Lecture Notes

05PC602
POWER SYSTEM ANALYSIS

VI Sem B.E (E & E)

Dr. P. Aravindhababu
Dr. L. Lakshminarasimman
 05PC602 POWER SYSTEM ANALYSIS

Unit–I : Modelling of Power Systems Components

Representation of power system components : Single phase solution of balanced three phase networks - One line diagram
- Impedance or reactance diagram - Per unit system - Per unit impedance diagram - Complex power - representation
of loads.

Review of symmetrical components - Transformation of voltage, current and impedance (conventional and power
invariant transformations) - Phase shift in star- delta transformers - Sequence impedance of transmission lines -
Sequence impedance and sequence network of power system components (synchronous
machines, loads and transformer banks) - Construction of sequence networks of a power system.

Unit–II : Bus Impedance and Admittance Matrices

Development of network matrix from graph theory - Primitive impedance and admittance matrices - Bus admittance and
bus impedance matrices – Properties - Formation of bus admittance matrix by inspection and analytical methods. Bus
impedance matrix: Properties - Formation using building algorithm - addition of branch, link - removal of link, radial
line - Parameter changes.

Unit–III : Power Flow Analysis

Sparsity - Different methods of storing sparse matrices - Triangular factorization of a sparse matrix and solution using the
factors - Optimal ordering - Three typical schemes for optimal ordering - Implementation of the second method of
Tinney and Walker. Power flow analysis - Bus classification - Development of power flow model - Power flow
problem - Solution using Gauss Seidel method and Newton Raphson method - Application of sparsity based
programming in Newton Raphson method - Fast decoupled load flow- comparison of the methods.

Unit–IV : Fault Analysis

Short circuit of a synchronous machine on no load and on load - Algorithm for symmetrical short circuit studies -
Unsymmetrical fault analysis - Single line to ground fault, line to line fault, double line to ground fault ( with and
without fault impedances ) using sequence bus impedance matrices - Phase shift due to star- delta transformers -
Current limiting reactors - Fault computations for selection of circuit breakers.

Unit–V : Short Circuit Study Based on Bus Admittance Matrix

Phase and sequence admittance matrix representation for three phase, single line to ground, line to line and
double line to ground faults (through fault impedances) - Computation of currents and voltages under faulted condition
using phase and sequence fault admittance models - Sparsity based short circuit studies using factors of bus admittance
matrix.

Text Books
1) Nagrath, I.J., Kothari. D.P., “Power System Engineering”, TMH, New Delhi; 2007.
2) Wadhwa, C.L., “Electric Power Systems”, Wiley Eastern, 2007.

Reference Books
1) Pai, M.A., “Computer Techniques in Power System Analysis”, TMH, 2007.
2) Stagg and El-Abiad, “Computer Methods in Power System Analysis”, McGraw Hill International, Student Edition,
1968.
3) Stevenson, W.D., “Element of Power System Analysis”, McGraw Hill, 1975.
4) Ashfaq Husain, “Electrical Power Systems”, CBS Publishers & Distributors, 1992.
5) Haadi Saadat, “Power System Analysis”, Tata McGraw Hill Edition, 2002.
6) Gupta, B.R., “Power System Analysis and Design, Third Edition”, A.H. Wheeler and Co Ltd., New Delhi, 1998.
7) Singh, L.P., “Advanced Power System Analysis and Dynamics, Fourth Edition, New Age International (P)
Limited, Publishers, New Delhi, 2006.
UNIT-I
Functions of power system analysis

 To monitor the voltage at various buses, real and reactive power flow between buses.
 To design the circuit breakers.
 To plan future expansion of the existing system
 To analyze the system under different fault conditions
 To study the ability of the system for small and large disturbances (Stability studies)
COMPONENTS OF A POWER SYSTEM
1. Alternator
2. Power transformer
3. Transmission lines
4. Substation transformer
5. Distribution transformer
6. Loads
SINGLE LINE DIAGRAM
A single line diagram is diagrammatic representation of power system in which the components are represented
by their symbols and interconnection between them are shown by a straight line(even-though the system is
three phase system).The ratings and the impedance of the components are also marked on the single line
diagram.
Purpose of using single line diagram
The purpose of the single line diagram is to supply in concise form of the significant information about the
system.
Per unit value.
The per unit value of any quantity is defined as the ratio of the actual value of the any quantity to the base value
of the same quantity as a decimal.
Per unit=Actual value / Base value

The components or various sections of power system may operate at different voltage and power levels. It will
be convenient for analysis of power system if the voltage, power, current and impedance rating of components
of power system are expressed with reference to a common value called base value.
Advantages of per unit system
i. Per unit data representation yields valuable relative magnitude information.
ii. Circuit analysis of systems containing transformers of various transformation ratios is greatly
simplified.
iii. The p.u systems are ideal for the computerized analysis and simulation of complex power system
problems.
iv. Manufacturers usually specify the impedance values of equivalent in per unit of the equipments
rating. If the any data is not available, it is easier to assume its per unit value than its numerical
value.
v. The ohmic values of impedances are refereed to secondary is different from the value as referee to
primary. However, if base values are selected properly, the p.u impedance is the same on the two
sides of the transformer.
vi. The circuit laws are valid in p.u systems, and the power and voltages equations are simplified since
the factors of √3 and 3 are eliminated.
Change the base impedance from one set of base values to another set
Let
Z=Actual impedance, Ω

Zb=Base impedance, Ω

The eqn 1 show that the per unit impedance is directly proportional to base megavoltampere and inversely
proportional to the square of the base voltage.
Using Eqn 1 we can derive an expression to convert the p.u impedance expressed in one base value ( old base)
to another base (new base)
Let kVb,old andMVAb,old represents old base values and kVb,new and MVA b ,new represent new base value
Let Zp.u,old = p.u. impedance of a circuit element calculated on old base
Zp.u,new = p.u. impedance of a circuit element calculated on new base
If old base values are used to compute the p.u.impedance of a circuit element, with impedance Z then eqn 1 can
be written as
If the new base values are used to compute the p.u. impedance of a circuit element with impedance Z, then eqn
1 can be written as

The eqn 4 is used to convert the p.u.impedance expressed on one base value to another base
MODELLING OF GENERATOR AND SYNCHRONOUS MOTOR

MODELLING OF TRANSFORMER

MODELLING OF TRANSMISSION LINE

MODELLING OF INDUCTION MOTOR

Impedance diagram & approximations made in impedance diagram

The impedance diagram is the equivalent circuit of power system in which the various components of power
system are represented by their approximate or simplified equivalent circuits. The impedance diagram is used
for load flow studies. Approximation: (i) The neutral reactances are neglected. (ii) The shunt branches in
equivalent circuit of transformers are neglected.
Reactance diagram & approximations made in reactance diagram
The reactance diagram is the simplified equivalent circuit of power system in which the various components of
power system are represented by their reactances. The reactance diagram can be obtained from impedance
diagram if all the resistive components are neglected. The reactance diagram is used for fault calculations.
Approximation:
(i) The neutral reactances are neglected.
(ii) The shunt branches in equivalent circuit of transformers are neglected.
(iii) The resistances are neglected.
(iv) All static loads are neglected.
(v) The capacitance of transmission lines are neglected
PROCEDURE TO FORM REACTANCE DIAGRAM FROM SINGLE LINE DIAGRAM

1. Select a base power kVAb or MVAb

2. Select a base voltage kVb

3. The voltage conversion is achieved by means of transformer kVb on LT section

= kVb on HT section x LT voltage rating / HT voltage rating

4. When specified reactance of a component is in ohms

p.u reactance=Actual reactance/Base reactance
specified reactance of a component is in p.u
EXAMPLE
1. The single line diagram of an unloaded power system is shown in Fig 1.The generator transformer
ratings are as follows.
G1=20 MVA, 11 kV, X’’=25%
G2=30 MVA, 18 kV, X’’=25%
G3=30 MVA, 20 kV, X’’=21%
T1=25 MVA, 220/13.8 kV (∆/Y), X=15%
T2=3 single phase units each rated 10 MVA, 127/18 kV(Y/∆), X=15%
T3=15 MVA, 220/20 kV(Y/∆), X=15%
Draw the reactance diagram using a base of 50 MVA and 11 kV on the generator1.

SOLUTION
Base megavoltampere,MVAb,new=50 MVA
Base kilovolt kVb,new=11 kV ( generator side)

side)
Example
2) Draw the reactance diagram for the power system shown in fig .Use a base of50 MVA , 230 kV in 30 Ω
line. The ratings of the generator, motor and transformers are
Generator = 20 MVA, 20 kV, X=20%
Motor = 35 MVA, 13.2 kV, X=25%
T1 = 25 MVA, 18/230 kV (Y/Y), X=10%
T2 = 45 MVA, 230/13.8 kV (Y/∆), X=15%

Solution
Base megavoltampere,MVAb,new=50 MVA
Base kilovolt kVb,new=230 kV ( Transmission line side)
Symmeetrical Coomponentts
An unb balanced ssystem off N relatedd vectors can be reesolved innto N systtems of bbalanced vectors.
v The
T N–
sets of balanced vectors are
a calledd symmetrrical comp ponents. Each set consists oof N – veectors whhich are
equal in
n length annd having
g equal phhase anglees betweenn adjacentt vectors.
Sequen
nce Imped
dance and
d Sequence Netwo
ork
The seq quence immpedances are imppedances offered by b the devvices or componennts for th he like seequence
compon nent of thee current .The singgle phase equivalen nt circuit oof a poweer system consisting of impeedances
to the cu
urrent of aany one sequence oonly is callled sequeence network.
Positiv
ve Sequen
nce Comp
ponents
The possitive sequuence com mponents are equall in magniitude and displayed d from eacch other by
b 120o with
w the
same seequence aas the orig ginal phasses. The positive
p seequence ccurrents and
a voltagges follow w the samme cycle
order off the origiinal sourcce. In the case of ty
ypical cou
unter clockkwise rotaation elecctrical systtem, the positive
p
sequencce phasor are shown in Fig . The samee case app plies for thhe positiv
ve current phasors. This sequ uence is
also callled the “aabc” sequeence and uusually deenoted by the symbbol “+” or “1”
Negative Sequence Components
This sequence has components that are also equal in magnitude and displayed from each other by 120o
similar to the positive sequence components. However, it has an opposite phase sequence from the original
system. The negative sequence is identified as the “acb” sequence and usually denoted by the symbol “-” or
“2” [9].The phasors of this sequence are shown in Fig where the phasors rotate anti- clockwise. This
sequence occurs only in case of an unsymmetrical fault in addition to the positive sequence components,

Zero Sequence Components

In this sequence, its components consist of three phasors which are equal in magnitude as before but with a
zero displacement. The phasor components are in phase with each other. This is illustrated in Fig . Under an
asymmetrical fault condition, this sequence symbolizes the residual electricity in the system in terms of
voltages and currents where a ground or a fourth wire exists. It happens when ground currents return to the
power system through any grounding point in the electrical system. In this type of faults, the positive and
the negative components are also present. This sequence is known by the symbol “0” .
EXAMPLE
1. The symmetrical components of a phase –a voltage in a 3-phase unbalanced system are

Determine the phase voltages Va ,Vb and Vc

THREE-SEQUENCE IMPEDANCES AND SEQUENCE NETWORKS

Positive sequence currents give rise to only positive sequence voltages, the negative sequence currents give
rise to only negative sequence voltages and zero sequence currents give rise to only zero sequence voltages,
hence each network can be regarded as flowing within in its own network through impedances of its own
sequence only.
In any part of the circuit, the voltage drop caused by current of a certain sequence depends on the
impedance of that part of the circuit to current of that sequence.
The impedance of any section of a balanced network to current of one sequence may be different from
impedance to current of another sequence.
The impedance of a circuit when positive sequence currents are flowing is called impedance, When only
negative sequence currents are flowing the impedance is termed as negative sequence impedance. With
only zero sequence currents flowing the impedance is termed as zero sequence impedance.
The analysis of unsymmetrical faults in power systems is carried out by finding the symmetrical
components of the unbalanced currents.
Since each sequence current causes a voltage drop of that sequence only, each sequence current can be
considered to flow in an independent network composed of impedances to current of that sequence only.
The single phase equivalent circuit composed of the impedances to current of any one sequence only is
called the sequence network of that particular sequence. The sequence networks contain the generated emfs
and impedances of like sequence. Therefore for every power system we can form three- sequence network
s. These sequence networks, carrying current Ia1, Ia2 and Ia0 are then inter-connected to represent the
different fault conditions.

SEQUENCE NETWORKS OF SYNCHRONOUS MACHINES

An unloaded synchronous machine having its neutral earthed through impedance, Zn, is shown in fig.
below. A fault at its terminals causes currents Ia, Ib and Ic to flow in the lines. If fault involves earth, a
current In flows into the neutral from the earth. This current flows through the
neutral impedance Zn. Thus depending on the type of fault, one or more of the line currents may be zero.
Thus depending on the type of fault, one or more of the line currents may be zero.

POSITIVE SEQUENCE NETWORK

The generated voltages of a synchronous machine are of positive sequence only since the windings of a
synchronous machine are symmetrical.
The positive sequence network consists of an emf equal to no load terminal voltages and is in series with
the positive sequence impedance Z1 of the machine. Fig.2 (b) and fig.2(c) shows the paths for positive
sequence currents and positive sequence network respectively on a single phase basis in the synchronous
machine.
The neutral impedance Zn does not appear in the circuit because the phasor sum of Ia1, Ib1 and Ic1 is zero
and no positive sequence current can flow through Zn. Since its a balanced circuit, the positive sequence N
The reference bus for the positive sequence network is the neutral of the generator. The positive sequence
impedance Z1 consists of winding resistance and direct axis reactance. The reactance is the sub-transient
reactance X”d or transient reactance X’d or synchronous reactance Xd depending on whether sub-transient,
transient or steady state conditions are being studied. From fig.2 (b) ,
the positive sequence voltage of terminal a with respect to the reference bus is given by:
Va1= Ea - Z1Ia1
NEGATIVE SEQUENCE NETWORK
A synchronous machine does not generate any negative sequence voltage. The flow of negative sequence
currents in the stator windings creates an mmf which rotates at synchronous speed in a direction opposite to
the direction of rotor, i.e., at twice the synchronous speed with respect to rotor.
Thus the negative sequence mmf alternates past the direct and quadrature axis and sets up a varying
armature reaction effect. Thus, the negative sequence reactance is taken as the average of direct axis and
quadrature axis sub-transient reactance, i.e.,
X2 = 0.5 ( X”d + X”q ).
It not necessary to consider any time variation of X2 during transient conditions because there is no normal
constant armature reaction to be effected. For more accurate calculations, the negative sequence resistance
should be considered to account for power dissipated in the rotor poles or damper winding by double
supply frequency induced currents. The fig.below shows the negative sequence currents paths and the
negative sequence network respectively on a single phase basis of a synchronous machine. The reference
bus for the negative sequence network is the neutral of the machine.
Thus, the negative sequence voltage of terminal a with respect to the reference bus is given by:
Va2= -Z2Ia2

ZERO SEQUENCE NETWORK

No zero sequence voltage is induced in a synchronous machine. The flow of zero sequence currents in the
stator windings produces three mmf which are in time phase. If each phase winding produced a sinusoidal
space mmf, then with the rotor removed, the flux at a point on the axis of the stator due to zero sequence
current would be zero at every instant.
When the flux in the air gap or the leakage flux around slots or end connections is considered, no point in
these regions is equidistant from all the three –phase windings of the stator.
The mmf produced by a phase winding departs from a sine wave, by amounts which depend upon the
arrangement of the winding.
Zero
o Sequen
nce Equ
uivalent Circuitss of Three‐Phasse Transsformerrs 
UN
NIT‐II    Graaph TTheoryy 

 
 
 
 

 
 

 
 
 
 
 
 
 
 

 
 
 

 
 
  

 
 Flow Chart for Inspection Method

START

Read line data & bus

data

Initialize the Y-bus matrix Yij  0

for all values of i and j

k=1

i  sb ( k ) & j  eb ( k )
Yii  Yii  1 ( rij  jxij )  YCk
Y jj  Y jj  1 ( rij  jxij )  YCk
Yij  Y ji   1 ( rij  jxij )

k=k+1

Yes Is

k<=NL

No

Print y-bus matrix

STOP
 UNIT-III SPARSITY TECHNIQUES
INTRODUCTION
 Sparsity is the condition of not having enough of something.
 If a matrix contains less number of non-zero elements, then that matrix is considered as sparse matrix. In
power systems, most of the matrices like Ybus matrix and Jacobian matrix are sparse matrices.
 Sparsity technique is a programming technique is a digital programming technique by which sparse
matrices are stored in a compact form in computer memory.
 Only non-zero elements are stored and calculations are done on non-zero values, thereby not only
reducing the computer memory requirement but also reducing the computation time.
 Most the software programs use sparsity techniques effectively in solving very large problems like
power flow of Indian Power System.

SPARSITY TECHNIQUES
1. Compact Storage Scheme
2. LU Factorization
3. Optimal Ordering

COMPACT STORAGE SCHEME

While storing non-zero elements of sparse matrices in computer memory, a systematic procedure must be
adapted so that the non-zero element can be accessed, altered, included or removed. To handle sparse matrices,
two methods are popularly used.

 Entry-Row-Column Method
 Chained Data-Structure Method

Entry-Row Column Method

0 1 0 
 Consider a sparse matrix A  3 0 0
0 0 2

 The above matrix can be stored in compact form as follows:

STO RN CN
1 1 2
3 2 1
2 3 3
where
STO : Stored Non-Zero Values
RN : Row Number
CN : Column Number

 It is very clear from the above example that there are three linear vectors to store non-zero values.
 These three vectors contain all the data present in the original [A] matrix.
 This is the simplest method but it has some drawbacks.
 The main drawback is that data retrieval is not so fast.
 This method is not followed in practice.

Chained Data-Structure Method

1 0 0 1
4 3 0 0
 Consider a sparse matrix A
0 0 0 2
 
0 0 1 0

 The above matrix can be stored in compact form as follows:

 STO CN . . . . N-First
(1) --> 1 1 1 --> First row starts from array index (1)
(2) --> 1 4 3 --> Second row starts from array index (3)
(3) --> 4 1 5 --> Third row starts from array index (5)
(4) --> 3 2 6 --> Forth row starts from array index (6)
(5) --> 2 4 7
(6) --> 1 3
(7) -->
(8) -->

The value-1 in NX vector indicates that there are some more values in the respective row.
If NX=0, there are no more non-zero values in the respective row.

 This method replaces the RN vector by RFirst vector, whose size equals only the number of rows in the
given matrix, which further reduces the memory requirement.
 The numbers in the RFirst arrays indicate the index numbers of STO/CN arrays and represent where the
a row starts in STO/CN arrays.
 This method is widely used in all practical applications.

LU FACTROIZATION OR TRIANGULAR FACTORIZATION

 

 
  

 
   
 

 
PROBLEM: Perform Optimal Ordering by Tinney-Walker Method-2 for the
following Matrix, where X represents non-zero elements.
 EM: C
PROBLE Compute the n
number of fill-iins in th
he abov
ve probllem, if we do LU
fa
actoriza
ation wiithout optimal
o orderin
ng.
UNIT-III

 
 
REPRESENTATION - POWER FLOW VARIABLES
Bus Voltage….

Vi  Vi  i  Vi e j i  Vi cos  i  j sin  i   ei  j f i

Ybus element…..

Yik  Yik  ik  Yik e j ik  Gik  j Bik

Bus Current….
n
Ii   Yij V j
j 1

Bus Power….
n
Si  Pi  j Qi  Vi I i  Vi *
 Yij* V j*
j 1

Hybrid Form….

n

j ( i  j )
Si  Pi  j Qi  Vi V j e (Gij  jBij )
j 1

Separating the real and imaginary parts …..

n
Pi   Vi V j Gij cos( i   j )  Bij sin( i   j )
j 1

n
Qi   Vi V j Gij sin( i   j )  Bij cos( i   j )
j 1

Polar Form……

n
 Vi V j Yij e
j ( i  j   ij )
Si  Pi  j Qi 
j 1
Separating……

n
Pi   Vi V j Yij cos( i   j  ij )
j 1

n
Qi   Vi V j Yij sin( i   j  ij )
j 1

 
 Rectangular Form……

n
Si  Pi  j Qi   (ei  j fi )(Gij  jBij )(e j  j f j )
j 1
Separating……

n
Pi   ei (Gij e j  Bij f j )  fi (Gij f j  Bij e j )
j 1

n
Qi   fi (Gij e j  Bij f j )  ei (Gij f j  Bij e j )
j 1

POWER FLOW ANALYSIS

Power flow analysis is the determination of steady state conditions of a power system for a specified power
generation and load demand. It basically involves the solution of a set of non-linear equations for the real and
reactive powers at each bus.

It is used in the planning and design stages as well as during the operational stages of a power system. Certain
applications, especially in the fields of power system optimization and distribution automation, require repeated
fast power flow solutions. Due to a large number of interconnections and continuously increasing demand, the
size and complexity of the present day power systems, have grown tremendously and it becomes very difficult
to obtain power flow solutions, which is ideally suitable for real time applications. The three traditional
methods used for power flow are

 Gauss Seidel (GS)

 Newton Raphson (NR)
 Decoupled NR
 FDLF

GS method was one of the most common method in power flow studies. This is the GS expression that may be
solved iteratively for the solution of power flow problem. This method is simple, requires less computer
memory but this method is slow due to poor rate of convergence, number of iterations increases directly with
the system size and choice of slack bus affects the convergence of this algorithm. Because of these drawbacks,
this method is not used for present day power systems.

NR method is very powerful technique in solving power flow problem. This is a gradient technique and needs
the jacobian matrix to be formed during the iterative process. This Jacobian matrix provides the optimal
direction for finding the solution. This method has several advantages. It reliably converges. It is insensitive to
selection of slack bus. No of iterations is independent of system size. It requires less no of iterations. But it is
very inefficient in the sense that it requires large computer memory and takes large computation time. That is
why this algorithm is not suitable for real-time applications.

Simplifications in the jacobian tend to alter the direction, generally increasing the number of iterations. If the
simplifications are done properly, an improvement in overall computational performance may be achieved.
Whatever be the simplifications made, the final solution should remain unchanged.

There is weak coupling between Real power flow and Reactive power flow in power systems. Based on this
weak coupling the real and reactive set of equations are decoupled and the problem is split into two sub-
problems in FDLF. In this method, the jacobian matrices are made constant and need not be recomputed during
the iterative process. It is developed with the following assumptions.

 the voltage magnitudes, V , are close to 1 p.u

 the phase angles,  , are not large in magnitude
 r« x .
This algorithm is fast and requires very less computer memory. This algorithm is predominantly used in the
energy management systems, even for real time applications. However, it diverges, if any of the assumptions
becomes invalid.
 

Classification of Buses

Bus Specified Computed

Slack V,  P,Q
Generator P, V Q, 
( PV )
Load P, Q V, 
( PQ )
 

Example System with Known and Unknown variables 
 
 

 
1 2 3 4 5
 

  6
 
Example Power System
 
7
 

  13 12 11 10 9 8
   

Slack  Generator 
  Load Buses
bus  Buses 
V 11  V 2  V 3 V 4 V 5                
Specified 
  P2  P3  P4 P5 P6 P7 P8 P9 P10  P11  P12 P13
Unknown 
  2  3  4 5 6 7 8 9 10  11  12 13
12 
           
Specified            Q6 Q7 Q8 Q9 Q10  Q11  Q12 Q13
Unknown 
          V6 V7 V8 V9 V10  V11  V12 V13
8 
                           
 

   
 
 
 
 
 
 
 
  

 
 

Flow Chart of NR Method 

 
 
  

 
 
 
 
Flow Chart of FDLF Method 

 
 
 
  
 
UNIT‐IV  SHORT CIRCUIT STUDIES 

 
 
 
  
 

 
  

Zero Sequence Equivalent Circuits of Three‐Phase Transformers 

 
 
 
  
   
 
 
FAULT IMPEDANC
CE AND AD
DMITTANC
CE MATRIC
CES

 
 
UNIT‐V