1

CHAPTER-1
INTRODUCTION
1.1 OVERVIEW
A power system is a complex network comprising of numerous
generators, transmission lines, variety of loads and transformers. As a
consequence of increasing power demand, some transmission lines
are more loaded than was planned when they were built. The
increasing cost of addition and modifications has made it imperative
that utilities consider a range of design options, and perform detailed
studies of the effect on the system of each option, based on a number
of assumptions: normal and abnormal operating conditions, peak and
off-peak loadings, and present and future years of operation. One of
the important studies that must be included in designing an electric
power system is Transient Stability. So with increased power
transfer, transient stability is increasingly important for secure
operation. Transient stability evaluation of large scale power systems
is an extremely intricate and highly non-linear problem.
An important function of transient evaluation is to appraise the
capability of the power system to withstand serious contingency in
time, so that some emergencies or preventive control can be carried
out to prevent system breakdown. In practical operations correct
assessment of transient stability for given operating states is
necessary and valuable for power system operation.
Transient stability entails the evaluation of a power systems
ability to withstand large disturbances, and to survive transition to a
normal operating condition. These disturbances can be faults such as:
a short circuit on a transmission line, loss of a generator, loss of a
load, gain of load or loss of a portion of transmission network. The
resulting system response involves large excursions of generator rotor
angles and is influenced by the nonlinear power angle relationship.

2

Stability depends upon both the initial operating conditions of the
system and the severity of the disturbance.
Transient stability studies place an important role in power
systems Which provide information related to the capability of a power
system to remain in synchronism during major disturbances resulting
from either the loss of generation or transmission facilities, sudden or
sustained load changes, in the voltages, currents, powers, speeds and
torques of the machines of the power systems as well as the changes
in system voltages and power flows, during and immediately following
the disturbance.
The transient stability of a generator depends on the difference
between mechanical and electrical power. During a fault, electrical
power is reduced suddenly while mechanical power remains constant,
thereby accelerating the rotor. To maintain transient stability, the
generator must transfer the excess energy towards the system.
The transient stability of a system also depends on the critical
clearing time of the fault, which in turn depends on the distance
between the fault location and the generating stations. The maximum
allowable value of the clearing time for the system to remain stable is
known as critical clearing time. So for the system to be stable the fault
on the system must be cleared before the critical clearing time.
For most of the faults in a multi-machine system, it was
observed that only one machine (or a small group of machines)
becomes severely disturbed and is called the critical machine (or
critical group). The critical machine (or critical group) is usually
responsible to initiate instability for an unstable situation.
Large number of simulations are carried out regularly during
planning stages to gain knowledge of this system. Yet, even a well
designed and normally operated system may face the threat of
transient instability. The aim of the investigation is to analyze the
behaviour of the synchronous machine in particular the angular
position of the rotor with respect to time after the fault occurs in the
system.

3

1.2 LITERATURE SURVEY
Transient stability is an important factor to be considered in
designing any power system. Transient stability of a system refers to
the stability when subjected to large disturbances such as faults and
switching of lines. Stability depends upon both the initial operating
conditions of the system and the severity of the disturbance [1]. There
are two popular numerical methods for solving the power-flow
equations. These are the in Gauss-Seidel (G-S) and the Newton-
Raphson (N-R) Methods. In this paper load flow study is carried out by
using Newton-Raphson method [2,3].
A power system can be considered as first swing stable if the
angles of all machines in centre of angle reference frame increases
(decreases) until a peak (valley) is reached where the angle starts
returning to the stable operating point [4,5]. For performing a stability
study, the necessary starting point is to determine the description of
the differential equations for the system. These differential equations
can be solved by using RANGA-KUTTA method. Here The modelling
and the transient stability analysis of the IEEE 14 test bus system is
done by using Matlab Power System Toolbox (PST) package [6,7]. A
first swing stable system can be considered as stable if the system has
adequate damping in subsequent swings. It has been found that, for a
small disturbance, continuous control is very effective in improving
damping. Usually the machine speed is used as an input signal to
change the mode of operation of the device [9]. During disturbances,
the machines of each group swing more or less together, that is they
retain approximately their relative angular position although these
vary greatly with respect to the machines of the other group. For
purposes of analysis the machines of each group can be replaced by
one equivalent machine [10].




4

1.3 PROBLEM FORMULATION
In transient stability studies a load flow calculation is made first
to obtain system conditions prior to disturbance. In this calculation
the network is composed of system buses, transmission lines and
transformers. The network representation for transient stability
studies includes, in addition to those components, equivalents circuits
for machines and static impedances or admittances to ground for
loads. After the load flow calculation, therefore, the impedance or
admittance matrix of the network must be modified to reflect the
changes in the representation of the network. Two first order
differential equations are required for the simplest representation of
synchronous machines. A transient stability analysis is performed by
combining a solution of the algebraic equations describing the
network with a numerical solution of the differential equations.
The solution of the network equations retains the identity of the
system and there by provides access to system voltages and currents
during transient period. The RUNGA-KUTTA method has been applied
to the solution of the differential equations in the transient stability
studies. In this thesis we can use RUNGA-KUTTA method for the
solution of differential equations in transient stability studies.
1.4 OBJECTIVE OF THE THESIS
The objective of this thesis is to develop a model of the power system
network for doing the load flow analysis of the system. The transient
stability of the power system can be done by considering a three phase
fault on the system. The objective is to analyze the behaviour of the
synchronous machine in particular the angular position of the rotor
with respect to time after the fault occurs in the system. The solution
of swing equation gives the relationship between the angular position
of the rotor and time. The effect of distance between the fault location
and the generating station is studied which affect the critical clearing
time.

5

1.5 ORGANIZATION OF THE THESIS
The organization of this thesis is as follows:
In chapter-1, it is discussed about the introduction of transient
stability, critical clearing time, literature survey, and objectives of this
thesis and organization of this thesis.
In chapter -2, it is discussed about definition, modeling of power
system.
In chapter-3, it is discussed about the factors influencing
transient stability, methods for improving transient stability limit,
swing equation, load flow studies, simulation of a fault, numerical
integration of swing equation by RUNGA-KUTTA method, and stability
analysis for single machine system and multi machine system.
In chapter-4, it is discussed about the results and analysis,
swing curves of IEEE 14-bus test system are plotted and power angle
characteristics are discussed.
In chapter-5, the conclusion and future scope of the same work
has been discussed.















6

CHAPTER-2
MODELING OF POWER SYSTEM
2.1. INTRODUCTION
Power system stability is the ability of an electric power system,
for a given initial operating condition, 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.
This Chapter Deals with modeling of power system i.e.,
representation of synchronous machine model,rotor and stator
winding equations, representation of loads.
2.2 MODELING OF THE SYSTEM
2.2.1 Synchronous machine model
Mathematical models of a synchronous machine vary from
elementary classical models to more detailed ones. In the detailed
models, transient and sub transient phenomena are considered. Here,
the transient models are used to represent the machines in the
system, according to following equations.
2.2.1.1 Rotor winding equations
To represent transient effects two rotor circuits, one field
winding on the d-axis and a hypothetical coil (damper winding) on the
q-axis are adequate.
'
' ' '
( )
q
do q fd d d d
dE
T E E X X i
dt
!
---(2.1)
'
' ' '
( )
d
qo d q q q
dE
T E X X i
dt
! ---(2.2)


7

Where
Tdo is the d-axis open circuit transient time constant
Tqo is the q-axis open circuit transient time constant
Efd is the field voltage
2.2.1.2 Rotor Mechanical Equations
The rotor mechanical dynamics are represented by the swing
equation
2
m
m e m
dS
H T T DS
dt
! ---(2.3)
m b
d
S
dt
H
[ !
---(2.4)
Where Sm is slip, b is the base synchronous speed and D is the
damping coefficient. Tm is the mechanical torque input and Te is
electrical torque output and is expressed as
' ' ' '
( )
e q q d d d q d q
T E i E i X X i i !
---(2.5)
2.2.1.3 Stator winding equations
Stator transients are neglected and the stator reduces to simple
impedance with reactance components in the d- and q- axes. The
stator is represented by dependent current source IG in parallel with
the admittance YG. The equivalent circuit is shown in fig.2.1.
Ia

I
G


Y
G





Fig. 2.1. Stator representation

8

The YG and IG are defined as:
'
1/ ( )
G a d
Y R JX ! ----(2.6)
Where
' ' '
[ ( )]
j
G G q d dc
I Y E j E E e
H
! ----( 2.7)
' '
( )
dc d q q
E X X i ! ---(2.8)
The synchronous machine is represented by a voltage source, in
back of a transient reactance, that is constant in magnitude but
changes in angular position. This representation neglects the effect of
saliency and assumes constant flux linkages and a small change in
speed. If the machine rotor speed is assumed constant at
synchronous speed, a normal and accepted assumption for stability
studies, then M is constant. If the rotational power losses of the
machine due to such effects as windage and friction are ignored, then
the accelerating power equals the difference between the mechanical
power and the electrical power [6]. The classical model is shown in
fig2.2 and is described by the following set of differential and algebraic
equations:
Differential:

2
2
2
m e
d
f
dt
d d f
P P
dt dt H
H
[ T
H [ T
!
! !
---(2.9)

Algebraic:
' '
t a t d t
E E r I jx I !
---
(2.10)

Where
E=voltage back of transient reactance
Et=machine terminal voltage
It=machine terminal current
ra=armature resistance

9

'
d
x =Transient reactance

Fig.2.2: Generator Classical model

2.2.2 Representation of Loads
Power system loads, other than motors represented by
equivalent circuits, can be treated in several ways during the transient
period. The commonly used representations are either static
impedance or admittance to ground, constant real and reactive power,
or a combination of these representations. The parameters associated
with static impedance and constant current representations are
obtained from the scheduled busloads and the bus voltages calculated
from a load flow solution for the power system prior to a disturbance
[5]. The initial value of the current for a constant current
representation is obtained from
*
lp lp
po
p
P jQ
I
E

!
---(2.11)

The static admittance Ypo used to represent the load at bus P,
can be obtained from
po
po
p
I
Y
E
!
---(2.12)

Where
Ep is the calculated bus voltage
Plp and Qlp are the scheduled busloads.

10

Diagonal elements of Admittance matrix (Y Bus) corresponding
to the load bus are modified using the Ypo.

V
I
G


Y
L



Fig. 2.3. Load representation

2.3 REPRESENTATION OF BUSES
All the buses of the power system network are generally
classified into three categories viz. generation bus, load bus and the
slack bus. Two of the four quantities as mentioned above, are
specified at each of the buses as shown below:
1. Generation Bus: This is also called as the P-V bus and here the
voltage magnitude |V| and the real power |P| are specified.
2. Load Bus: This is also called as the P-Q bus and here real power
|P| and reactive power |Q| are specified.
3. Slack or Swing Bus: This is also known as the reference bus and
the voltage magnitude |V| and phase angle are specified here. This
bus is selected to provide additional real and reactive power to supply
the transmission losses since these are unknown until final solution
is obtained. If slack bus is not specified, then a generation bus usually
with maximum real power |P| is taken as a slack bus.



11

2.4. NETWORK REPRESENTATION
A power transmission network can be represented by the
algebraic equation:
[I]=[Y][V] ---(2.13)
[Y] is network admittance matrix. It is sparse, complex, and
symmetrical, and constant in between disturbance such as faults and
line switching operations. YG and YL being constants are also included
in the network admittance matrix [Y]. [V] is vector of bus voltages and
[I] is the vector of current injections.
In general
[I]=
G
L
D
I
I
I





---(2.14)
Where IG is the generator current injection, IL is the current due
to the loads (IL=0 for constant impedance loads) and ID is the current
injection due to the additional network based devices like SVC,
STATCOM, TCSC, UPFC etc which may be incorporated in to the
system.
2.5. SUMMARY
In this chapter modeling of power system and for transient stability
evaluation is presented. Algorithmic procedure for transient stability
evaluation is presented in the following chapters. Simulation steps of the
transient stability analysis program and results of IEEE 14 bus test system
are presented in the following chapters.





12

CHAPTER-3
LOAD FLOW FORMULATION

3.1 INTRODUCTION

Load flow studies are one of the most important aspects of
power system planning and operation. The load flow provides the
sinusoidal steady state of the entire system - voltages, real, reactive
powers and line losses. It provides solution of the network under
steady state condition subjected to certain inequality constraints such
as nodal voltages, reactive power generation of the generators. Load
flow studies gives the voltage magnitudes and angles at each bus in
the steady state. This is rather important as the magnitudes of the
bus voltages are required to be held within a specified limit. Once the
bus voltage magnitudes and their angles are computed, the real and
reactive power flow through each line can be computed. Based on the
difference between power flow in the sending and receiving ends, the
losses in a particular line can also be computed.

3.2 CLASSIFICATION OF BUSES

The system consists of three basic individual components:
generators, transmission lines and loads. The system buses can be
classified as slack bus, generator or voltage controlled buses and load
buses. There are four system variables namely real power injection

, reactive power injection

, voltage magnitude

and phase
angle

are used to characterize each bus i; usually for each bus two
system variables are controllable and other two are uncontrollable. A
slack or swing bus, voltage controlled buses and load buses are
distinguished by different set of controllable variables.


13

Slack or Swing bus: Controllable variables are voltage magnitude (V),
and phase angle ().
Load buses: Controllable variables are real and reactive power
injection

and

respectively.
Voltage controlled buses: Controllable variables are real power
injection,

and voltage magnitude

.

3.3 Y BUS FORMULATION

For an interconnected system if the value of admittance for each
interconnecting circuit is known then the admittance matrix may be
assembled as follows:
The diagonal element of each node is the sum of the admittance
connected to it.
The off-diagonal element is the negated admittance between the
nodes.
If any shunt branch is present then shunt admittance is added to
the diagonal elements corresponding to the nodes at which they are
connected. Admittance Y is used to describe the topological properties
of transmission network.
The entries of Y,

are given by:

= -

---(3.1)
where

is the admittance of transmission line pq as, Y provides a

convenient notation of relating various system variables.

3.4 LOAD FLOW SOLUTION
The load flow problem models the nonlinear relationships
among bus power injections, power demands, and bus voltages and
angles, with the network constants providing the circuit parameters. It
is the heart of most system-planning studies and also the starting
point for transient and dynamic stability studies. This section

14

provides a formulation of the load-flow problem and its associated
solution strategies. There are two popular numerical methods for
solving the power-flow equations. These are the Gauss-Seidel (G-S)
and the Newton-Raphson (N-R) Methods. The N-R method is superior
to the G-S method because it exhibits a faster convergence
characteristic. However, the N-R method suffers from the
disadvantage that a flat start is not always possible since the
solution at the beginning can oscillate without converging toward the
solution. In order to avoid this problem the load-flow solution is often
started with a G-S algorithm followed by the N-R algorithm after a few
iterations. There is also an approximate but faster method for the
load-flow solution. It is a variation of the N-R method called the fast-
decoupled method, which was introduced by Stott and Alsac (1974).
3.5 DEVELOPMENT OF LOAD FLOW EQUATIONS
The nodal current equations for n bus system are given by

. --- (3.2)

--- (3.3)
Now

--- (3.4)
or

--- (3.5)

--- (3.6)

has been substituted by the real and reactive powers because

normally in a power system these quantities are specified.



15

3.5.1 NEWTON RAPHSON LOAD FLOW FORMULATION IN
CARTESIAN FORM
We know that at any bus p,
Pp-jQp = Vp
*
Ip = Vp
*

--- (3.7)
Vp = ep+jfp and Ypq = Gpq-jBpq --- (3.8)
Pp-jQp = (ep+jfp)
*

jf

--- (3.9)
=(ep-jfp)

jf

--- (3.10)
Separating the real and imaginary parts we have
Pp =

f

B

--- (3.11)
Qp = f

f

B

--- (3.12)
|Vp|
2
= ep
2
+fp
2
---(3.13)
Jacobian form for PQ buses

---(3.14)
Jacobian form for PQV buses

---(3.15)
In general, jacobian form is

---(3.16)

16

The off-diagonal elements and the diagonal elements of J1 i.e.

respectively are

= epGpq-fpBpq , qp ---(3.17)

= 2epGpp+fpBpp-fpBpp+

f

B

---(3.18)
=2epGpp+

f

B

---(3.19)
The off-diagonal elements and the diagonal elements of J2 i.e.

respectively are

pBpq+fpGpq , qp ---(3.20)

= 2fpGpp+ f

---(3.21)
The off-diagonal elements and the diagonal elements of J3 i.e.

respectively are

= epBpq+fpGpq , qp ---(3.22)

= 2epBpp - f

---(3.23)
The off-diagonal and diagonal elements of J4 i.e.

respectively are

= - epGpq+fpBpq qp ---(3.24)

= 2fpBpp+

f

B

---(3.25)


17

The off-diagonal and diagonal elements of J5 i.e.

respectively are

= 0, qp ---(3.26)

= 2ep ---(3.27)
The off-diagonal and diagonal elements of J6 i.e.

respectively are

= 0, qp ---(3.28)

= 2fp ---(3.29)
3.5.2 NEWTON RAPHSON LOAD FLOW FORMULATION IN POLAR
FORM
Power flow equations in polar form
S = VI
*
---(3.30)
Vp = |Vp|p ---(3.31)
Vq = |Vq|q ---(3.32)
Ypq = Gpq+jBpq ---(3.33)
S = Vp

---(3.34)
= (|Vp|p)

u jB

---(3.35)
= cos jsin cos jsin

jB

---(3.36)
Pp = u

cos B

sin

---(3.37)
=Vp
2
Gpp+ u

cos B

sin

---(3.38)


18

Qp = u

sin B

cos

---(3.39)
=-Vp
2
Bpp+ u

sin B

cos

---(3.40)
Jacobian form

---(3.41)
The off-diagonal and diagonal elements of

respectively are

= |Vp||Vq| [Gpqsin

Bpqcos

] ---(3.42)

=|

|
2
Bpp- u

sin B

cos

---(3.43)
= -Qp-|

|
2
Bpp ---(3.44)
The off-diagonal and diagonal elements of

respectively are

= |

||

|[Gpqcos

Bpqsin

] ---(3.45)

= 2|

|
2
Gpp+ u

cos B

sin

---(3.46)
=Pp+|

|
2
Gpp ---(3.47)
The off-diagonal and diagonal elements of

respectively are

= -|

||

| [Gpqcos

Bpqsin

] ---(3.48)
= -

---(3.49)

19

= -|

|
2
Gpp+ u

cos B

sin

---(3.50)
= Pp-|

|
2
Gpp ---(3.51)
The off-diagonal and diagonal elements of

respectively are

= |Vp||Vq| [Gpqsin

Bpqcos

] ---(3.52)
=

---(3.53)

= -2|

|
2
Bpp+ u

sin B

cos

---(3.54)
= Qp-|

|
2
Bpp ---(3.55)

3.6 ALGORITHM FOR LOAD FLOW SOLUTION

The algorithm for Newton-Raphson load flow method is as follows:
Step1: Read system line data, bus data and limits for reactive power
and voltage.
Step 2: Form Y bus using line data.
Step 3: Assume the initial values of voltage magnitudes

of all load
buses and angles

of the voltages of all the buses except the slack

bus.
Step 3: Set convergence criteria value as =0.0001
Step 4: Set bus count i=2.
Step 5: Calculate real and reactive powers

and

as follows
Pp = Vp
2
Gpp+ u

cos B

sin

Qp = -Vp
2
Bpp+ u

sin B

cos

20

Also calculate the mismatch powers (i.e. Pp and Qp)
Pp=

- Pp
Qp=

-Qp
Step 6: Check the type of bus, if the bus is generator bus then check
the reactive power limits. If the value exceeds the limit, fix the reactive
power generation to the corresponding limit.
Step 7: Increase bus count i= i+1 and go to step 5.
Step 8: determine the largest of the absolute of the mismatch power
and if it is less than the convergence criterion then compute line flows
and slack bus power and terminate the process.
|

|
|

|
Step9: Compute elements of Jacobian matrix.
Step 10: update voltage and phase angles
p = p +p
|Vp|= |Vp| + |Vp|
Step 11: Evaluate cos and sin of all voltages.
Step 12: Check for voltage limits and set the voltages to their limits.
Step 13: advance the iteration counter K=K+1 and go to step 4.
Step 14: print the voltages, phase angles and line losses.

3.7 SUMMARY

This chapter mainly explains about load flow solutions and the
methods used to solve load flows. Also this chapter explains the
algorithm implemented to solve load flow problem according to Newton
Raphson method.






21

CHAPTER-4
DEVELOPMENT OF TRANSIENT STABILITY SOLUTION
4.1 INTRODUCTION
Following a disturbance in power system, the rotor angular
position of the machine experiences transient deviations (swings). If it
can be ascertained by the analysis that all the individual rotor angles
will settle down to new post fault steady state values then the system
is considered to be transient stable. The important aspect is the
prediction of the rotor swings accurately which depends upon the
dynamic model of the machine.
This chapter out lines the development of a generalized
transient stability simulation program.
4.2 DESCRIPTION OF TRANSIENTSTABILITY PROBLEM
4.2.1 SWING EQUATION
In order to determine the angular displacement between the
machines of a power system during transient conditions, it is
necessary to solve the differential equations describing the motion of
the machine rotors. The net torque acting on the rotor of a machine,
from the laws of mechanics related to rotating bodies, is
T=W (R
2
) /g --- (4.1)
Where
T=algebraic sum of all torques, ft-lb.
W (R
2
) =Moment of inertia lb-ft
2

g= acceleration due to gravity, equal to 32.2ft/sec
2

= Mechanical angular acceleration, rad/sec
2

The electrical angle e is equal to the product of the mechanical angle m and
the number of pairs of poles P/2, that is,
e =P/2 m --- (4.2)


22

The frequency f in cycle per second is
f= p rpm/(120) ---(4.3)
Then from equations (4.1) and (4.2) electrical angle in radians is
e=60 f m /rpm ---(4.4)
The electrical angular position in radians, of the rotor with
respect to a synchronous rotating reference axis is
= e o t ---(4.5)
Where
o= Rated synchronous speed in rad/sec
t = time in sec
Then, the angular velocity or slip with respect to the reference axis is
d/dt = (de/dt) - o ---(4.6)
and the angular acceleration is
d
2
/ (dt)
2
= d
2
e / (dt)
2
---(4.7)
taking the second derivative of equation(4.4) and substituting,
d
2
/ (dt)
2
= (60 f / rpm) d
2
m / (dt)
2
---(4.8)
Where
d
2
m / (dt)
2
= ---(4.9)
Then, substituting into equation (4.1), the net torque is
T = (W(R
2
rpm) / (g 60 f)) d
2
/ (dt)
2
---(4.10)
It is desirable to express the torque in per unit. The base torque
is defined as the torque required developing rated speed, that is
Base torque =base kVA (550/0.746) / (2(rpm/60)) ---(4.11)
Therefore, the torque in per unit is
T = {{(W R
2
2(rpm/60)
2
(550/0.746))/(gf)} / (base kva)} d
2
/ (dt)
2

---(4.12)
The inertia constant H of a machine is defined as the kinetic
energy at rated speed in kilowatt seconds per kilovolt ampere.
K.E =W(R
2
) (o)
2
/ (2g) ---(4.13)
Where
o= 2 rpm/60 ---(4.14)
and rpm is the rated speed.


23

Therefore
H = ({W(R
2
) / (2g) (2)
2
((rpm/60) )
2
) (0.746/550)} / base kva ---(4.15)
Here T = (H/f) (d
2
/ (dt)
2
) ---(4.16)

The torques acting on the rotor of a generator include the
mechanical input torque form the prime mover, torques due to
rotational losses (Friction, wind age, and core losses) electrical output
torques, and power system.
The electrical and mechanical torques acting on the rotor of the
motor are of opposite sign and are a result of the electrical input and
mechanical load. Neglecting damping and rotational losses, the
accelerating torque Ta is
Ta = Tm Te --- (4.17)
Where
Tm = Mechanical torque
Te = Electrical air gap torque
Thus equation becomes
H/(f) ( d
2
/ (dt)
2
)) = Tm Te ---(4.18)
Since the torque and power in per unit are equal for small
deviations in speed, equation becomes
d
2
/ (dt)
2
= f ( pm pe) / H ---(4.19)
where
pm = Mechanical power
pe = Electrical air gap power
This second order differential equation can be written as two
simultaneous first order equations:
d
2
/ (dt)
2
= d / dt = f ( pm pe) / H ---(4.20)
d /dt = (de / dt )- o ---(4.21)
Since the rated synchronous speed in radians per second is 2f,
equation (2.8) becomes
d /dt = -2f ---(4.22)

24

The swing equation can be represented using the single
synchronous machine connected to infinite bus bars, governed by the
nonlinear differential equation
M d
2
/ dt
2
= pm pe ---(4.23)
Where
pe = pmax sin
=> M d
2
/ dt
2
= pm pmax sin ---(4.24)
4.3 SIMULATION OF FAULTS
A fault at or near a bus is simulated by appropriately changing
the self-admittance of the bus. For a three-phase fault, the fault
impedance is zero and the faulted bus has the same potential as the
ground. This involves placing infinite shunt admittance, so that the
bus voltage is in effect zero. The fault is removed by restoring the
shunt admittance to the appropriate value depending on the post fault
system configuration.
4.3.1 Simulation of fault in a power system studies
A symmetrical fault is simulated in one of the lines at a time.
The simulation is done in three phases:
1. The pre-fault system is run for a small time (say 1 second) till the
system is initialized.
2. The fault is then applied at one end of the line. Simulation of this
faulted condition continues till the line is disconnected from the buses
at both the ends of the faulted line after a time tcl. The time gap
between the tripping of breakers at the two ends is negligible
compared to the clearing time. Hence the disconnection of the line at
the two ends can be considered simultaneous.
3. Next is the post-fault system simulation where the faulted line is
totally disconnected from the system. Simulation is carried out for a
longer time (say 10-20seconds) to observe the nature of the transients.

25

4.3.2 Runga-Kutta method
In the application of the Runga-Kutta fourth-order
approximation, the changes in the internal voltage angles and
machine speeds, again for the simplified machine representation, are
determined from




1 2 3 4
1 2 3 4
1
2 2
6
1
2 2
6
i i i i i t t
i i i i i t t
k k k k
l l l l
H
[
(
(
( !
( !
---(4.25)

i=1,2,,no. of generators.
The ks and ls are the changes in H i and [
i respectively,
obtained using derivatives evaluated at predetermined points. For this
procedure the network equations are to be solved four times.
4.4 TRANSIENT STABILITY ANALYSIS OF MULTI MACHINE
SYSTEM
In multi-machine case, two steps are required, they are:
i. The steady state pre-fault conditions for system are calculated
using load flows.
ii. The pre-fault network representation is determined and then
modified to account for the fault and for the post-fault conditions.
From the first step we know the values for power, reactive power
and voltage at each generator terminal and load bus with all angles
measured with respect to swing bus. The transient internal voltage of
each generator is calculated by using
E
|
= Vt + jXd --- (4.26)
Where
V t Terminal voltage
I Output current
Each load is converted into a constant admittance to ground at
its bus using the equation:
YL = PL-jQL/ | VL |
2
--- (4.27)

26

The bus admittance used for pre-fault load flow calculation is
augmented to include the transient reactance of generator and shunt
load admittance.
The second step determines the modified bus admittance
matrices corresponding to the faulted and post-fault conditions.
To find bus admittance matrix corresponding to the faulted
condition, all the elements of the row and column which correspond to
the faulted bus should be replaced in pre-fault admittance matrix by
zeros and then the reduction method is applied.
To find bus admittance matrix corresponding to the post-fault
condition, the faulted line is isolated first by replacing the admittance
of the faulted line by zero and calculating the diagonal elements of the
buses connected by the faulted line before the fault occurs.
Since only generator internal buses have injections, all other
buses can be eliminated to reduce the size of matrix and matrix size is
equal to number of generators.

For elimination of n
th
bus
Yij(new) = Yij(old) Yin Ynj / Ynn --- (4.28)

During and after the fault the power flow into the network from
each generator is given by
Pei = | Ei| |Ej| |Yij| cos( i-j-ij) --- (4.29)
Where Yij = |Yij|< ij
Yij= Admittance between i
th
and j
th
nodes, the swing equation is given
2Hi d
2
i / s dt
2
= Pa = Pmi - Pe ---(4.30)
to represent the motion of each rotor for during fault and post-fault
periods.
In a multi-machine system a common system base must be
chosen.
Let
Gmech = machine rating (base)

27

Gsystem = system base
Then the swing equation can be written as
(Hsystem/f) (d
2
/ dt
2
) = Pm Pe pu in system base ---(4.31)
Consider the swing equations of n machines on common base
(Heq / f) (d
2
/ dt
2
) = Pm Pe ---(4.32)
Where
Pm = Pm1+Pm2++Pmn
Pe = Pe1+Pe2++Pen
Heq = H1 + H2 ++Hn ---(4.33)
Then machines swinging coherently are thus reduced to a single
machine.
4.5. Network Reduction Technique: (Algorithm)
Data preparation
In the performance of a transient stability study, the following
data are needed;
1. A load-flow study of the pre-transient network to determine the
mechanical power Pm of the generators and to calculate the
values of Eii0 for all generators. The equivalent impedances of
the loads are obtained from the load bus data.
2. System data follows:
(a) The inertia constant H and direct axis transient reactance Xd
for all Generators.
(b) Transmission network impedances for the initial network
Conditions and the subsequent switching such as fault clearing and
breaker reclosings.
3. The type and location of disturbance, time of switching and the
maximum time for which a solution is to be considered.
Preliminary calculations
To prepare the system data for stability study, the following
preliminary calculations are made:
1) All system data are converted to a common base; a system base of
100MVA is frequently used.

28

2) The loads are converted to equivalent impedances or admittances.
The needed data for this step are obtained from the load flow
study. Thus if a certain load bus has a voltage VL, power PL,
reactive power QL, and current IL flowing into a load admittance
YL=GL+j BL, then
PL+j QL=VL IL* = VL {VL*(GL-jBL)} =VL
2
(GL-j BL) ---(4.34)

The equivalent shunt admittance at that bus is given by
YL=PL / VL
2
j ( QL / VL
2
) ---(4.35)

3) The internal voltages of the generators Eii0 are calculated from
the load flow data. These internal angles may be computed from
the Pretransient terminal voltages V as follows. Let the terminal
Voltage be used temporarily as a reference. If we define I=I1+j I2 ,
then from the relation P+j Q = VI* we have I1+j I2 = ( P j Q)/V, But
since E = V+jxdI,
We compute E= (V + QXd/V) + j (PXd/V)
The initial generator angle 0 is then obtained by adding the
Pretransient voltage angle to , or 0= +
4) The Y matrix for each network condition is calculated. The
following steps are usually needed:
a. The equivalent load impedances (or admittances) are connected
between the load buses and the reference node; additional nodes are
provided for the internal generator voltages (nodes 1,2,., n) and
the appropriate values of Xd are connected between these nodes and
the generator terminal nodes. Also, simulation of the fault impedance
is added as required, and the admittance matrix is determined for
each switching condition.
b. All impedance elements are converted to admittances.
c. Elements of the Y matrix are identified as follows: Yii is the sum of
all the admittances connected to node I, and Yij is the negative of the
admittance between node j and I.

29

5) Finally, we eliminate all the nodes except for the internal generator
nodes and obtain the Y matrix for the reduced network. The reduction
can be achieved by matrix operation if we recall that all the nodes
have zero injection currents except for the internal Generator nodes.
This property is used to obtain the network reduction as shown below.
Let
I=YV ---(4.36)

Where I=

Now the matrices Y and V are partitioned accordingly to we get

---(4.37)

Where the subscript n is used to denote generator nodes and
the Subscript r is used for the remaining nodes. Vn has the
dimension (n*1) and Vr, has the dimension (r*1)

Expanding the above matrix form we get
In=YnnVn+YnrVr
0=YrnVn+YrrVr ---(4.38)

From which we eliminate V, to find
In= (Ynn-YnrYrr
-1
Yrn)Vn ---(4.39)

The matrix (Ynn-YnrYrr
-1
Yrn) is the desired reduced matrix Y. It
has the dimensions (n*n) where n is the number of generators. The
network reduction illustrated by above equations is a convenient
analytical technique that can be used only when the loads are treated
as constant impedances. If the loads are not considered to be constant
impedances, the identification of the load buses must be retained.
Network reduction can be applied only to those nodes that have zero
injection current.

30

4.6. SUMMARY
It is discussed about the factors influencing transient stability,
swing equation, load flow studies, simulation of a fault, and numerical
integration of swing equation by RUNGA-KUTTA method, and stability
analysis for single machine system and multi machine system.

























31

CHAPTER-5
RESULTS AND ANALYSIS
5.1. INTRODUCTION
The stability problem of such a power system concerns
transmission of the power from one group of synchronous machines
to another. During disturbances, the machines of each group swing
more of less together, that is they retain approximately their relative
angular position although these vary greatly with respect to the
machines of the other group. For purposes of analysis the machines of
each group can be replaced by one equivalent machine [10].
This chapter deals with the solution for improving transient
stability (algorithm) and the proposed algorithm is tested with single
machine connected with infinite bus system and IEEE 14-bus test
system.
5.2. ALGORITHM
In the performance of a transient stability study, the following
data are needed;
1) All system data are converted to a common base; a system base of
100MVA is frequently used. Form Ybus and run load flow
a) The mechanical power input is taken as (Pm = Pinj ) of the
generators .
b) The loads are converted to equivalent impedances or
admittances. The needed data for this step are obtained from the load
flow study. Thus if a certain load bus has a voltage VL, power PL,
reactive power QL, and current IL flowing into a load admittance
YL=GL+j BL, then
PL+j QL=VL IL* = VL {VL*(GL-jBL)} =VL
2
(GL-j BL) ---(5.1)
The equivalent shunt admittance at that bus is given by
YL=PL / VL
2
j ( QL / VL
2
) ---(5.2)

32

c) The internal voltages of the generators Eii0 are calculated from
the load flow data. These internal angles may be computed from the
pretransient terminal voltages V as follows. Let the terminal voltage
be used temporarily as a reference. If we define I=I1+j I2, then from the
relation P+jQ = VI*. we have,
I1+j I2 = (P j Q)/V ---(5.3)
But since E = V+jxdI
We compute,
E= (V + QXd/V) + j (PXd/V) ---(5.4)
d) The initial generator angle 0 is then obtained by adding the
pretransient voltage angle to , or 0= +
2) The Ytrbus matrix for each network condition is calculated. The
following steps are usually needed:
a) The equivalent load impedances (or admittances) are connected
between the load buses and the reference node; additional nodes
are provided for the internal generator voltages (nodes 1,2,.n)
and the appropriate values of Xd are connected between these
nodes and the generator terminal nodes. Also, simulation of the
fault impedance is added as required, and the admittance matrix is
determined for each switching condition.
b) All impedance elements are converted to admittances.
c) Elements of the Y matrix are identified as follows: Yii is the sum
of all the admittances connected to node I, and Yij is the negative of
the admittance between node j and i.
d) The Y matrix for the reduced network. The reduction can be
achieved by matrix operation if we recall that all the nodes have
zero injection currents except for the internal generator nodes. This
property is used to obtain the network reduction as shown below.
For each load bus
Ytrbus (i, i)=ybus (i, i)+complex (P (i), -Q (i))/(conj (v (i))*v (i))
---(5.5)
For each generator bus
Ytrbus (i, i) =ybus (i, i) +1/complex (ra (i), xdp (i)) ---(5.6)

33

3) System data follows:
(a) The inertia constant H and direct axis transient reactance Xd
for all Generators.
(b) Transmission network impedances for the initial network
conditions and the subsequent switching such as fault clearing and
breaker reclosings.
(c) The type and location of disturbance, time of switching and the
Maximum time for which a solution is to be considered.
(d) Find Ytrbus for various network conditions during fault, post
fault (Faulted line cleared), after line reclosure.
(e) For faulted mode, find generator outputs from power angle
equations and solve swing equations by R.K fourth order
method etc.
Voltages at each bus is obtained by v=inv (ytrbus)*Inor
(f) Keep repeating the above step for post fault mode and after line
reclosure mode.
(g) Examine (t) plots for all the generators and establish the
answer to the stability question.
5.3. DESCRIPTION OF TEST SYSTEM
In this chapter 5 generator 10 bus test system (IEEE 14-bus Model) is
considered for analyzing the effect of fault location in conjunction with
the fault clearing time.
A three-phase fault is considered at two locations in Fig IEEE
14-bus Model. One of them is on line 2-4 to the generating station 2,
which has the smallest inertia value. The second is on line 13-14, very
close to bus 13 and far away from all the generating station. Thus, the
effect of the distance between the fault location and the generating
stations and the effect of fault clearing time is analyzed.



34

5.4. TRANSIENT STABILITY ANALYSIS
Fault on line 2-4
This study is performed with the intention of analyzing the effect
of fault location in conjunction with the fault clearing time.
Two faults located on two different lines are considered one
closer to the generating stations and the other one far from the
generating stations. Now consider fault located closer to the
generating stations.
Fig.5.1. shows the results of the angle differences of the
machines in the system, when fault is occurred on line 2-4.

Fig.5.1. Plots of angle differences for machines 2, 3, 4&5.
Fault on bus 2 faults cleared at 0.4sec

Fig.5.2. shows the results of the angle differences of the
machines in the system, when fault occurred on line 2-4 and fault is
cleared in 0.5 sec.
0 0.5 1 1.5
-150
-100
-50
0
50
100
Phase angle difference (fault c leared at 0. 4s)
t, sec
D
e
l
t
a
,

d
e
g
r
e
e

35


Fig.5.2. Plots of angle differences for machines 2, 3, 4 & 5.
Fault on bus 2 faults cleared at 0.5sec.
Since after 0.4 sec, the machines go out of step. So the fault
should be cleared in 0.4 sec.Fig.2 shows the swing curves for
machines fall out of synchronism if fault is cleared 0.5 sec. so the
critical clearing time here is 0.4 sec. the fault should be cleared in 0.4
sec.
Fault on line 13-14
This study is performed with the intention of analyzing the effect
of fault location in conjunction with the fault clearing time. Two faults
located on two different lines are considered one closer to the
generating stations and the other one far from the generating stations.
Now consider fault located far from the generating stations.
Fig.5.3. shows the results of the angle differences of the
machines in the system, when fault is occurred on line 13-14,
0 0. 5 1 1. 5
-1200
-1000
-800
-600
-400
-200
0
200
P hase an gle di fferenc e (faul t cl eared at 0. 5s )
t , se c
D
e
lta
, d
e
g
re
e

36


Fig.5.3. Plots of angle differences for machines 2, 3, 4&5.
Fault on bus 13 fault cleared at 0.6sec.
There are many factors affecting the critical clearing time. Here,
the effect distance between the fault location and the generating
stations is studied. Two fault locations for the same values of the
damping and inertia constants are considered. One of the faults is on
line (2-4), very close to bus 2 which is connected to machine 2, and
the second fault is on line (10,11), very close to bus 10 . Machines 4
and 5 have the smallest value of inertia, so they are expected to go out
of step first. The fault that is closer to the generating station must be
cleared rapidly than the fault on the line far from the generation
station. Rapid clearing of the faults promotes power system stability.
5.5. SUMMARY
In this chapter solution steps to analyze transient stability
(algorithm) and also swing curves for IEEE 14-bus test system are
plotted. It also includes power-angle characteristics showing
improvement in critical clearing time and critical clearing angle when
the distance between the generating station and fault location is
increased.

0 0. 5 1 1. 5
-20
-15
-10
-5
0
5
10
15
20
Phas e angle difference (fault c leared at 0. 6s)
t , s ec
D
e
l
t
a
,

d
e
g
r
e
e

37

CHAPTER-6
CONCLUSIONS AND FUTURE SCOPE
6.1 INTRODUCTION
The work reported in this thesis explained the development of a
generalized program for transient stability evaluation of large
interconnected power systems.
The stability has determined by plotting the swing equations
which are nonlinear. For a three-phase fault at two different locations
in the system, the reduction method has been applied to determine
the reduced connection matrices Yfault and Ypostfault during and
after clearing the fault. The two different locations of the fault are
chosen to observe the effect of the distance of the fault from the
generators on the stability. It is therefore recommended that power
system designers must do proper inspections on the design with
regard to transient stability as it is the most cumbersome aspect in
power system design.
Here the 5 generator 14 bus test system is considered.

6.2 OBSERVATIONS AND CONCLUSIONS
The developed program is tested for different locations of the
three phase fault and major observations and conclusions which are
drawn from the above test is given below.
y Critical clearing time is the main factor to be considered while
analyzing the transient stability of the power system.
y The critical clearing time can dependent on the distance
between the generating station and fault location.
y The fault that is closer to the generating station must be
cleared rapidly than the fault on the line far from the generation
station.


38

6.3 SCOPE FOR FURTHER WORK
The present work has identified further areas of interest:
y Incorporating FACTS devices in power system network to
improve the transient stability of the system.
y Testing should be done on extremely large real time power
systems.

6.4 SUMMARY
In this chapter, observations, conclusions and future scope of
the works has been briefly discussed.






















39

REFERENCES

[1] P. Kundur, Power System Stability and Control, New York: Mc
Graw-Hill, 1994 pp 104-120.
[2] Glover, J. D., and M. Sharma. 1994. Power System Analysis, and
Design, 2nd ed. Boston: PWS Publishing.
[3] Lynn Powell, Power System Load Flow Analysis, New York:
McGraw-Hill, 2004 pp.24-109.
[4] M.H. Haque and A.H.M.A. Rahim, Determination of first swing
stability limit of a multimachine power system, IEE Proc., Part-C, Vol.
136, No. 6, 1989, pp. 373-379.
[5] P.W. Sauer and M.A. Pai, Power System Dynamics and Stability,
New Jersey: Prentice-Hall, 1998 pp. 161-180.
[6] G. W. Stagg and A. H Fl-Abiad, Computer Methods in Power System
Analysis. New York: McGraw-Hill, 1968 pp. 150-164.
[7] D. Hanselman and B. Littlefield, Mastering Matlab 6, New Jersey:
Prentice Hall, 2001pp. 176-190.
[6] W.D Stevenson, Elements of Power System Analysis. New York,
McGraw-Hill. 1982 pp 402
[7] B.M Weedy and B.J Cory, Electric Power Systems, New York: John
Wiley & Sons, 1998 pp. 95-157
[9] G. Maria, C.Tang, J., Kim, Hybrid Transient Stability Using Energy
Functions, IEEE Trans. 1990 May, Vol. PWRS-5, No. 2, pp. 384-393
[10] Arthur H. Seidman and H. Mahrous, Handbook of Electric Power
Calculations. New York: McGraw- Hill, 1983 pp. 143-150





APPENDICES

40

APPENDIX-A
5 MACHINE 14 BUS TEST SYSTEMS:
GENERATOR DATA:
UNIT R X
q
X
q

X
d
X
d

T
do

T
qo

H D
1 0 0.646 0.646 0.898 0.3 7.4 0 5.148 2
2 0 0.98 0.36 1.05 0.19 6.1 0.3 6.54 2
3 0.003 0.98 0.36 1.05 0.185 6.1 0.3 6.54 2
4 0.001 1.22 0.715 1.25 0.232 4.75 1.5 5.06 2
5 0.004 1.22 0.715 1.25 0.232 4.75 1.5 5.06 2

IEEE 14-bus Model:

12 13 14

11 10

G 9
1 8 C
6 7
C 4
5


2
G
3
Fig.A.1 IEEE 14BUS system C
9
G- GENERATOR
C-SYNCRONOUS
COMPENSATOR


41

7 8 C

4


BUS DATA:
Bus P
Gen(pu)
Q
Gen(pu)
P
Load(pu)
Q
Load(pu)
Bus
Type
Q Gen
max(pu)
Q Gen
min(pu)
1 2.32 0 0 0 2 10 -10
2 0.4 0.424 0.217 0.127 1 0.5 -0.4
3 0 0 0.942 0.19 2 0.4 0
4 0 0 0.478 0 3 0 0
5 0 0 0.076 0.016 3 0 0
6 0 0 0.112 0.075 2 0.24 0.06
7 0 0 0 0 3 0 0
8 0 0 0 0 2 0.24 0.06
9 0 0 0.295 0.166 3 0 0
10 0 0 0.09 0.058 3 0 0
11 0 0 0.035 0.018 3 0 0
12 0 0 0.016 3 0 0 0
13 0 0 0.135 0.058 3 0 0
14 0 0 0.149 0.05 3 0 0
Bus Type:
(1) Swing bus,
(2) Generator bus (PV bus)
(3) Load bus (PQ bus)
LINE DATA:
THREE WINDING TRANSFORMER
EQUIVALENT CIRCUIT

42

FROM
BUS
TO BUS R(pu) X(pu) B(pu) TAP
1 2 0.0194 0.05917 0.053 1
1 5 0.054 0.22304 0.049 1
2 3 0.047 0.19797 0.044 1
2 4 0.0581 0.17632 0.037 1
2 5 0.057 0.17388 0.034 1
3 4 0.067 0.17103 0.346 1
4 5 0.0134 0.04211 0.013 1
4 7 0 0.20912 0 1
4 9 0 0.55618 0 1
5 6 0 0.25202 0 0.9
6 11 0.095 0.1989 0 1
6 12 0.1229 0.25581 0 1
6 13 0.0662 0.13027 0 1
7 8 0 0.17615 0 1
7 9 0 0.11001 0 1
9 10 0.0318 0.0845 0 1
9 14 0.1271 0.27038 0 1
10 11 0.0821 0.19207 0 1
12 13 0.2209 0.19988 0 1
13 14 0.1709 0.34802 0 1






43