POWER SYSTEM

The network of generators,

transformers and transmission
lines which generates, transmits
and distributes electrical Power
to the Consumer is called
Electrical Power System.
STRUCTURE OF POWER SYSTEM

 An Electrical Power System consists of

the following sections
1. Generation – the section that generates
electrical power
2. Transmission – the part of the power
system which transports electrical power
from the generating stations to load centers
3. Distribution – this is last section that
distributes the electric power to the different
consumers.
 Generating Station 6.6kV, 10.5 kV, 11 kV,
13.8 kV, 15.75 kV
Step Up Transformer
Sending End Substation 11 kV / 220kV or 132kV
or 400kV or 765 kV
Primary Transmission

Step down Transformer

Receiving Station
220 kV/33 kV or
66 kV
Secondary Transmission

Secondary Substation
Step down Transformer
33 kV/11 kV or 6.6
kV
Primary distribution
Big Consumers
Distribution Step down
Distribution Station Transformer 11kV/400V
for 3Φ and 230V for 1Φ
Secondary Distributors
Distributors

Service mains Service mains

To Consumers
 COMPONENTS OF AN ELECTRIC
POWER SYSTEM

GENERATOR
 This is the essential component of power systems that
is used to convert mechanical energy into electrical
energy. Usually it is a three phase ac generator known
as synchronous generator or alternator
TRANSFORMER
 Transfer power or energy from one circuit to other
circuit without change of frequency (to increase or
decrease voltage level)
 The generated voltage needs to be stepped up to reduce
the current level in order to have reduced line losses.
TRANSMISSION LINE
 Transfers power from one location to another.
 POWER SYSTEM OPERATION

 Power system operation is necessary to

make the system to work according to the
demand.
 Symmetrical steady state is in fact, the most
important mode of operation. Three major
problems encountered in this mode of
operation of a power system
1. Load Flow Problem
2. Optimal Load Scheduling
Problem
3. Systems Control Problem

Here we consider the Load Flow Problem

NEED FOR LOAD FLOW ANALYSIS
 Load flow analysis is performed on a
symmetrical steady state operating
condition of a power system under normal
operating condition.
 The main information obtained from this
study comprises the magnitudes and phase
angle of load bus voltages, reactive powers
at generator buses, real and reactive power
flow on transmission lines, other variables
being specified.
 This information is essential for the
continuous monitoring of the current state
of the system and for analyzing the
effectiveness of alternative plans for future
system expansion to meet increased load
demand
 STEPS FOR LOAD FLOW
STUDIES
The following work has to be performed for
a load flow study.
 Representation of the system by single line
diagram.
 Obtaining the impedance diagram using the
information in single line diagram.
 Formulation of network equations.
 Solution of network equation.
NETWORK MODEL FORMULATION

The net Complex power injected at the ith bus is

Si  Pi  jQi
The Complex power supplied by the generator is
S Gi  PGi  jQGi
The Complex power drawn by the load is
S Di  PDi  jQDi
The Real Power is Pi  PGi  PDi

The Reactive Power is Qi  QGi  QDi

Where i  1,2,...n

Sample four bus system

Let the generators be G1 , G2 , G3 & G4 respectively
S G1, S G 2, S G 3 & S G 4
Let the generator buses be respectively
The load buses be S D1, S D 2, S D 3 & S D 4

The equivalent power source at the ith bus injects current I i

The voltages at nodes 1,2,3 & 4 are V1 ,V2 ,V3 & V4

The line admittance between ith node kth node be Yik

Now it is taken as Yik  Yki

Applying Kirchhoff’s current law at the nodes we get the

following equations
I1  V1 y10  V1  V2 y12  (V1  V3 ) y13
I 2  V2 y20  V2  V1 y12  (V2  V3 ) y23  (V2  V4 ) y24
I 3  V3 y30  V3  V1 y13  (V3  V2 ) y23  (V3  V4 ) y34
I 4  V4 y40  V4  V2 y24  (V4  V3 ) y34

 I1   y10  y12  y13   y12  y13 0  V1 

I    y  y  y  y  y   y  y  V 
 2   12 20 12 23 24 23 24  2 
 I 3    y13  y23 y30  y13  y23  y34  Y34  V3 
    
I
 4  0  y 24  y34  y 40  y 24  y 
34  V4 
  I1  Y11 Y12 Y13 Y14  V1 
 I  Y Y22 Y23 Y24  V2 
 2    21
 I 3  Y31 Y32 Y33 Y34  V3 
    
 I 4  Y41 Y42 Y43 Y44  V4 
Y11  y10  y12  y13
Y22  y 20  y12  y 23  y 24
Y33  y 30  y13  y 23  y 34
Y44  y 40  y 24  y 34
Y12  Y21   y12
Y23  Y32   y 23

Yii is self admittance or driving point admittance

Yik is mutual admittance or transfer admittance

n
In compact form I 
i  Y V ; i  1,2,3,.....n
k 1
ik k

In matrix form
Impedance matrix
Vbus  Z bus I bus

Admittance matrix

I bus  YbusVbus
1
Ybus  Z bus
 PROPERTIES OF BUS
ADMITTANCE MATRIX
 Y – Bus is n x n matrix where n is the number of
buses.
 The diagonal elements of Y – bus are the driving point
admittances and the off diagonal elements of Y – bus
are the short circuit transfer admittances.
 Yij (i!=j)=0 if ith and jth buses are not connected.

 Y – bus matrix is symmetric matrix (Yij=Yji) if the

nxn  n n(n  1)
regulating
 n  transformers are not involved. So only
2 2
terms to be stored for n bus system.
 Bus admittance matrix is symmetric along the leading
diagonal, and we need to store the upper triangular
admittance matrix only.
 Each bus is connected to only a few nearby buses. So
many off-diagonal elements are zero. Such a matrix is
called sparse.
 APPLICATIONS OF BUS ADMITTANCE
MATRIX

 Y – bus is used in solving load flow problems.

 It has gained applications owing to the simplicity in
data preparation.
 It can be easily formed and modified for any changes
in the network.
 It reduces computer memory and time requirements
because of sparse matrix.
ADDITION OF A LINE
 Addition of an element of admittance y between buses
i and j.
 This affects 4 entries in Y – bus, viz., Yii, Yij, Yji, Yjj as
follows.
 Changes in adding a line admittance y between buses i
Yiinew  Yiiold  y
and j are
Yijnew  Yijold  y
Y jjnew  Y jjold  y
Y jinew  Y jiold  y
REMOVAL OF A LINE.

Changes in removing a line admittance y between

buses i and j are
Yiinew  Yiiold  y
Yijnew  Yijold  y
Y jjnew  Y jjold  y
Y jinew  Y jiold  y

ADDING A SHUNT ELEMENT

Change in adding shunt element y from i bus to ground

is Yiinew  Yiiold  y
 LOAD FLOW EQUATIONS
The net complex power injected at the ith bus is

Si  Pi  jQi  Vi I i*
Where voltage at the ith bus with respect to ground is
the source current injected into the bus
Ii

Pi  jQi  Vi * I i
Where
n
Substituting I i   YikVk
k 1

Ii
n
Pi  jQi  Vi *
Y V
k 1
ik k

n
Pi  jQi   Vi Yik Vk  k   i   ik 
k 1

 * n 
Pi  Re al Vi  Yik Vk 
 k 1 
 * n 
Qi   Im ag Vi  YikVk 
 k 1 
The real power is
 * n 
Pi  Re al Vi  Yik Vk 
 k 1 
n
Pi   Vi Yik Vk cos( k   i   ik )
k 1

The Reactive power is

 * n 
Qi   Im ag Vi  Yik Vk 
 k 1 
n
Qi   Vi Yik Vk sin( k   i   ik )
k 1

Rewrite the above equation

n
Pi  Vi Yii cos  ii   Vi Yik Vk cos( k   i   ik )
2

k 1
i

n
Qi   Vi Yii sin  ii   Vi Yik Vk sin( k   i   ik )
2

k 1
i
 Where i  1,2,...n
In polar form
j i
Vi  Vi e
ik
Yik  Yik e
CLASSIFICATION OF BUSES
Generally two variables are specified much importance
for a bus based on this buses are classified as
PQ Bus or Load Bus
PV Bus or Voltage controlled Bus or Generator
Bus
Slack Bus or Swing Bus or Reference Bus
PQ Bus or Load Bus PGi  QGi  0

Pure load bus where

Pi & Qi

Known values are PGi & QGi

Vi ,  i
Specified parameters
 i are
Unknowns
Where = load angle
PV Bus
Generator bus Pi & Vi

Known values are PGi

Specified parameters Qi ,  i
Unknowns are

Slack Bus
Reference Bus
Specified values Vi ,  i

 i Phase angle normally set to zero

Normally one bus is set as reference bus for a bus

system
 SOLVING NON – LINEAR
EQUATIONS

The power flow equation is non linear

algebraic equation and it cannot be solved by
direct methods

Therefore we need an indirect method such as

1. Gauss – Seidal
2. Newton – Raphson
3. Decoupled
4. Fast Decoupled
 GAUSS SEIDAL METHOD

Iterative algorithm to solve a set of Non – Linear

algebraic equations
Assumptions
All buses other than the slack bus are PQ buses
Let n be the number of buses, one bus is a slack bus
and for the remaining (n-1) buses voltage magnitude
and phase angles are assumed
The voltage at the ith bus is
 
1  n
Vi  I i   YikVk 
Yii  k 1


 k i 

Current is determined from the power and voltage of
the buses Ii 
Pi  jQi 
Vi *

SubstituteI i inVi

The new bus voltage

1  Pi  jQi  i 1 n
old 
new
Vi   *old
  YikVk   YikVk 
new

Yii  Vi k 1 k i 1 
ACCELERATION FACTOR
To reduce the number of iteration in Gauss –
Seidel method the voltage at each bus is corrected by a
multiplying factor so as to bring the voltage closer to the
value it approaches. This factor is known as acceleration

factor ().


Vi new  Vi old   Vi new  Vi old 
 COMPUTER TECHNIQUE TO SOLVE THE
PROBLEMS BY USING GAUSS – SEIDAL METHOD

ALGORITHM FOR ITERATION METHOD

Step 1: Form Y – bus matrix.

0
Step 2: Assume Vk  Vk(spec) 0 at all generator buses.

Step 3: Assume Vk  10 0  1  j 0 at all load buses.

Step 4: Set iteration count =1 (iter = 1).

Step 5: Let bus number i =1.

Step 6: If ‘i’ refers to generator bus go to step. No. 7

otherwise go to step 8

Step 7(a): If ‘i’ refers to the slack bus go to step. No. 9

otherwise go to step 7(b).

Step 7(b): Compute Qi using
N * 
Q i
cal
  Im Vi YijV j 
 j 1 
QGi  Qical  QLi
ifQ i (min)  QGi  Qi max , thenQi ( spec )  Qical

ifQ i (min)  Q Gi thenQi ( spec )  Qi (min)  Q Li

ifQ i (max)  QGi thenQi ( spec )  Qi (max)  Q Li

If Q limit is violated then consider this bus as

P –Q bus till convergence is obtained.

Step 8: Compute 1Viusing
Pi  jQ
the 
i equationnew  j 1 n 
new
Vi     YijV j  Y V 
ij j
old

Yii  Vi*old i 1 i  j 1 

Step 9: if i is less than number of buses, increment i by 1

and go to step 6.
Vi new V jold
Step 10: compare two successive iteration values for Vi
If - < tolerance, gooldto step 12
V new
V V 
Step 11: update the new voltage as
old new
 V
V old  V new
 iter = iter + 1; go to step 5
Step 12: Compute relevant quantities.
Slack bus power,
n
S1  Pi  jQi  V I  Vi  
Y V
j 1
ij j

Line flows
S ij  Pij  jQij

  2
 Vi Vi   V j Yij series  Vi Yij
PLoss  Pij  Pji
QLoss  Qij  Q ji

Step 13: Stop the execution

NEWTON RAPHSON LOAD
FLOW ANALYSIS
Converges fast for large bus systems too, usually 4 to 5
iterations
It is successive approximation procedure it uses Taylor
series expansion
Consider a set of non-linear algebraic equations
f (x, y)  0
g (x, y)  0
Where x, y are unknowns
Let the initial value be x0 and y0
Let the corrections to be added are ∆x and ∆y.
The actual solutions now are
x = x0 + ∆x
y = y0 + ∆y.
Now the functions are
f(x 0  x, y 0  y)
g(x 0  x, y 0  y)
Expanding the equation by Taylor series

2 2
 f  x   f  x  f  y   2 f  y 2
f ( x0 , y 0 )     2   .........     2   .........  0
 x  xyxy00 1!  x  xyxy0 2! y
  y  y0
x  x 0 1! y
  y  y0
x  x 0 2!
0

 
 f  x   f  x  f  y   2 f  y 2
2 2

f ( x0 , y 0 )      2   .........     2   .........
 x  xyxy0 1!  x  x  x0 2!  y  xyxy00 1!  y  xyxy00 2! 
 0 y  y0 

 g  x   g  x  g  y   2 g  y 2
2 2
g ( x0 , y 0 )     2   .........     2  .........  0
 x  y  y00 1!  x  xyxy0 2!
x  x
 y  xyxy00 1!  y  xyxy00 2!
0

  2
 2
 2
 2 
 g  x  g x  g  y  g y
g ( x0 , y0 )      2   .........     2   .........
 x  xyxy0 1!  x  x x0 2!  y  x x0 1!  y  x x0 2! 
 0 y  y 0 y  y 0 y  y 0 

Neglecting the second and higher order terms, the above

two equations can be expressed in matrix form as

  f   f  
   x x  y  x  x 
 f ( x0 , y 0 )   x  y  y00   y  y00  x 
 g ( x , y )     g   y 
 0 0    g   y  x  x   
  x  xyxy0   y  y00 
 0

From xy,new x0y 0 values are obtained for next

iteration
 x0new  x0old  x
y 0new  y 0old  y

Values are used for updating the results of previous

iteration xy

After this next iteration is carried out

This procedure is followed until a solution is obtained,

i.e., until two successive iterations provide the result

 Pi & Qi can be represented as
Pi = P (δ, V)
Qi = Q (δ, V)
Load flow solution is achieved
when Pi = Pispecified and Qi= Qispecified
The above two equations are linearized by Taylor’s series
  P     2 P   2  Pi   V   2 Pi   V
2

Pi  Pi ( 0 , V0 )     i
 2 i
 .........      2
  ......... 
   2  1!   3  2!  V
 2  1!   V3  2! 

  P     2 P   2  P   V   2 Pi   V
2

Pi  Pi ( 0 , V0 )    i    2i   .........   i    2
  ......... 
   2  1!   3  2!   V2  1!   V3  2! 

  V   2 Pi   V
2
 P     2 Pi   2  P
Pi   i   2 
 .........  i   2
  .........
  2  1!   3  2!   V  1!   V  2!

 Qi   V   2Qi   V
2
 Q     2Qi   2
Qi  Qi ( 0 , V0 )   i   2  .........    2  .........
  2  1! 
 3  2!  V
  1!   V  2!

  Q     2Q   2  Qi   V   2Qi   V
2


Qi  Qi ( 0 , V0 )   i
  i
2 
 .........     2
  .........
   2  1!   3  2!   V  1!   V  2! 

  Q     2Q   2  Qi  V   2Q   V 2 
Qi    i
  i
2 
 .........      i
2 
 ......... 
   2  1!   3  2!   V  1!   V  2! 

Neglecting the second and higher order terms, the above

two equations can be expressed in matrix form as

   Pi   Pi 
   
 i     k
P    Vk    k 
Q     Qi
 
   Vk 
 i  Qi 
   
   k    Vk  

 Pi    k 
 Q   J  
 i    Vk 

where
  Pi   Pi 
   
   k    Vk
 

J  
 Qi   Qi 
   
   k
    Vk
 
 

H  N 

 J  L 
Steps Involving In N - R Method

Step 1 : Assume initial values of δ and V i.e., δ0, Vo

Step 2 : Calculate Pi0 and Qi0 and estimate mismatch

powers ∆Pi0 and ∆Qi0 also calculate elements

of J
Step 3: Solve equations for correction vector
Step 4: Add the correction vector with the old values to
get new values
Step 5: This procedure is repeated till acceptable
solution is obtained.
 FAST DECOUPLED LOAD
FLOW ANALYSIS
By decoupling the weak relationship between P & V
and Q & F many of the elements of Jacobin matrix can
be made zero. This enables fast convergence
As the N & J sub matrices are zero now ,
The load flow equations reduce to

P   H  
V 
Q   L  

 V 


With the following assumptions the above said

Decoupled load flow is changed into Fast Decoupled
load flow
cos  ij  1
sin  ij  0
Gij sin  ij  Bij
2
Qi  Bii Vi
The sub matrices are reduce as follows

H ij  Lij   Vi V j Bij fori  j

2
H ii  Lii   Vi Bii fori  j
The final equations are

P   Vi V j Bij  

'

Q   Vi 
2 " V 
Bii  
 V
 


where Bij’ and Bij” are elements of [-B] matrix

As it is a modified version of NR method the steps are
more are less the same
 COMPARISON OF LOAD FLOW
METHODS

Method/ G-S N-R Fast

feature Decoupled
complexity Simple More Less
complex complex
Memory Small Large Medium
requiremen
t
Number of More Usually 3 to 4 to 7
iterations 6
Time for Less More Between the
convergenc first two
e
Acceleratio Required No No
n factor
Slack bus Trail and Tolerant Tolerant
error method