rotor angle
difference) ................................................................................................................................. 45
Figure 43: Effect of high speed auto reclosing (Rotor angle difference, Voltage angle &
speed deviation)......................................................................................................................... 46
Figure 44: Effect of high speed auto reclosing (Active Power, Voltage profile &
Frequency) ................................................................................................................................. 47
Figure 45: Effect of high speed auto reclosing (Excitation voltage, Excitation current,
reactive power and Active power flow) .................................................................................... 48
Figure 46: From top to bottom: Rotor angle for KFG HPP machines, Voltage magnitude
and the rotor angle for KNBC HPP for a bus bar fault at LHill Substation. ............................ 49
Figure 47: Rotor angle, voltage angle and speed deviation on all major HPP machines for a
bus bar fault at LHill substation ................................................................................................ 50
Figure 48: Active power on the transmission lines to the North, Voltage profile on the
Copper belt Major substations and the frequency on all HPP generators for a bus bar fault
at LHill substation ..................................................................................................................... 51
Figure 49: Rotor angle and voltage profile for the fault at Leopards hill with the staggered
circuits for a bus bar fault. ......................................................................................................... 52
Figure 50: Active power flow to the north, voltage profile at the copper belt and the
generating frequency for a bus bar fault at LHill ...................................................................... 53
Figure 51: Relevant interconnected power sub-systems ........................................................... 59
ix

Figure 52: Single line diagram of the Zambia Power System ................................................... 64
Figure 53: Model3 of SVC and its parameters .......................................................................... 66
Figure 54: Simulated fault on KFG-Lhill Line close to KFG HPP ........................................... 89
Figure 56: Voltage magnitude and rotor angle for a fault close to KFG HPP .......................... 90
Figure 57: Close-up fault to Kafue gorge bus bar on KFG-LHill line ..................................... 91
Figure 58: Close-up fault to Kafue Gorge bus bar on KFG-LHill line .................................... 92
Figure 59: Auto reclosing on KFG-LHill line ........................................................................... 93
Figure 60: Auto reclosing on KFG-Lhill line............................................................................ 94
Figure 61: Auto reclosing on KFG-LHill line ........................................................................... 95
Figure 62: Clearing time at 300ms for a close up fault to KFG HPP on KFG-LHILL Line (
Voltage magnitude and angle difference) ................................................................................. 96
Figure 63: Clearing time at 300ms (Rotor angle, Voltage angle) for a KFG HPP close up
fault............................................................................................................................................ 97
Figure 64: Clearing time at 300ms for a close up fault at KFG HPP ( Active power,Voltage
magnitude and frequency) ......................................................................................................... 98
Figure 65: Close-up fault to Kariba bus bar on KNB-LHIll line .............................................. 99
Figure 66: Close-up fault to Kariba bus bar on KNB-LHIll line ............................................ 100
Figure 67: Close-up fault to Kariba bus bar on KNB-LHIll line ............................................ 101
Figure 68: Auto reclosing at Kariba end then at LHill end ..................................................... 102
Figure 69: Auto reclosing at Kariba end then at LHill end ..................................................... 103
Figure 70: Auto reclosing Kariba LHill line ........................................................................... 104
Figure 71: Fault on LHill-Kabwe line ..................................................................................... 105
Figure 72: Fault on LHill Kabwe line ................................................................................... 106
Figure 73: Fault on LHill-Kabwe line ..................................................................................... 107
Figure 74: LHill-Kabwe fault 400ms clearing time. ............................................................... 108
Figure 75: LHill-Kabwe fault and auto reclosing ................................................................... 109
Figure 76: LHill Kabwe fault and auto reclosing .................................................................... 110
Figure 77: Kabwe Kitwe line fault ........................................................................................ 111
Figure 78: Kabwe-Kitwe line fault.......................................................................................... 112
Figure 79: Kabwe-Kitwe line fault.......................................................................................... 113
Figure 80: Kabwe-Kitwe auto reclosing ................................................................................. 114
Figure 81: Kabwe-kitwe auto reclosing .................................................................................. 115
Figure 82: Kabwe Kitwe line auto reclosing ......................................................................... 116
Figure 83: Fault on Kabwe pensulo line ................................................................................. 117
Figure 84: Kabwe-pensulo line fault ....................................................................................... 118
Figure 85: Kabwe-pensulo line fault ....................................................................................... 119
x

Figure 86: KFG-Kfue west fault (close to Kafue west) .......................................................... 120
Figure 87: KFG-Kfwest line fault ........................................................................................... 121
Figure 88: KFG-KFWest line fault ......................................................................................... 122
Figure 89: KFG-KFWest auto reclosing ................................................................................. 123
Figure 90: KFG-KFWEST line auto reclosing ....................................................................... 124
Figure 91: KFG-KFWest line auto reclosing .......................................................................... 125
Figure 92: Fault on Luano-Kansanshi line .............................................................................. 126
Figure 93: Fault on Luano-Kansahi line ................................................................................. 127
Figure 94: Fault on Luano Kansanshi line .............................................................................. 128
Figure 95: Reactive Power, Excitation Voltage and current of the Generating Machines for
the fault close to KNBC HPP .................................................................................................. 128
Figure 96: Reactive power, excitation voltage and current of the machines for a fault close
to KFG HHP ............................................................................................................................ 129
Figure 97: Power swing for a fault close to KFG HHP .......................................................... 129
Figure 98: Power Swing for a fault close to KNBC HPP ....................................................... 130
Figure 99: Excitation voltage after a fault close to Vic Falls .................................................. 130
Figure 100: Power swing for a fault close to Vic. falls on Kariba HPP Lhill Line and
KFG HPP - LHill Line ............................................................................................................ 131

xi

LIST OF TABLES
Table 1: Loss of supply in the past 20 years. .............................Error! Bookmark not defined.
Table 2: Unit commitment ........................................................................................................ 65
Table 3: Reactive power compensation and SVC Models ........................................................ 66
Table 4: Loading on the lines at peak load ................................................................................ 72
Table 5: Load flow with Kabwe-Kitwe line out at peak load ................................................... 72
Table 6: Load flow with Kariba north and LHill line out (Load reduced to account for
increased loses). ........................................................................................................................ 73
Table 7: Voltage profile with Kariba North- LHill line out. ..................................................... 75
Table 8: Load flow with KFG-LHil line out. ............................................................................ 76
Table 9: Voltage profile with KFG-LHil line out ..................................................................... 77
Table 10: Load flow with KFG-Kafue west line out. ............................................................... 78
Table 11: Load flow with Lhill-Kabwe line out (load reduced to 98% to account for
increased transmission loses). ................................................................................................... 80
Table 12: Load flow calculations with Kabwe- Kitwe line out (load reduced to 93% to
account for increased losses). .................................................................................................... 82
Table 13: Load flow calculation with Kabwe-Luano line out. ................................................. 84
Table 14: Detailed parameters of the AVRs ............................................................................. 87
Table 15: Detailed parameters of speed Governors. ................................................................ 88

xii

Chapter 1: Introduction
ZESCO power system has in the past twenty years under gone several of system
disturbances and the latest being early last year [1]. The most common disturbances are the
ones to do with rotor angle and voltage instability following a severe system fault near
generating stations or on lines to major load centres. It is in this regard that there is need to
do some system stability studies to find a better solution to such disturbances. Therefore
the stability studies should be based on 330 kV, 220 kV and 88 kV transmission system up
to 2011. The highest voltage level is 330 kV in Zambia.
The Power system stability studies on the Zambian Power network are based on the fact
that by 2011 the Power demand will have increased from 1500 MW to about 1670 [1].
Generation capacity also will have been up rated by 11%. Therefore the network studies
are performed with the view that there will be this kind of Generation and demand by 2011
and hence the need to perform stability studies to ascertain the stability of the system.
Some severe situations have been examined to ascertain the stability of the Zambian
network. The critical period for the study is the annual peak load demand. Some load flow
simulations have also been done on 80% of the full peak load demand of the Zambian
network to represent the off peak load demand. The unity commitment of the peak load is
characterised as follows; all the six generating units at Kafue Gorge operate at almost full
capacity seconded by the four units at Kariba North Bank Power station. The rest operate at
almost full capacity as well.
All power plants are modelled as Hydro power plants (HPP) in Dig Silent software. This
study reports the static and dynamic behavior of the Zambian power system with the view
of the increased generation and load demand by 2011. For the transient study, the
electromechanical transients are considered, few seconds after the disturbance occurrence,
in order to detect a potential loss of synchronism on groups.
The description of the power system is detailed enough to simulate the electromechanical
transients. The study is focused on the behavior of the generators in terms of rotor angle
without any power exchange with the Zimbabwe or Congo DR system.
Currently the Zambian network is operated with the spinning reserve of 70 MW since it is
always interconnected with ZESA to the south and SNEL of Congo to the North. But this
spinning reserve is not enough if the Zambian system were to stand on its own. The
recommended spinning reserve in this situation is 180 MW representing the capacity of one
of the largest machines [1].
This report has the following details;
A brief discussion on the Zambian network is highlighted to give a wide
understanding of the network.

The basic concepts on Power system stability is illustrated in chapter three where
Equal Area Criterion is explained on how it helps to give the critical clearing angle or
the critical clearing time and their importance.
The methodology used to perform the Power System Stability study
The different steps of the proposed work

Stead state analysis covers the load flow and contingencies

Transient state analysis which covers significant defaults to check the behavior of the
rotor angle of all the interconnected machines to ascertain synchronism.
The results of the study that consists of permanent state and transient state analysis

Chapter 2: Overview of the Zambian

Power network
This chapter gives the overview of the Zambian power system with the consideration of the
neighboring country interconnections.
The Zambian electrical power system is operated as part of an interconnected power
system linking South Africa and Zimbabwe to the south and the Democratic Republic of
Congo (DRC) to the north, through tie-lines as shown in Figure 1.
The Zambian Electricity Supply Industry (ESI) presently consists of three major market
players namely ZESCO, CEC and LHPC involved in generation, transmission, distribution
and supply. System operation is coordinated by the ZESCO National Control Center in
conjunction with the various control centers.

2.1

ZESCO System

ZESCO owns and operates most of the power stations, transmission system, the
distribution system and the National Control Center.
ZESCO produces almost all of its electric power from hydropower stations which account
st

for 1668 MW, as at 31 May 2010 [1]. The remainder, about 1.6 MW is produced from
isolated diesel power stations.
2.1.1 Kafue Gorge Power Station
This is the largest Power Station in Zambia. It is located on the Kafue River and has been
in operation since 1971. The Station has an installed capacity of 990 MW with 6 generators
of 165MW each. Two generators feed one transformer consisting of three single-phase
generators. There are therefore ten single-phase generators, including a spare generator.
2.1.2

Kariba North Bank Power Station (KNBPS)

KNBC is the second largest power station after Kafue Gorge. It has a total capacity of 720
MW consisting of four generators of 180 MW each. Each generator feeds one three phase
transformer. The power station is located on the north bank of Kariba dam on the Zambezi
River. It has been in operation since 1977.Originary it had 4 by 150 generators but there
has been an uprating exercise on all the four generators to 180 MW. Currently there are
two more 180 MW generators being constructed to be completed in 2012. To the south of
this power station is another power station of 600 MW belonging to ZESA of Zimbabwe.
The two power stations are interconnected by two by 330 kV transmission lines of 1.5 km
long.
2.1.3 Victoria Falls Power Station
The station is a run-of-river station located on the Zambezi River at Victoria Falls. It has an
installed capacity of 108 MW and comprises three power stations namely stations A, B and
C. Station A was commissioned in 1936. It has an installed capacity of 8 MW consisting of
two generators of 1 MW each and another two of 3 MW each. Station B was commissioned
in 1968 and has an installed capacity of 60 MW consisting of 6 generators of 10 MW each.

Station C was commissioned in 1968 and has an installed capacity of 40 MW consisting of

4 generators of 10 MW each.
2.1.4 Small Hydropower Stations
There are four (4) Small hydro power stations with a combined capacity of 23.75 MW.
These are Lusiwasi (12 MW) in Central province, Musonda Falls (5 MW) in Luapula
province, Chishimba Falls (6 MW) and Lunzua (750 kW) in Northern Province.
2.1.5 Diesel Power Stations
ZESCO operates several small Diesel Power Stations throughout the country. The stations
are at Lukulu, Luangwa, Chama, Zambezi, Mufumbwe, Mwinilunga, Chavuma, Kabompo
and Kaputa. [1]
2.1.6 Transmission System
The transmission system has a total of 4,638 km of transmission lines spread as follows:
2,008 km of 330 kV lines, 348 km of 220 kV, 85 km of 132 kV, 704 km of 88 kV, and
2,823 km of 66 kV lines. The total transformer installed capacity is about 3,000 MVA. [1]
Transmission systems transports power from generating stations to all bulk distributions
end points, the ZESCO major customer, Copper belt Energy Corporation on the Copper
belt and Imports or Exports to neighboring counties
2.1.7 Distribution and Supply
The distribution system comprises the main network from the national grid and isolated
networks fed from stand-alone diesel power stations. Zambia imports power at distribution
voltage level from ESCOM (Malawi) to Lundazi District and ZESA (Zimbabwe) to supply
Siavonga District. [1]

2.2 Lunsemfwa Hydro Power Company

Lunsemfwa Hydro Power Company owns and operates two (2) hydropower stations
namely Lunsemfwa and Mulungushi power stations with installed capacities of 18 MW
and 20 MW respectively. These stations are connected to the ZESCO power network.

2.3 Copper belt Energy Corporation

The Copper belt Energy Corporation owns and operates part of the transmission and
distribution infrastructure on the Copper belt. The infrastructure includes 208 km of 220
kV lines, 590 km of 66 kV lines, 3 bulk purchase points, 5 by 220 kV substations, 28 by 66
kV substations, and 1,275 MVA of 220/66 kV transformer capacity and 1,750 MVA of
66/33/11 kV transformer capacity. [2]
CEC also owns and operates Emergency Gas Turbine Alternators (GTAs) with an installed
capacity of 80 MW. They also own the 40 km long 220 kV tie line to the DRC.

2.4 System Operation

The National Control Center (NCC) coordinates all the operations in the system, which
includes transmission switching, generation dispatch and exports. The Control Center uses
a SCADA (Supervisory, Control and Data Acquisition) system, which was commissioned

in June 1996 and upgraded in 2005. The SCADA system covers 24 transmission stations,
which include four (4) Power Stations and twenty (20) Substations.
CEC operates a regional control center for its system on the Copper belt. ZESCO and
Lunsemfwa Hydro Power Company also operate control rooms at their stations.
System operation is the function of maintaining the power system in a stable operating
condition. This involves:
a. Monitoring changes in demand and scheduling and dispatch of generating plant in
accordance with expected requirements.
b. Monitoring and taking corrective action to maintain frequency and voltage in the system.
c. Monitoring transmission lines, to ensure that the electrical and thermal stability limits are
not exceeded.
d. Ensuring an acceptable level of system security.
e. Being prepared to take appropriate action in the event of a failure of any equipment that
may jeopardize the security position.
f. Having a contingency plan to recover the system in the event of a complete system
failure i.e. a black-start capability.

2.5 Regional Interconnections

The Zambia power system is connected to the regional interconnected system via two 330
kV transmission tie-lines to Zimbabwe to the south, and one 220 kV line to Congo power
system to the north as shown in Figure 1.

Figure 1: The Zambian Power System

The regional interconnected system is operated under the auspices of the Southern African
Power Pool (SAPP), which falls under SADC. There are twelve (12) member utilities of
SAPP of which nine are interconnected, including ZESCO. The SAPP objectives are to
coordinate the planning and operation of member systems and to promote energy trading
among members in the region. SAPP activities are coordinated by the SAPP Coordination
Center based in Harare Zimbabwe.

Chapter 3: Basic concepts on Power

System Stability.
The chapter describes the classification of the power system stability and its basic concepts.
On the basic concepts, the chapter clearly explains various cases of system transient faults,
their effects on the power network and how the system is brought back to the stead state
with time.

3.1 Classification of Power System Stability.

Power system stability was first recognized as an important problem in the 1920s [l]. From
the beginning, for convenience in analysis, gaining a better understanding of the nature of
stability problems and developing solutions to the problems, it has been the usual practice
to classify power system stability basing on the following consideration [2]:

The physical nature of the resulting instability;

The size of the disturbance considered;
The devices, process, and time span that must be taken into consideration in order to
determine stability; and
The most appropriate method of calculation and prediction of stability.

The traditional large disturbance stability problem is related to the short term or transient
period, which is usually limited to a few seconds following the disturbance. It is concerned
with the system response to a severe disturbance, such as a transmission system fault.
Much of the electric utility industry effort and interest related to system stability to date
have been concentrated on the short-term response, and as a result the system is designed
and operated so as to meet a set of reliability criteria concerning transient stability [3]. Well
established analytical techniques and computer programs exist for the analysis of transient
stability. In recent years, the need for studying the response of the system for longer
periods has been recognized, and the terms mid-term and long-term stability have been
introduced. Analytical tools for studying these aspects of system stability are evolving.
As power system have evolved through continuing growth in interconnections, use of new
technologies and controls, and the increased operation in highly stressed conditions,
different forms of system instability, frequency stability and inter-area oscillations have
become greater concerns than in the past [4]. Voltage stability, frequency stability and
inter-area oscillations have become greater concerns than in the past. Historically transient
angle instability has been the dominant stability problem on most systems, and has been the
focus of much of industrys attention concerning system stability.
A typical modern power system is a high-order multivariable process whose dynamic
response is influenced by a wide array of devices with different characteristics and
response rates. Depending on the network topology, system operating condition and the
form of disturbance different sets of opposing forces may experience sustained imbalance
leading to different forms of instability. Below, in

Figure 2, is the summary of the classification of the system stability, giving the overall
picture of the power system stability problem and identifying its categories and
subcategories [4].
The classification of power system stability is for convenience case to identify causes of
instability, applying suitable analysis tools, and developing corrective measures. In any
given situation, however, any one form of instability may not occur in its pure form. This is
particularly true in highly stressed systems and for cascading events; as systems fail one
form of stability may lead to another form. However, distinguishing between different
forms is important for understanding the underlying causes of the problem in order to
develop appropriate design and operating procedures.
While classification of power system stability is an effective and convenient means to deal
with the complexities of the problem, the overall stability of the system should always be
kept in mind. Solutions to stability problems of one category should not be at the expense
of another. It is essential to look at all aspects of the stability phenomena and at each aspect
from more than one view point.

Power System
Stability

Rotor Angle
Stability

Small
disturbanc
e Angle

Frequency
Stability

Voltage Stability

Transient
Stability

Short
Term

Large
Disturbance
Voltage
Stability

Short
Term

Long
Term

Short
Term (110sec)

Small
Disturbance
Voltage
Stability

Long
Term
(160min)

Figure 2: Classification of power system stability.

As already stated, transient angle instability has been the center of all forms of instability.
In this thesis therefore we analyze further the rotor angle instability in response to shortcircuit faults.

3.2 An Elementary View of Transient Stability.

Power system stability is the ability of an electric power system, for a given initial
operating conditions, to regain a state of operating equilibrium after being subjected to a
physical disturbance, with most system variables bounded so that practically the entire
system remains intact[3]. The system response to such disturbances involves large
excursions of generator rotor angles, power flows, bus voltages, and other system variables.
The stability is influenced by the nonlinear characteristics of the power system. If the
resulting angular separation between the machines in the system remains within certain
bounds, the system maintains synchronism. Loss of synchronism because of transient
instability, if it occurs, will usually be evident within 2 to 3 seconds of the initial
disturbance [5] [6].
3.2.1 Loss of one Parallel Line
Consider Figure 3 below where the generator is delivering power to the infinite bus bar
through two parallel lines. The infinite bus bar represents a voltage source of constant
voltage magnitude and constant frequency;

Figure 3: Generator connected to infinite bus bar through two parallel lines.

In Figure 4 below all resistances are neglected.

Figure 4: Electrical representation.

The electrical power transfer to the infinite bus before one line is isolated is given by
(1)
Where is the angle between the internal voltage of the machine and that of the infinite
bus.
E is the voltage behind the transient reactance

V is the infinite bus voltage,

is the electrical power output and finally
Pm is the mechanical power in put.
After the line is isolated the electrical power transfer becomes
(2)
| | | |

If one of the circuits is out of service the effective reactance

the corresponding maximum power is lower.

(3)

in Figure 4 is higher and

Basically the two equations, equation ( ) and ( ) give the power angle curves as shown
below in Figure 5 where Pmax1 and Pmax2 power curves are clearly shown.

10

Figure 5: Equal area criterion for the loss of one parallel line
When the system is perturbed, the magnitude of E remains constant at its pre-disturbance
value and changes as the generator rotor speed deviates from synchronous speed .
3.2.2 Equal Area Criterion;
The equation of motion or the swing equation may be written as
( 4)
Where Pm

= mechanical power input, in Pu

Pmax = maximum electrical power output, in Pu

H

= inertia constant, in MW.s/MVA

= rotor angle in elect. rad

time, in s

From the equation ( ) we have following relationship between the rotor angle and the
accelerating power [2];
=

( )

- )

Now Pe is a nonlinear function of , and therefore the above equation cannot be solved
directly. If both sides are multiplied by 2d/dt, then
(

11

Or
[

Integrating gives
[ ]

( )

The speed deviation d is initially zero. It will change as a result of the disturbance. For
stable operation, the deviation of angle must be bounded, reaching a maximum value (as
at
on the line2 out- power curve on Figure 5 above) and then changing direction. This
requires the speed deviation d/dt to become zero at some time after the disturbance.
Therefore, from the equation( ), as a criterion for stability we may write;

( )

Where
is the initial rotor angle and
is the maximum rotor angle. As illustrated in the
last figure above. Thus, the area under the function Pm-Pe plotted against must be zero if
the system is to be stable. In the last figure above this is satisfied when area
. Kinetic energy is gained by the rotor during acceleration when
changes from to . The energy gained is
= (

The energy lost during deceleration when changes from

= (

= area

to

is;

= area

( )

( )

Since the losses are not considered the energy gained is equal to the energy lost which
means that area
in equation ( ) equal to area in equation( ). And this forms the
basis of equal area criterion.
The criterion enable us to determine the maximum swing of and hence the stability of the
system without computing the time response through formal solution of the swing
equation.
We now consider the loss of one parallel line. The initial operating point of the system
would be at point a (from figure4) where the mechanical power Pm is equal to electrical
power Pe. When the line is opened the operating point suddenly changes to point b because
the transfer reactance has now changed. Because of the inertia angle cannot change
12

instantly. The mechanical power is greater than the electrical power at this point hence
there will be some accelerating power that will make the rotor angle shift from
to
at
point c. At this point the mechanical power is equal to electrical power but because of the
acceleration power which makes the speed of the rotor higher than the synchronous speed
and the rotor angle continues to rise until it reaches maximum value
where the rotor
speed is now equal to synchronous speed and where the kinetic energy gained during this
acceleration will have expended by transferring the energy to the system. That is area
is
equal to area
At this point the electrical power will be greater than mechanical power
and consequently there will be some deceleration power and the rotor speed will be seen to
be reducing. Eventually we see that there will be some oscillations about point c until these
oscillations are damped.
If we now suddenly increase the mechanical power from the initial operating point shown
in Figure 5, we see a situation where we need more decelerating power available because
area
has now increased. At this point we search for the critical point, a point at which
the two areas will just be equal so that we dont lose stability of the system.
3.2.3 Fault at Either end of the Line
Another interesting situation is when we have a fault on one of the parallel lines but close
to either end of that line. See the Figure 6 below.

Figure 6: Short circuit at the end of the line

From the last figure above we see that there is no power transfers to the infinite bus
because the voltage just after the machine reactance becomes zero. See Figure 7 below.
During the fault the operating point is as shown in figure5 as

13

Figure 7: Power angle relationship after a fault at either end of the parallel line
In this case when a fault occurs the operating point suddenly shifts from point a to point b.
Before the fault the power angle relationship is given by
Pelect = PmaxI Sin and during the fault the power transfer is zero and after the fault is
cleared by opening the faulted line the power transfer becomes; PmaxII Sin as shown in
Figure 7 above.
The rotor angle wont change immediately owing to the inertia and it will move along the
zero line to
when the fault is cleared. At this point the new operating point will be at c.
and because rotor speed is greater than the synchronous speed the rotor continues to
accelerate until point d where the rotor speed is equal to the synchronous speed. At this
point the Electrical power is higher than the mechanical power hence the deceleration in
the rotor speed which gives out some oscillations until the oscillations are dumped.
From the figure above we see that if we increase the clearing time the acceleration area
becomes larger and this limits the deceleration area available to a critical clearing angle .
Similary if we reduce the fault clearing angle the smaller is the accelerating area compared
to the deceleration area available and hence more chance of the system remaining stable.
And if the initial operating point is raised a similar situation occurs where the decelerating
area available becomes smaller compared to the acceleration area.
We therefore conclude that the stability of the system is dependent on two things in this
case, the initial operating angle and the fault clearing time.
3.2.4 Fault in the Middle of the Line
Now let us take another scenario where the fault occurs in the middle of the line. See
Figure 8 below;

14

Figure 8: Fault in the middle of the line.

Figure 9: Circuit model for the fault in the middle of the line

Figure 10: Star delta transformation

15

Figure 11: Final circuit representation where only

reactance

is considered as the transfer

Figure 12: Equal Area criterion for the fault in the middle of the line
From the above figures, Figure 8, Figure 9, Figure 10, Figure 11 and Figure 12, we see that
the only difference between the previous case and this one is that in this case where the
fault is in the middle of the line the power transfer never goes to zero hence more chances
of the system remaining stable unless the fault clearing time is increased in such a way that
the accelerating area is larger than the decelerating area.

16

Figure 13: Critical clearing angle

The critical clearing angle is found at the point where the accelerating angle is just equal to
the decelerating area as we increase the fault clearing time. See Figure 13 below;

3.2.5 Auto-reclosing on to Either Transient or Permanent Fault

Another interesting scenario is the case of auto-reclosing on to the transient fault or
persistent fault. We consider the case where we have a fault in the middle of the line. See
Figure 14 below

Figure 14: Auto reclosing to either transient or persistent fault

17

If we auto-reclose to a transient fault we see that the decelerating area becomes larger due
to the fact that the new operating point goes back to the initial operating point. But if we
auto reclose to a persistent fault the decelerating area reduces and we may go into a
runaway situation because now the accelerating power is larger than the decelerating power
available. This situation can be avoided if the auto-reclosing is only allowed once and more
time set for the initiation of the auto-reclosing, say 10 seconds and by then the system will
have been stabilized. Delayed auto reclosing only applies to three phase faults because this
is when we have critical situation for the rotor stability.

18

Chapter 4: Power System Modeling

This chapter describes how various power system components are modeled in the Power
Factory Software. On the generating end side voltage regulators and speed governors are
modeled accordingly and finally the load side the load characteristics are modeled to give
desired results in the power system simulations.

4.1 Voltage Regulations

On all the generators connected to the Zambian transmission system, the units are equipped
with an Automatic Voltage Regulation (AVR) and a speed governor. Whatever the real
regulations used to equip the future HPP, the behavior will be very close of the proposed
models. The retained models are displayed hereafter. See Figure 15 and Figure 16 for
AVR models. Concerning the speed governor model, the IEEEG3 [13] model was
proposed for all the generators connected to the Zambian transmission system. It is
normally used for hydro generators. See Figure 17. The parameters of the regulations are
displayed in Appendix 11, Table 15.

Figure 15: AVR Model IEEET1A

Figure 16: AVR Model SCRX/SEXS

19

Figure 17: Speed Governor Model IEEEG3

4.2 Load Modeling

4.2.1 General Description
Load characteristics have long been known to have a significant effect on system
performance, and transient stability results are known to be highly dependent upon the load
characteristics assumed [7]. Because of the uncertainty of the actual load characteristics
utilities have attempted to use characteristics that would lead to conservative designs. It has
been shown, however, that there is no single load characteristic that leads to a conservative
design for all system configurations [8].
Dynamic load models are of growing importance to the studies of power system dynamics
[9]. If the load representation is not of sufficient accuracy, the simulation results will not
correspond to the actual response of the load. This will affect the assessment of system
stability limit [10]. Frequently, both power industry engineers and academic researchers
study system stability and planning by utilizing static load models (i.e. constant impedance,
constant current, constant power, and combinations of these models) to represent the
relationship between power and voltage. Because these load models are static and time
invariant, they are not sufficiently accurate to describe the load behaviors under various
operation conditions. The uncertainty regarding load composition and the sufficiency of
these static load models have been questioned in some publications [11]. However, the load
behavior is mostly dynamic, with the real and reactive powers being changed at any instant
of time [12]. For this reason, dynamic load models are considered.
In power systems, electrical load consists of various different types of electrical devices,
from incandescent lamps and heaters to large arc furnaces and motors. It is often very
difficult to identify the exact composition of static and dynamic loads in the network. This
load composition can also vary depending on factors such as the season, time of day etc.
[7].
20

Additionally, the term load can be used for entire MV-feeders in case of an HV-system or
LV-feeders if an MV system is in the center of interest.
4.2.2 Load Modeling in Power Factory Software.
In Power Factory, most of the loads are modeled as General loads where they represents:
A complete feeder
A combination of dynamic and static loads
The general load model representation is shown in Figure 18 below;

Figure 18: Power Factory general load model

For load flow analysis, in Power Factory software [13], there is an option to specify for
balanced or unbalanced loads. Furthermore the user can specify the input parameters for
the load. Based on the available data, the user can select the relevant combination of
parameters from S (apparent power), P (real power), Q (reactive power), cos(phi) (power
factor) and I (current). The other parameters such as voltage dependence loads and the
number of phases used are specified in Load type as shown in the window below in Figure
19.

21

Figure 19: Load flow calculation window indicating the voltage dependence terms for load
flow calculations in Power Factory software.
4.2.3 Three phase load modeling for transient Stability Simulations.
For transient simulations a three-phase load can be modeled as a percentage of static and
dynamic load. The static portion is modeled as constant impedance whereas the dynamic
load can be modeled as either a linear load or a non-linear load [13].

Z
%

Dynamic
Load %

Figure 20: Diagram indicating the mixture of static and dynamic load model used for
stability studies.
According to [13] the models used to approximate the behavior of linear and non-linear
dynamic loads are given in Figure 21 and Error! Reference source not found.
respectively.

22

Figure 21: Model used to approximate the behavior of the linear dynamic load.

Figure 22: Model used to approximate the behavior of the non-linear dynamic load.
23

Figure 21 and Figure 22 represent small signal model and are valid in limited voltage
range. This voltage range is defined as
and
and outside this range the power is
adjusted as shown in Figure 23 below [13].

Figure 23: Low/High voltage approximations used in the non-linear dynamic load model.

24

Chapter Five: Power System Simulations

Here the chapter describes the power system simulations in a detailed manner. The chapter
starts with the planning criteria to show some limits which are considered in the simulation
results followed by the methodology to show in which state the system is considered in a
particular simulation and finally the analysis of the various simulation results of different
cases.

5.1 Planning Criteria

The Zambia transmission system must satisfy the N-1 criteria, where it is meshed. So,
following the tripping of any network component (overhead lines, transformers, generators)
no overload must appear on the remaining components. Thermal ratings of equipment
should not be exceeded during the outage. The operational limits must be respected
whatever the situation.
The operational limits, in normal and emergency situations, are described hereafter. In
normal conditions the flows must be kept below the thermal rating of each network
element. The voltage at each bus bar must be kept within 95% and 105% of its nominal
value. The generators must operate within their reactive capabilities (generation and
absorption).
In emergency situation (outage of one element), the thermal rating of any equipment should
not be exceeded. The voltage should be kept within 95% and 105% of its nominal value. A
15% over-voltage will be acceptable during 5 s and a 20 % over-voltage will be acceptable
during 1s or 2s. This is according to IEEE 1453-2004 standard.
In transient state, the power system must remain stable following a three-phase fault
cleared within 100 ms. This time includes the current transformer errors, protection relay
response and the breaker operating time.
Transient stability is that ability of the power system to maintain synchronism when
subjected to a severe transient disturbance such as a fault on transmission facilities, loss of
generation, or loss of large load [14]. The generators must operate in synchronism without
inter-area oscillations. Power and voltage oscillations must be damped.
Inter-area oscillations can appear on the power system if the power exchange between two
parts of the power system (areas) is close to the maximum transmitted power [14]. They
can also appear between one power plant and the transmission system.
Spinning or primary reserve corresponds to the power delivered by generators due to the
action of their speed governor following a decline of the system frequency. All the units
may participate to the primary reserve. For the purpose of the load flow calculations, it has
been assumed that each unit can participate to the primary reserve up to 10% of its net
output. The primary reserve of the Zambian system alone will amount to 180 MW by 2011,
25

corresponding to loss of one large unit at Kariba North Bank or Kafue Gorge power
stations.

5.2 Methodology
The operation of the Zambian power system was simulated to analyze its behavior and its
ability to operate within its capacity in stead state and in transient state.
In permanent-state the electrical parameters that characterize the behavior of the system,
currents, voltages, generation, angles of the machine, position of the automatic on-load tapchange, are assumed to be constant. The load flow calculation is used to simulate the
behavior of the power system in permanent state and to calculate its electrical parameters.
The behavior of the power system is satisfactory if, and only if [14]:

The currents in the lines are below their rating;

The flows through the transformers are below their rating
The voltage profile is within the acceptable limits;
The reactive power output of the generators respects their capability curve.

All load flow calculations were carried out for the peak and the off-peak load demands.
In transient-state the system is affected by a fault, such as a short-circuit, a tripping of a
unit or a disconnection of a load, all these parameters evolve very quickly during a brief
period until the system recovers a new permanent state situation. If the system fails to do it,
a loss of synchronism between some units and other units of the system may occur [14].
The transient-state analysis determines if the up rating of the machines can improve the
control of the voltage and the frequency of the system. The electromechanical behavior of
the power system is analyzed versus time for few seconds. The generators and their control
system are represented in detail, taking into account their inertia, their turbine with the
speed governor, the electrical parameters of their alternator with their excitation system and
voltage regulator. Moreover, the voltage and frequency dependent characteristics of loads
are represented. A fault is applied on a line connected to Hydro Power Plant and then is
cleared by permanent tripping of the faulted element following a normal operation of the
protections. The dynamic behavior of the Machines connecting to the system is satisfactory
if [14]:
No loss of synchronism of units occurs;
No sustained voltage oscillations are detected on substations;
No sustained oscillations of the rotor angle appeared on the machines;
Voltage profile and currents remain within the acceptable limits.

5.3 Analysis in Stead State

5.3.1 Presentation
The simulations were performed for the year
capacity as indicated in Appendix1. They were
the network elements in operation. Kafue Gorge
output minus the primary reserve. They were
following N -1 cases were examined:
26

2011 with the 10% uprated generation

performed in normal conditions, with all
Power plant operates at almost maximum
also performed in N-1 conditions. The

Tripping of one of the two 330 kV circuits connecting Kariba North to the Zambian
system.
Tripping of one of the two 330 kV circuits between Kafue Gorge and LHill
substation.
Tripping of 330 kV Kafue Gorge-Kafue west line.
Tripping of 220 kV Vic.Falls-Muzuma line
Tripping of one of the three 330kV circuits between Leopards Hill and Kabwe.
Tripping of one of the two circuits between Kabwe and Luano and between Kabwe
and Kitwe.
Tripping of the 330 kV Luano-Kansanshi Mine.
Tripping of the 330 kV Kabwe-Pensulo Line.

The generation had been adjusted to the Zambian load in order to have no transit on the
two
circuits interconnecting the Zambian and the Zimbabwe systems, i.e. between Kariba North
Bank and Kariba South Bank substations. The power system must operate within the limits
defined by the Zambian grid code. If it does not, improvements (such as capacitor banks)
will be proposed to allow a satisfactory operation.
The load flow calculations were performed in normal and in N-1 conditions for the peak
and the off-peak.

5.4

Load flow Calculations at Peak Period

Load flow gives us the analysis of the power system in terms of power flow, real and
reactive power, on various lines and the voltage magnitudes and respective angles at
various bus bars at a particular time instant for given condition of Generation and loads
[13].
The power system operation is considered as Quasi-static, the static analysis of power
network. The real power balance, reactive power balance, transmission flow limits and bus
voltage limits is studied.
In Power Factory software the following is specified: the generating bus bars as PV buses
simply because at these buses the voltage magnitude is maintained by the excitation of the
generators. The Kafue Gorge bus bar has been selected as slack bus bar. This is a reference
bus bar where voltage magnitude and angle are specified. This means that all loses are
assigned to this bus bar. The real and reactive power is allowed to vary at this bus bar.
5.4.1 Analysis in Normal Conditions
The simulation was performed in normal conditions; all the network elements were in
operation. The generation is located in the south of the transmission system and about 45%
of the total demand is located in the north of the system (Copper belt area).See Figure 24
and Figure 25 below. The North of the system is then supplied from the South through four
330 kV lines between Kabwe and Luano substations of which two are direct lines ,then
one goes via Kitwe substation to Luano the fourth one goes via Kitwe and Chambishi
substations. See Figure 1. Kariba North substation is linked to Kariba south substation
through two 330kV circuits. These two border systems are balanced in order to have no
27

load on these two circuits. The unit commitment of the Zambian network is displayed in
Table 2. Figure 24 below shows the Generated power, loses and the installed capacity
As can be seen Figure 24,the system is operating at almost full capacity hence the spinning
reserve cannot hold system stability in case of loss of one generator in Kafue Gorge or
Kariba unless the interconnections from other mentioned countries are considered as it is
shown in the N-1 analysis below.
In case of voltage profile, on the Copper belt area, the voltage can be controlled in a
satisfactory way with two Static Var compensator units, which have been implemented in
two main 330kV substations (i.e. one in Kitwe and one in Luano). To operate the system
within the voltage limits, the reactive compensation means were connected as shown in
Appendix7 Table 3. Despite heavy loads on the copper belt, which accounts for 45% of the
total load, the voltage profile is within the limits. The power factor is still ok with this
compensation despite the high inductive load on the copper belt.

Figure 24: Calculated load flow at peak period

28

Figure 25: Copper belt load under calculated load flow.

The loading on Copper belt lines are within limits. As at now the lines are only loaded
25% to 36% of the capacity as shown in Table 4. Therefore for N-1 criterion the lines will
still be safe in terms of their loading capacity.
With these capacitor banks on the copper belt main load centers and two Static Var
Compensators, the behavior of the Zambian transmission system was generally
satisfactory.
The voltage profile of Kafue town 88 kV substation exceeded the lowest limit (0.95 p.u.),
see Figure 26 below. This is because the reactor which prevents high voltages during light
loads is on, but this is normally out during heavy loads and the voltage profile in this
earlier improves.

29

Figure 26: voltage profile for the calculated load flow.

The voltage profile of Leopard Hill 88 kV and Kabwe 330 kV substations (1.036 p.u) was
the closest to the upper limit. Luano recorded the lowest voltage on the copper belt about
0.94 p.u as seen from Figure 26. This is because of the increased load on the two new mine
Kansanshi and Lumwana connected to this bus bar. At Lumwana we have capacitor banks
that improve the voltage profile in this area. See Figure 27 below,

30

Figure 27: Voltage profile for the calculated load flow under normal system condition

5.5 Analysis in N-1 Conditions at Peak Load

Tripping of lines between Major Hydro Power Plants and the Copper belt area were
analyzed at peak load:

One of the three 330 kV circuits between Kariba North and Leopards Hill
One of the three 330 kV circuits between Leopards Hill and Kabwe
One of the two 330 kV circuits between Kabwe and Luano
One of the two 330 kV circuits between Kabwe and Kitwe
Pensulo line from Kabwe

After each tripping, the Zambian network impedance is increased and so the losses. Then,
the flow over the two circuits between Kariba North Bank substation and Kariba South
Bank substation will correspond to the additional losses.
5.5.1 Tripping of one of the two 330 kV circuits between Kariba North and Leopards Hill
Following the tripping of the circuit, the behavior of the system was satisfactory, no
overload
31

Occurred on the transmission system. The initial power flowing through the tripped circuit
Kariba North - Lhill 1A was the same as that of the other parallel circuit. See appendix8.
The tripped line was loaded to 27% of its capacity. Therefore the remaining line was
loaded to 54% after a tripping.
The voltage profile of the Copper belt area kept to almost its initial value thanks to the
SVC of Kitwe and Luano. The reactive power on Kitwe SVC bank was raised from 2 Mvar
to 18 Mvar and from 3 Mvar to 24 Mvar for Luano substation. See Table 3 in appendix 7.
The most significant voltage drop is at Leopards Hill substation with a dip of 1% (i.e. from
340 kV to 336.5 kV). See Figure 28 below.

32

Figure 28. Voltage profile with Kariba N-LHill line out.

After the tripping, the voltage profile stayed within the acceptable limits:

346 kV at Kariba North (1.05p.u.)

320 kV at Luano(0.97p.u.)
318 kV at Kitwe(0.966p.u.)
343 kV at Kafue Gorge (1.04p.u.)
339.62 kV at Lumwana (1.02)

The lowest voltage in the Copperbelt area reached 0.95p.u. at Michelo and in the south at
Kafue town. See Table 7 in Appendix 10. Kariba North units generated 30.7 MVAr, which
were within their reactive limits. Kafue Gorge units generated 53 MVAr, which were
within their reactive limits.
5.5.2 Tripping of one of the three 330 kV Circuits between Leopards Hill and Kabwe
Following the tripping of the circuit, the behavior of the system was satisfactory, no
overload occurred on the transmission system. The initial power flowing through the
tripped circuit Leopards Hill Kabwe was 34.6% for each line. After a tripping the power
for one line was equally dispatched between the two remaining circuits. See Table 5 in
Appendix 8.
The voltage profile of the Copper belt area kept to its initial value. Much of the reactive
power was supplied by the SVC of Kitwe and Luano to the north. The Generated reactive
power on all Machines at the big Generating stations in the south was within limits. The
most significant voltage drop is at Kafue 88 kV substation in the south (0.84 p.u.) and
Michelo 220 kV close to the lower limit (0.95 p.u.) in the north. After the tripping, the
voltage profile stayed within the acceptable limits

33

5.5.3 Tripping of one of the two 330 kV Circuits between Kabwe and Luano
Following the tripping of the circuit, the behavior of the system was satisfactory, no
overload occurred on the transmission system. The initial power flowing through the two
lines to Luano from Kabwe was 30% on each line. The power flow on the tripped circuit
Kabwe - Luano had been dispatched between the remaining circuit Kabwe - Luano and the
two circuits Kabwe - Kitwe which were currying 36% each initially. After the tripping, the
circuit Kabwe - Luano was loaded at 50% (350 MW) and the two circuits Kabwe Kitwe
were loaded at 55% each (382 MW). See Table 5 in Appendix 8.
The voltage profile of the Copper belt area was virtually kept to its initial again with the
help of the SVC of Kitwe and Luano
Kariba North units generated 50% of their reactive power limits and Kafue Gorge units
generated 66% of their reactive power limits. The results of the load flows calculations are
displayed in Appendix 8.
5.5.4 Tripping of one of the two 330 kV circuits between Kabwe and Kitwe
Following the tripping of the circuit, the behavior of the system was satisfactory, no
overload occurred on the transmission system. A similar situation to the previous scenario
prevailed
5.5.5 Tripping of Kabwe-Pensulo Line
After the tripping of this line the system was stable. No overloads on the lines and no major
under voltages recorded. The only problem was on the Generators on Lusiwasi s/s which
could trip on under frequency because the Northern part load is transferred to these
generators of limited capacity.
All critical N-1 contingency calculations are displayed in Table 6 to Table 13 in appendix
9.

5.6 Analysis at off peak for both normal and N-1 contingencies.
This operation was similar to the above analysis. This is made possible for the fact that the
generation was reduced according to the adjusted load and the system compensation
followed suit.

5.7 Analysis in transient state

The analysis in transient state is based on the theory presented in chapter three. Here the
classification of the transient stabilities is highlighted to give a clear picture of what to
consider in the presentation in this thesis. The short term stability is considered in the
analysis of the rotor stability for the fact that rotor stability is the predominant among the
three types of instabilities. The three types being rotor angle, voltage and frequency
instability. The time frame considered in this short time stability studies is 10 seconds.
Also in chapter three is the analysis of equal area criterion for four different cases. The
switching off of one line for two parallel lines, the application of the fault at the end of any
line, the application of the fault in the middle of the line and clearing it and finally the
effect of Auto reclosing on either transient or persistent fault. This is to get the critical

34

clearing angle at which the system would remain stable in any situation. This of course
depends on the fault clearing time.
All these scenarios are simulated and analyzed on the Zambian network. From chapter
three it is shown that the initial operating point is crucial for the stability of the system. The
higher the initial maximum operating power output of the generator the shorter is the
critical clearing time. This is the reason for basing the analysis on the peak period in the
system.
To summarize the analysis presented in chapter three, the following are the factors
influencing the transient stability of the generator;

How heavily the generator is loaded

The generator output during the fault. This depends on the fault location and type.
The fault clearing time.
The post fault transmission system reactance.
The generator reactance. A lower reactance increases peak power and reduces initial
rotor angle.
The generator inertia. The higher the inertia, the slower the rate of change in angle.
This reduces the kinetic energy gained during fault, that is area
is reduced.
The generator internal voltage magnitude (E). This depends on field excitation.
The infinite bus voltage magnitude

5.7.1 Transient State Presentation

The purpose of the study in transient state is to demonstrate that, following a severe fault
normally cleared, the power system is able to recover a satisfactory behavior without loss
of the synchronism of any generator and without sustained voltage or power oscillations.
The voltage profile and the system frequency must remain within the acceptable limits.
The fault consists of a three-phase to ground short-circuit located close to Generating
stations on one connection circuits to either 330 kV LHill Substation from Kariba and
Kafue Gorge or 220 kV Kafue town substation. It is the most severe fault for the
generators. The fault is assumed to be cleared by opening of the two end breakers
simultaneously, isolating the faulted element.
The simulations provide the evolution versus time of the main parameters that are of great
importance in power system stability: rotor angle, excitation voltage, active power and
power frequency.
This part of the thesis presents hereafter the hypothesis, the main results and conclusions of
the study in transient state.
5.7.2 Hypotheses
5.7.2.1 Reserve Power
This is the difference between the Pmax of the units in operation and their generation output.

35

The primary reserve of Zambian system amounts to 180 MW, representing one of the
biggest generating units. With the interconnections to the neighboring countries, SAAP in
particular, Zambia gives a power reserve of about 70 MW [14].
The effective primary reserve of the HPP units depends on the opening time of the valves
of the turbine.
5.7.3. Simulated Faults
Three phase faults were applied near generating stations to ascertain the stability of the
Zambian Network. Following cases were examined:

Three phase to ground fault eliminated by definitive tripping of the faulted line in 100
ms. That is both ends of the faulted line being isolated simultaneously.
Three phase to ground fault eliminated by definitive tripping of the faulted line in 100
ms then the faulted line is automatically reclosed after 10s from the generating side of
the line and the other end auto reclosing 5seconds thereafter. In all auto reclosing
cases there was no switch on to fault. The fault was first cleared before auto reclosing.

The following faults were examined:

Three phase fault on the transmission line close to 330 kV Kafue Gorge HPP bus bar
Three phase fault on the transmission line close to 330 kV Kariba North HPP bus bar
Three phase fault on the transmission line close to 220 kV Vic. Falls HHP bus bar
Three phase fault on the transmission line close to Kafue Gorge HPP on KFG-Kafue
west line
Three phase fault on the 330 kV transmission line between LHill and Kabwe
substations
Three phase fault on the 330 kV transmission line between Kabwe and Luano
substations
Three phase fault on the 330 kV transmission line between Kabwe and Kitwe
substations
Three phase fault on the 330 kV transmission line between Kabwe and Pensulo
substations
Three phase fault on the 330 kV Kansanshi transmission line from Luano

5.7.3.1 Close up fault to Kafue Gorge on KFG-LHill Line at peak load

A three phase to ground fault was applied near Kafue gorge Power station with a capacity
of 990 MW. See Figure 29 below. This fault was applied after 3 seconds from the start of
simulation just to give initial conditions to compare with the post fault results. The fault
was applied at a position 5% of the line from the Hydro power plant.
The fault was Isolated after 100ms. Within this time the current transformer errors, relay
time and breaker contact opening period is taken care of. For the SF6 breakers that we have
in circuit, it takes only one cycle, 20ms to have the contacts opened fully. Infact the
recommended period for the contacts to open fully from the time of fault initiation is 70ms
for ZESCO protection philosophy for the system with SF6 breakers.

36

The behavior of the system during and after the fault was satisfactory. The generators
operated above the reactive limit during the fault but eventually stabilized. See Figure 30
below.
All groups operated in synchronism. A voltage dip close to zero for 100ms (during the
fault) was recorded on Kafue Gorge 330 kV busbar. The maximum rotor angle on Kafue
Gorge HPP was 80 degrees peak to peak as can be seen from Figure 30. The oscillations
were damped within 10seconds see Figure 30.

Figure 29: Three phase to ground fault on KFG-LHill line

Figure 30: From top to bottom: Excitation voltage, Excitation Current, Machine reactive
power, Voltage dip and Rotor angle variations before, during and after the fault close to
KFG HPP
The power oscillations on the remaining KFG-LHill line reached first high value of 600
MW from initial stead state value of 210 MW before stabilizing to a value of 425 MW. The
power oscillations were damped within 10seconds. No sustained oscillations were seen.
See Figure 31 below;

37

Figure 31: Power swings on the four lines from Main Generating stations to Lhill
Substation
Frequency deviations were not alarming on all Generators from all major hydro power
plants. They were within acceptable limits as can be seen from Figure 32.

Figure 32: Frequency deviation on major Hydro Power Plants

A voltage dip of 70% was noticed on major substations on the copper belt far away from
generating stations. However the voltage was recovered just after 100ms. See Figure 33
below.

38

Figure 33: Voltage dip on major Load centers on the Copper belt after a fault close to
KFG-HPP
Another interesting case was the increase of the clearance time to see if the line would still
be stable if the fault would be cleared in 400ms according the zone two settings of the lines
on the distance relay. Zone 2 is the backup protection to zone one. If the fault is not cleared
in zone one then zone two should clear it.
The best clearing time that was satisfactory was 200ms further than that the system was
very unstable. Oscillations were never damped. See
Figure 34 below for a 300ms fault. For more variables on this simulation see appendix 13.

Figure 34: Voltage and rotor angle oscillations for a 300ms fault on KFG-Lhill line

39

5.7.3.2 Close up fault to Kafue Gorge HPP on KFG-LHill Line at peak load with auto
reclosing
Another necessary simulation on the same line was auto reclosing to a transient and
permanent fault. For a transient fault, a three phase fault was applied 1 second from the
initiation of the simulation. After the fault was cleared, at t =1.1 seconds during simulation
time, the line auto-reclosed after 10 seconds according to the relay settings. The first end of
the line near, KFG HPP was reclosed at t = 11.1 then five seconds later the other end was
reclosed. This line is set to be synchronized at Lhill S/S.
The system operated satisfactory after this operation. There were no sustained oscillations.
See the results in Figure 35.

Figure 35: Auto reclosing of the KFG-Lhill line on transient fault

The voltage dip on KFG 330 kv bus bar reached almost zero. On Kariba 330 kV bus bar
the voltage dip reached 65% of the nominal value and that of Vic Falls the voltage dip
reached 78% see Figure 35.
The power swings on the remaining healthy lines were damped within 10 seconds. See
Figure 36.
The voltage dips on major Copper belt load bus bars were recorded to have dropped to 27%
because of the KFG busbar short cicuit. They were recovered just after fault clearance as
can be seen from Figure 36 below.
Frequency deviations were recorded high about 1.1% on KFG machines as compared to the
rest of the Machines on Kariba and Vic Falls Power Station because flow from KFG
machines was interrupted hence the power input was higher than the power output..After
40

the reclosure the frequency deviation was seen to be lower than the nominal due to
increased power flow to the north. See Figure 36 below.

Figure 36: From top to bottom, power swings on main generating stations, Voltage dips on
main Load bus bars and Frequency deviations after auto reclosing the KFG-LHill line.
5.7.3.3 Auto reclosing on to a persistent fault on KFG-Lhill.
The operation of the system after auto reclosing to a fault 10 seconds after the line was
isolated was satisfactory. All oscillations were damped. The maximum rotor angle reached
upon auto reclosing was almost like at the first fault. This is because the system was almost
stable after 10 seconds. See Figure 37 and 38 below;

41

Figure 37: Auto reclosing on persistent fault and locking out

Figure 38. Auto reclosing on persistent fault

42

There were no major power swings on the copper belt loads. The voltage dip reached 27%
of the nominal value due to system losses. The frequency on the Kafue Gorge machines
went up higher than the rest of the machines on the other hydro power plants and was seen
to stabilize at a lower value due to increased load flow. See Figure 39.

Figure 39: Auto reclosing on persistent fault- power swing on the copper belt lines, voltage
profile and machine frequency on major generating stations.
The reactive power on the Kafue Gorge machines exceeded the limits but was stabilized.
See Figure 40. The increase of the reactive power on the machines is as a result of the fault
current flows through pure reactance to the fault hence only reactive power flows [2].

43

Figure 40: Excitation voltage, current and reactive power on all machines on major
generating stations-auto-reclosing on a persistent fault
The power swings on the four main lines from major generating stations were damped. See
Figure 41 .

Figure 41: Power swing on the four transmission lines to LHill substation from Kariba and
Kafue Gorge HPP after auto-reclosing on to a persistent fault.

44

5.7.3.4 High speed auto reclosing.

In this simulation the auto reclosing was deliberately initiated to close 70ms after the trip
out on to a persistent fault to just find out the effect of this operation otherwise it does not
apply to the Zambian system unless on single pole auto reclosing. See Figure 42 to Figure
45 for the auto-reclosing effect.
From the Figure 42 below the rotor angle was raised to almost double the initial highest
value. This poses a danger for instability as explained in chapter three.

Figure 42. Effect of high speed auto reclosing (Rotor angle, Voltage profile and rotor angle
difference)

45

Figure 43: Effect of high speed auto reclosing (Rotor angle difference, Voltage angle &
speed deviation)

46

Figure 44: Effect of high speed auto reclosing (Active Power, Voltage profile &
Frequency)

47

Figure 45: Effect of high speed auto reclosing (Excitation voltage, Excitation current,
reactive power and Active power flow)

48

5.7.3.5 Faults on the remaining lines close to the main Hydro power plants.
The behavior of the whole system after a fault close to the main hydro power plants on the
remaining circuits was satisfactory. The pattern of variable oscillations was similar to the
one analyzed here before. The rotor angle oscillations were damped within 10seconds. No
sustained oscillations of any kind were recorded for a fault cleared in 100 ms. See Figure
64 to Figure 99.
The voltage dips were recorded on the major substations on the copper belt but were not
sustained. The post fault voltages were close to the initial value which means the loses
were not all that high after the isolation of the faulted circuit.
5.7.3.6 Busbar faults
The only scenario on power system stability that was different from the one just analyzed
here before is that of a bus bar fault. A complete bus bar fault was simulated at Leopards
Hill substation. This is the focal point of all the power to the copper belt that accounts for
about 45% of total power from major Generating stations. The fault was applied 1 second
after the start of the simulation. The fault duration was 50 milliseconds. This fault duration
accounts for CT error, relay and breaker opening times.
The power system operation after the fault was not satisfactory. This was expected since
45% of the total load was suddenly interrupted that means more power input than the
power output.
The rotor angle and voltage oscillations for Kafue Gorge and Vic. Falls were not damped
except for the Kariba North Hydro power plant machines. See figure 40 below. The KNBC
machines were spared because of the interconnection to the Southern load that provided
some inertia. Infact the machines were lightly overloaded at this stage as can be seen from
the new operating angle. It dropped to 32 degrees from the initial 40 degrees.

Figure 46: From top to bottom: Rotor angle for KFG HPP machines, Voltage magnitude
and the rotor angle for KNBC HPP for a bus bar fault at LHill Substation.

49

The frequency deviation on the remaining machines was not satisfactory. The one for Vic
falls looks like it was about to stabilize because of the light load to the western part of the
country that are connected to the Vic falls Power Station machines. The Kariba North Bank
HPP machine frequency deviation was satisfactory. There was a bit of deviation because of
the light load after the fault. See Figure 48.

Figure 47: Rotor angle, voltage angle and speed deviation on all major HPP machines for a
bus bar fault at LHill substation
As can be seen from Figure 48 below, the power flow to the north was suddenly disrupted.
The voltage totally collapsed. The frequency on the KNBC HPP machines looks ok. It
slightly went up for the light load as already mentioned.

50

Figure 48: Active power on the transmission lines to the North, Voltage profile on the
Copper belt Major substations and the frequency on all HPP generators for a bus bar fault
at L.Hill substation
.
The solution to the above problem is to stager the circuits on the Leopards Hill substation
bus bars. At this substation there are two bus bars coupled by a bus coupler breaker. If a
fault occurs on one of them the only faulted bus bar will be Isolated and the other will
remain intact to save copper belt load. With this kind of simulation the results are shown in
Figure 49 and Figure 50 below. All oscillations were damped and the voltage profile on the
copper belt looks ok despite the added loses with the tripping of two lines connecting the
next substation, Kabwe. This also reduces the power flow to copper belt as can be seen
from Figure 50. With this power flow reduction the frequency value went up but it was
within limits, see Figure 50.

51

Figure 49: Rotor angle and voltage profile for the fault at Leopards hill with the staggered
circuits for a bus bar fault.
Only slight voltage drop was recorded on the copper belt due to the increased loses as can
be seen from Figure 50. There was some slight power swing on the northern power lines.

52

Figure 50: Active power flow to the north, voltage profile at the copper belt and the
generating frequency for a bus bar fault at LHill

53

Chapter 6: Conclusions and

Recommendations
6.1 Conclusions
Generally the Zambian system operation is very stable. On transient stability analysis for a
severe fault close to major Hydro Power Plants and cleared within 100ms the operation of
the system is satisfactory. But this type of fault cannot be sustained for more than 200ms. If
the fault is cleared after this time the system becomes unstable. The fault clearing time
reduces further if the power reserve is low. With the increase of power demand and less
generated power the effect is the instability of the power system for faults close to the
generating stations as the case will be in 2011 for the Zambian System if more Generating
stations will not be built by then. But this situation can be avoided with the increased
interconnections to the neighboring countries like Congo Dr to the North and Zimbabwe,
Mozambique and South Africa to the South. This scenario will increase the reserve power
for the Zambian Power System and hence the increase in the stability of the Power System.
Auto-reclosing on to a sustained fault has no major effect on the stability of the Zambian
system if it is initiated 10seconds after the isolation of the fault. 10 seconds is enough
period for the system to be stable following a major system disturbance. If for any reason
the auto reclosing is initiated after 1ms the system is likely to become unstable and this is
only applicable to three phase auto-reclosing. High speed auto-reclosing is applicable to
single phase only.
For a bus bar fault the effect is pronounced if the circuits are not staggered at that particular
bus bar. But this does not guaranty a good voltage profile on the Northern system because
of losses resulting from increased line impedance. However this voltage profile is eased by
the use of the SVCs. SVCs are better than the Capacitor banks for intermitted voltage
variations.
The voltage profile on the Zambian system is generally ok both in normal and in N-1
contingencies. The installations of Capacitor banks in substations like Luano, Chambishi,
Kansanshi and Lumwana has really helped in improving the voltage profile on the
Copperbelt area. But these capacitor banks will not be as effective as SVC because of the
low short-circuit power in the Copper belt area. A low short-circuit power means that a
small reactive power variation will have a significant impact on the voltage level. The SVC
is continuously bound to a voltage set point and so adjusts its reactive power to the wanted
voltage value. Thus, for N-1 and peak loads the SVC, with a smooth voltage adjustment,
will be a better solution on the voltage variations on the Copper belt than a bank of
capacitors which will respond more bluntly and thus causing unwanted voltage variations.
The other factor affecting the voltage profile is the location of the Generating stations in
relation to the load center. The generating station is in the south and the major load center
in the north with the distance of about 800 km gives a great risk for a voltage collapse if a
major line fault on the lines connecting the two centers occurs. There are some limitations
on the operation of SVCs to arrest the voltage collapse for a situation like this one where
the distance between the generating stations and the load center is long unless we had some
generating stations in the north.
54

6.2 Recommendations
The stability of the Zambian Power System can be further enhanced if more generating
plants are implemented as the load grows. As at now the spinning reserve when operating
at peak loads is very small, about 70 MW. Already two by 180 MW Hydro Power
Generators are being implemented at Kariba North Bank to be commissioned in 2012 but
this will be taken up by the ever increasing demand with the increase in the number of
mining industry.
A research to come up with the specifications to acquire and install SVCs on major
substations on the copper belt is recommended for the reason already given in the
conclusion remarks. There should, also, be a thought to come up with the Generating
stations in the North for System stability especially voltage stability. If Generation capacity
is up graded in Lusiwasi which is part of Northern part of Zambia then we cannot have
further instability problems that arise from the faulted Kabwe-Pensulo line.

6.3 Future Work

Short term static and transient stability studies should be done for the inclusion on the
system of the two 180 MW generators at Kariba North Bank Power Plant to ascertain their
effect.
Long term Transient simulations can be done to ascertain the suitability of the said SVCs.
This thesis only covers static and short term transient simulations.

55

REFERENCES
[1]

Energy regulation Board Report of Zambia, Power outage report 4th June, 2006

[2]

P. Kundur, Power System Stability and Control,

System Engineering series-1994 edition.

[3]

A Volume in the EPRI Power

P.Kundur, G.J.Rogers, D.Y.Wang, L.Wang, M.G.lauby, A computer program

Package for small signal stability Analysis of power systems, ALEE Trans.On power
System, Vol.5, No.4, November 1990.

[4]

P.Kundur, J. Paseba, S.Vitet, Overview on Definition and classification of Power

System Stability. IEEE/CIGERE joint Task force on stability term and Definition.

[5]

R.T. Byerly and E.w. Kimbark, Stability of Large Electric Power Systems, IEE
Press, 1974

[6]

P. Kundur, Power system stability and Control, A Volume in the EPRI Power
System Engineering series.

[7]

IEEE Task Force on Load representation for dynamic performance, Bibliography on

Load Models for Power and Dynamic performance Simulation, IEE Trans. On
Power System Vol. 10, No. 1, 1995.

[8]

M.Sedighizadeh, A.Rezazadeh, Load modeling for power flow and transient stability
Computer studies at Bakhtar, Network World Academy of Science, Engineering and
Technology 36 2007.

[9]

M.h Nehrir, P.S.Dolan, Development of physically Based computer model for

Predicting winter electric loads, IEEE Trans. on, Power Systems ,Vol 10, No.1,
P.266-272, February 1995.

[10]

L. Hajagos, Behnam, Danai, Laboratory measurement and models and their effects
on Voltage studies, IEEE Trans.On power system vol.13, No.2, PP.584-592, May
1998.

[11]

W.S. kao, Dynamic Load modeling in Power System Stability studies, IEEE Trans
56

on Power systems, vol.10, No.2PP.907-913 May 1995.

[12]

W.S.Kao, comparison simulated power system dynamics applying various load with
actual recovered data , IEEE Trans.On power systems vol.9, No.1PP284-254,
February.

[13]

Digsilent, Power Factory technical reference manual, Version 4.0.

[14]

EDF. Power System & Transmission Engineers of Germany, Consultancy Report on

Kariba North Extension - Zambia. June 2009.

57

Appendix 1: System upgrades

1. ZESCO System Upgrades
1.2 Completed generation and transmission system works.
Between 1999 and 2003 ZESCO embarked upon the rehabilitation of the entire
transmission system of Zambia. The specific works that were carried out are as follows:
New Transmission Line: Kafue west - Lusaka west
a) This involved the construction of a 43 km long 330 kV transmission line by from
Kafue West Substation to the new Lusaka West Substation with most of the materials
(steel towers and conductors) coming from the abandoned 330 kV Tika line project.
b) The objective of these works was to increase reliability of supply to Lusaka by
providing a second Bulk Supply Point for Lusaka.
c) Victoria falls Katima mulilo 220 kv interconnector
d) This project involved the construction of a 230 km long 220 kV transmission line
from Victoria Falls Power Station in Zambia to Katima Mulilo in Namibia. The line is
principally to facilitate exports of up to 200 MW to from Zambia to Namibia. On the
Namibian side a high capacity HVDC line will be constructed to facilitate the
interconnection of the two systems. The line was completed and commissioned by
August 2006.
e) Refurbishment and uprating of Generators
f) All three major power stations at Kafue Gorge, Kariba North Bank and Victoria Falls
are being refurbished to give them another 30 years of reliable service life. In addition
at Kafue Gorge the six generators will be uprated from 150 MW to 165 MW each
giving a total capacity of 990 MW from the current 900 MW. The four machines at
Kariba North will be uprated from 150 MW to 180 MW to increase the installed
capacity from the current 600 MW to 720 MW. The uprating also applies to the
generator transformers. Protection and control systems are also being modernized.

58

Appendix2: Identification of relevant sub-systems of the

network
The Zambian power network has all the major hydro power stations located in the southern
part of the country. The major power stations are linked to Leopards Hill substation in
Lusaka, which in turn is connected to the Copper belt via Kabwe step down substation.
The Zambian power system is linked to the Zimbabwean System through two 330 kV
transmission lines from Kariba North Bank power station to Kariba South power station in
Zimbabwe. It is also linked to the SNEL power system in Congo through the 220kV
Luano-Karavia transmission line.
The relevant interconnected power sub-systems are depicted in Figure 51.
The fact that all the major power stations are located in the south and the major load
centers in the north linked by long transmission lines creates a reliability challenge.
Further, although the transmission lines are a strong link for bulk transfer of electrical
power, they are all located within a common way-leave making them vulnerable to
common cause damage.
Nevertheless, system reliability could be enhanced by distributed generation, strengthening
of inter-connectors with neighboring countries, and reinforcing the nation grid with
alternative transmission lines through different geographical routes.
Currently, the link to the south is very strong with a capacity of 1400 MW which is capable
of carrying most of the current Zambian peak demand of 1670 MW. On the other hand, the
connection to DRC is weak with a capacity of 250 MW. To strengthen the interconnection
to DRC, CEC and ZESCO are considering building new lines.

Figure 51: Relevant interconnected power sub-systems

59

Appendix 3: ZESCO protection philosophy

3. ZESCO transmission protection philosophy
Prior to commencement of the Power Rehabilitation Project (PRP), the ZESCO 330 kV
system consisting of transmission lines, power transformers, generators and bus-bars, were
protected using old electromechanical relays. However, these old relays became obsolete
due to changes in technology. This meant that there were no longer spare parts to repair
broken down equipment as manufacturers had moved to new numerical technology. This
made maintenance of relays and control equipment very difficult and the system was at risk
of becoming unsafe due to inadequate protection as some equipment became inoperable.
The main purpose of the rehabilitation project was to enhance system security by replacing
old equipment that had become obsolete with the introduction of modern protection
equipment.
Therefore, the Corporation decided to replace all old relays with new numerical ones. In
2004 ZESCO Limited completed the transmission system rehabilitation in which the main
focus was on the upgrade and rehabilitation of the 330 kV protection, control, metering and
primary equipment such as circuit breakers.
To enhance the protection system further, the old bulk oil and air blast circuit were
replaced with faster and more reliable SF6 gas breakers with two trip coils for increased
reliability.
The DC system at the major substations, that consisted of battery sets and charger were
also replaced. To increase the reliability two sets of batteries and chargers are each used to
supply one of the two trip coils.

3.1 Old Protection Philosophy

Before the transmission rehabilitation was implemented, the protection philosophy was
based on the following; Main 1: Electro-mechanical Distance relay type TS and H were
used on the 330 kV transmission lines and the scheme used was Permissive Under-reach
based on mho characteristics for fault loops.
Backup Protection: Electro-mechanical Over current and Earth Fault relay with IDMT
Characteristics.

3.2 New Protection System Philosophy

As part of the preparation for the transmission rehabilitation project, Nor consult was hired
to carry out an intensive review of the state of all the plant on the transmission system. This
work was carried out over a period of one year. One of the main recommendations of the
study was on the need for adopting a new protection philosophy. Nor consult and ZESCO
formulated the new protection philosophy jointly. The main principles of the philosophy
are; Main 1 protection that is Numerical Differential, Main 2 protection which is
Numerical Distance and Breaker failure protection. Backup protection: Directional earth
fault (DEF) with a permissive overreach feature. Over-current and earth fault with IDMT
characteristics. Bus bar protection with two zones: main and check whose Primary function
is to protect bus-bars,Breaker Failure protection which is fast local backup and Two
tripping Supply system.
60

The set time for differential protection is instantaneous tripping and for distance protection
the time is graded in three zones. Zone 1 is instantaneous tripping, zone 2 we have 400
seconds and zone 3 is 1.5 s.

61

Appendix4: Previous major faults

In order to do a better stability study on the Zambian power network, there is need to
review the past major system disturbances.
Previous experiences affecting the Zambian system in the past 20 years include the 1989
Kafue Gorge Power Plant fire that led to the disconnection of all the six generators and
gave severe impact on the ZESA system in neighboring Zimbabwe. The 1994 Kariba South
Power Station bus-bar fault that tripped out Kariba North Bank Power Plant. The Kabwe
step down bus-bar fault, when Kafue Gorge power station was lost and the system
survived. The table below depicts incidents of loss of supply in the past 20 years.
Table 1: Loss of supply in the past 20 years.

ITE
M
NO.
1.

INCIDE
NT
DATE

LOCATIO
N

DESCRIPTIO
N
OF
INCIDENT

INCIDENT
DURATION
&
LOAD
LOST

COMMENT

26/03/19
89

Kafue
Gorge
Power
Station

The
power
station
experienced
extensive
damage due to
fire

Recommission

1.ZESA
power
system was
affected

G12/12/1989
G2-2/12/1989
G3-24/2/1990
G4- 6/2/1990
G5-8/8/1991

2.
Import
support from
neighbors
Zimbabwe
and Congo
DR

G610/12/1991
2.

1994

Kariba
South
Power
station

Bus bar fault

in the switch
yard

N/A

3.

14/11/19
97

Kafue West
Leopards
Hill 330Kv
line

Collapsed
towers on the
Kafue WestL/Hill 330kV
line.

15:01
15:14hrs

Kafue Gorge
power station
and all lines
from
power
station tripped
following
attempted high
62

Tripping of
Kariba North
power
station
to

SAPP
frequency
dropped to
45.7Hz. Tie
lines
to
ZESA
tripped on
under
frequency.

speed
auto
reclose onto
faulty line

4.

5.

17/12/20
05

24/12/20
05

Kabwe Step
Down
Substation

Bus bar fault

Loss
Kafue
Gorge

Mudslide
20:52hrs

of

17:25
18:13hrs;

to

Loss
of
whole
Copper belt
and tie line
to DRC.

First
two
generators
restored
at
07:22hrs on
25/12/2005.
Third
generator on
29/12/2005.
Fourth
generator on
23/01/2006

No load loss.
Import
support from
SAPP

530 MW

63

at

Figure 52: Single line diagram of the Zambia Power System

64

Appendix6: Unit Commitment

Table 2: Unit commitment

65

Appendix 7: Reactive power compensation and SVC Models

Table 3: Reactive power compensation and SVC Models

Figure 53: Model3 of SVC and its parameters

66

Appendix8: Loading on the lines at peak load

Name

Loadin
g

Active
Power

Total
Active
Power

Power
Factor

Active
Power

Reactive
Power

Terminal
i in MW

Terminal
i in MW

Terminal
i

Terminal
j in MW

Terminal
i
in
Mvar

ChambKitwe 330

8.8018
4

7.54172
6

7.54172
6

0.14700
49

7.57042
5

50.7451
8

ChambLuano 330

5.0636
4

12.6161
6

12.6161
6

0.36562
18

12.6222

32.1169
7

CosakKitwe 220
A1

3.8441
7

12.2688
5

12.2688
5

0.99306
3

12.2574
9

1.45269
3

CosakLuano 220
1A

11.347
9

37.9737
8

37.9737
8

0.99723
68

-37.9425

2.82882
3

Coventry
StreetLeopards
Hill 132
1A

112.72
2

100.829
6

100.829
6

0.85659
73

105.119
6

-60.7364

Coventry
StreetLusaka
West 132
1A

36.593
6

32.6993
9

32.6993
9

0.85571
93

32.8125
9

19.7728
2

KabweKitwe 330
1A

30.222
6

191.565
5

191.565
5

0.93433
34

194.555
4

73.0723
9

KabweKitwe 330
2A

30.222
6

194.555
4

194.555
4

0.99351
3

191.565
5

22.2690
7

KabweLeopards
Hill 330
1A

34.629
6

246.087
1

246.087
1

0.99915
24

248.085
9

10.1387
7

KabweLeopards
Hill 330

34.629
6

248.085
9

248.085
9

0.99926
47

246.087
1

9.51910
6

67

2A
KabweLeopards
Hill 330
3A

34.629
6

246.087
1

246.087
1

0.99915
24

248.085
9

10.1387
7

KabweLuano 330
1A

26.107
9

167.538
9

167.538
9

0.99996
06

165.001
6

1.48695
9

KabweLuano 330
2A

26.107
9

165.001
6

165.001
6

0.92593
54

167.538
9

-67.3029

KabweMulugushi
66 1A

60.024
4

28.5431
2

28.5431
2

0.99994
37

26.9979
7

0.30296
44

KabwePensulo
330 1A

11.268
3

15.4599
3

15.4599
3

0.40689
37

-15.303

-34.7075

Kafue
GorgeKafue
West 330
1A

33.925
5

229.260
1

229.260
1

0.95046
32

228.427
5

74.9766
9

Kafue
GorgeLeopards
Hill 330
1A

46.143
8

323.266
6

323.266
6

0.97695
74

324.969
5

70.6236
1

Kafue
GorgeLeopards
Hill 330
2A

46.143
8

324.969
5

324.969
5

0.97926
62

323.266
6

67.2254
3

Kafue
TownKafue
West 330
1A

10.316
6

48.1657
3

48.1657
3

0.65812
63

-48.1603

55.1024
1

Kafue
TownMapepe 88
1A

73.878
7

36.6559
8

36.6559
8

0.82585
29

38.8257
7

-25.0284

Kafue
TownMazabuka
88 1A

63.163
8

30.1838
3

30.1838
3

0.80452
65

27.6941
1

22.2822
7

68

Kafue
TownMuzuma
220 1A

16.075
8

37.2707
9

37.2707
9

0.96730
94

36.7168
1

9.77126
6

Kafue
TownNampund
we 88 1A

66.451

25.3492
6

25.3492
6

0.83825
46

23.2093
8

16.4895
2

Kafue
WestLeopards
Hill 330
1A

14.149
3

100.572
2

100.572
2

0.99820
39

100.752
3

6.03596
1

Kafwe
WestLusaka
West 330
1A

13.794
6

79.5095
5

79.5095
5

0.88209
55

79.3814
7

42.4609
5

Kansanshi
-Luano
330 1A

28.357
1

153.393
7

153.393
7

0.79251
87

151.151
3

118.037
3

Kansanshi
-Lumwana
330_1

12.363
4

59.9999
9

59.9999
9

0.84453
48

60.1513
3

38.0446
6

KansuswaKitwe 220
1A

21.065
8

62.0513
4

62.0513
4

0.90306
02

-61.8534

29.5128
9

KansuswaLuano 220
1A

18.355
2

53.5284
2

53.5284
2

0.87370
11

53.6963
5

29.8037
8

KansuswaMufulira
66 1A

39.658
4

31.4579
8

31.4579
8

0.89082
49

31.2724
1

16.0444
9

KaomaMongu 66
1A

5.6858

0.93826
67

0.93826
67

-0.94

0.96224
68

0.34054
54

Kapiri
MposhiMpongwe
88 1A

4.2614
7

2.81577
4

2.81577
4

0.99556
17

2.80367
2

0.26617
82

Kariba
NorthKariba
South 330

14.294
3

105.003
5

105.003
5

0.99999
36

-105

0.37676
28

69

1A
Kariba
NorthKariba
South 330
2A

14.294
3

105.003
5

105.003
5

0.99999
36

-105

0.37676
28

Kariba
NorthLeopards
Hill 1A

27.044
2

184.396
5

184.396
5

0.98853
98

182.951
1

28.1593
1

Kariba
NorthLeopards
Hill 2A

27.044
2

182.951
1

182.951
1

0.94338
42

184.396
5

64.3271
7

KasamaLubushi T
66 1A

8.4560
4

-2.64167

-2.64167

0.66085
41

2.69318
4

3.00007
5

KasamaMbala 66
1A

4.0368
1

1.59504
9

1.59504
9

0.71327
2

1.57561
9

1.56735
4

KasamaMpika 66
1A

6.1263

2.19033
5

2.19033
5

0.99722
92

2.24681
3

0.16339
24

Katima
MuliloSesheke
66 1A

13.518
2

5.33754
1

5.33754
1

0.92859
97

5.37388
9

2.13296
5

Kawambw
aMporokos
o 66 1A

3.9943
5

0.45299
75

0.45299
75

0.95830
56

0.46342
35

0.13507
36

KazungluVictoria
Falls
66
1A

11.393
5

6.04020
1

6.04020
1

0.94580
43

5.92034
6

2.07387
4

Kazungula
-Sesheke
66 1A

5.5309
2

1.89116
5

1.89116
5

0.93056
22

1.92200
4

0.74408
75

KitweLuangwa
66 1A

10.617

5.41154
9

5.41154
9

0.95466
33

5.39999
9

1.68745
6

KitweMaposa
220 1A

21.768
9

61.9711
5

61.9711
5

0.85670
64

61.8877
8

37.3115

70

KitweMaposa
220 2A

21.768
9

61.8877
8

61.8877
8

0.84722
92

61.9711
5

38.8047
3

KitweMill 66 1A

42.668
3

27.6958
2

27.6958
2

0.71102
07

27.4905
4

27.3900
6

Leopards
HillMapape 88
1A

88.765
2

51.7938
5

51.7938
5

0.80855
6

49.0371
9

37.6924
9

Leopards
Hill-Roma
132 1A

101.21
5

92.1358
2

92.1358
2

0.86446
69

95.5928
5

53.5772
3

Leopards
HillWaterwor
ks 88 1A

90.677
3

55.0707
3

55.0707
3

0.84001
22

-52.92

35.5704
2

Leopards
HillWaterwor
ks 88 2A

96.964
4

50.5431
1

50.5431
1

0.72158
14

48.0908
1

48.4941
8

LuanoMichelo-T
220 1A

43.869
5

109.745
8

109.745
8

0.75628
32

108.115
6

94.9387
8

MaambaMuzuma
88 1A

19.438
1

11.0174
6

11.0174
6

0.85532
36

10.7766
2

6.67359
4

MuzumaVictoria
Falls 220
1A

20.576

58.2286
2

58.2286
2

0.92521
46

59.1219
4

23.8804
2

PensuloSerenje 66
1A

2.1577
8

1.12669
1

1.12669
1

0.98691
69

1.12445
8

0.18406
42

SenangaSesheke
66 1A

15.632
6

3.67272
9

3.67272
9

0.77202
73

4.06520
7

3.02365
9

ShesekeVictoria
Falls
220_1

6.6872
5

11.9845
5

11.9845
5

0.63100
39

12.0530
5

14.7342
5

71

Table 4: Loading on the lines at peak load

Table 5: Load flow with Kabwe-Kitwe line out at peak load

72

Appendix 9: Load flow calculations on N-1

Table 6: Load flow with Kariba north and LHill line out (Load reduced to account for
increased loses).

73

74

Table 7: Voltage profile with Kariba North- LHill line out.

75

Table 8: Load flow with KFG-LHil line out.

76

Table 9: Voltage profile with KFG-LHil line out

77

Table 10: Load flow with KFG-Kafue west line out.

See voltage profile below with the Kafue Gorge- Kafue West Line out

78

79

Table 11: Load flow with Lhill-Kabwe line out (load reduced to 98% to account for
increased transmission loses).

See voltage profile below

80

81

Table 12: Load flow calculations with Kabwe- Kitwe line out (load reduced to 93% to
account for increased losses).
See corresponding voltage profile below.

82

83

Table 13: Load flow calculation with Kabwe-Luano line out.

See corresponding voltage profile below

84

85

86

Appendix 10. Detailed parameters of the AVRs

Table 14: Detailed parameters of the AVRs

87

Appendix 11. Detailed parameters of speed Governors.

Table 15: Detailed parameters of speed Governors.

88

Appendix 12. Transient Simulations and results

Figure 54: Simulated fault on KFG-Lhill Line close to KFG HPP

89

Figure 55: Voltage magnitude and rotor angle for a fault close to KFG HPP

90

Figure 56: Close-up fault to Kafue gorge bus bar on KFG-LHill line

91

Figure 57: Close-up fault to Kafue Gorge bus bar on KFG-LHill line

92

Figure 58: Auto reclosing on KFG-LHill line

93

Figure 59: Auto reclosing on KFG-Lhill line

94

Figure 60: Auto reclosing on KFG-LHill line

95

Figure 61: Clearing time at 300ms for a close up fault to KFG HPP on KFG-LHILL Line (
Voltage magnitude and angle difference)

96

Figure 62: Clearing time at 300ms (Rotor angle, Voltage angle) for a KFG HPP close up
fault.

97

Figure 63: Clearing time at 300ms for a close up fault at KFG HPP ( Active power,Voltage
magnitude and frequency)

98

Figure 64: Close-up fault to Kariba bus bar on KNB-LHIll line

99

Figure 65: Close-up fault to Kariba bus bar on KNB-LHIll line

100

Figure 66: Close-up fault to Kariba bus bar on KNB-LHIll line

101

Figure 67: Auto reclosing at Kariba end then at LHill end

102

Figure 68: Auto reclosing at Kariba end then at LHill end

103

Figure 69: Auto reclosing Kariba LHill line

104

Figure 70: Fault on LHill-Kabwe line

105

Figure 71: Fault on LHill Kabwe line

106

Figure 72: Fault on LHill-Kabwe line

107

Figure 73: LHill-Kabwe fault 400ms clearing time.

108

Figure 74: LHill-Kabwe fault and auto reclosing

109

Figure 75: LHill Kabwe fault and auto reclosing

110

Figure 76: Kabwe Kitwe line fault

111

Figure 77: Kabwe-Kitwe line fault

112

Figure 78: Kabwe-Kitwe line fault

113

Figure 79: Kabwe-Kitwe auto reclosing

114

Figure 80: Kabwe-kitwe auto reclosing

115

Figure 81: Kabwe Kitwe line auto reclosing

116

Figure 82: Fault on Kabwe pensulo line

117

Figure 83: Kabwe-pensulo line fault

118

Figure 84: Kabwe-pensulo line fault

119

Figure 85: KFG-Kfue west fault (close to Kafue west)

120

Figure 86: KFG-Kfwest line fault

121

Figure 87: KFG-KFWest line fault

122

Figure 88: KFG-KFWest auto reclosing

123

Figure 89: KFG-KFWEST line auto reclosing

124

Figure 90: KFG-KFWest line auto reclosing

125

Figure 91: Fault on Luano-Kansanshi line

126

Figure 92: Fault on Luano-Kansahi line

127

Figure 93: Fault on Luano Kansanshi line

Figure 94: Reactive Power, Excitation Voltage and current of the Generating Machines for
the fault close to KNBC HPP

128

Figure 95: Reactive power, excitation voltage and current of the machines for a fault close
to KFG HHP

Figure 96: Power swing for a fault close to KFG HHP

129

Figure 97: Power Swing for a fault close to KNBC HPP

Figure 98: Excitation voltage after a fault close to Vic Falls

130

Figure 99: Power swing for a fault close to Vic. falls on Kariba HPP Lhill Line and KFG
HPP - LHill Line
.

131