EEEN 4371

Power Flow Analysis

0
Single-Line Or One-Line
Diagram:
In Power System, the circuit models for transformers and transmission lines are developed
using a simple single-line diagram that represents that actual three-phase system.

a Z Line
_+

Z Line
_+ b

c Z Line
_+

Dr. A.M.
Gaouda
UAE University
ELEC572,
04/05
a Z Line A
_+

Dr. A.M. N
Gaouda UAE
University
ELEC572, 04/05

1
Single-Line Diagram: Part of UAE Power System Network-ADWEA

KHAZNA 400 kV
SALAMAT
RAMAH WATHBA 220 kV
AL AIN PS 132 kV
MAHAWI
33 kV
C.CENTRE
WAGEN
MZYAD SWEIHAN
HAYER HABSHAN

ARAD MIRFA

SENAIYA
AL AIN SW
MD ZAYD
ADCO

GAYATHI
LIWA

ZAKHER
RUWAIS
MOSAFAH SHUWAIHAT

DAHMA

BEACH(E48)
ADNOC
OHA (TAKREER)

ADCO
E-19
UNPS

SHAHAMA SILA
GIC

ADPS-400
TAWEELAH UNPS

2
 EXAMPLE:

For the three-phase power system shown below:

1. Define each element in the system (Generators, Transformers, Loads,

Circuit Breakers, Transmission Lines, Grounding points).

2. Draw the Single-Line or One-Line Diagram.

ABRK

0.0001 A
B BRKA
0.1

B
C A 0.0001
C B
B
C
BRK A BRK A C
ABRK 0.02 [MVA] A BRK 0.02 [MVA] A BRK
A 0.0012 1.0 A
0.0001 A A
A
B B
1.0

#1 B #1 B
B B #2 0.0012 1.0 B #2
C B C B
B 0.2 0.2 0.2 0.2
C C
C C 0.0012 1.0 C
C C
C
BRK

A 1.0 0.0012

BRK
B 1.0 0.0012
0.0012 1.0 A
340.0

340.0

340.0

C 1.0 0.0012
340.0

340.0

340.0

0.0012 1.0 B

0.0012 1.0 C

3
 ...

Generator 1
ABRK Generator 3
0.0001 A
B BRKA
0.1

B
C A 0.0001

C Transformer 1 Transmission Line Transformer 2 B

B

C
BRK A BRK A C
ABRK 0.02 [MVA] A BRK 0.02 [MVA] A BRK
A 0.0012 1.0 A
0.0001 A A A
B B B B
1.0

B #1 #2 #1 #2
B B 0.0012 1.0 B B
B C 0.2 0.2 C C 0.2 0.2 C
C C C 0.0012 1.0 C C

Generator 2 C
BRK

A 1.0 0.0012

BRK
B 1.0 0.0012
0.0012 1.0 A

340.0

340.0

340.0
C 1.0 0.0012

340.0

340.0

340.0
0.0012 1.0 B

Load 2
BusBar
0.0012 1.0 C

Load 1
BusBar
The Single-Line or One-Line Diagram

BusBar BusBar
Gen. 1
0.0001
Gen. 3
BRK
0.1

0.0001

Transformer 1 Transmission Line Transformer 2 BRK

0.0001
#1 #2 #1 #2
BRK
1.0

BRK BRK 0.0012 1.0 BRK BRK

Gen. 2
CB CB CB CB
BRK 1.0 0.0012
340.0
340.0

0.0012 1.0 BRK

BRK
#1 #2
BRK BRK
#1 #2
BRK
Load2
Load1
4
4
Formulation of the Bus Admittance Matrix
The first step in developing the mathematical model describing the power flow in
the network is the formulation of the bus admittance matrix. The bus admittance
matrix is an n×n matrix (where n is the number of buses in the system)
constructed from the admittances of the equivalent circuit elements of the
segments making up the power system. Most system segments are represented by
a combination of shunt elements (connected between a bus and the
reference node) and series elements (connected between two system buses).
Example: Formulate the bus admittance matrix for the network
shown in the Figure. The Impedance diagram of the
G1 G2
system is as indicated. Shunt elements are ignored.
j1.0
Bus 1 Bus 2
j0.8 Solution:
The node voltage method is commonly used for the power
j0.4
system analysis. Where,
j0.2
j0.2
 I 1  = Y11 Y12 Y13 Y14  V1 
Y  V 
 I2 
Y Y
Bus 3 21 24   2 
j0.08 Y 22 23

Y33 Y34  V3 

 I 3 
Y31 Y32
 
Bus 4  I4  Y41 Y42
Y44  V4 
Impedance diagram
Or I bu  Y[Ybu ] Vbu 5
43
The system can be represented in terms of its y 01 y 02 I2
I1
admittance elements as shown, where:
1
yij 
Bus 1 Bus 2
y12

Zij y y
1
 y12 
13 23
y 01
j1.0   j1.0  Bus 3
y 34
1 y13  y 
y 02   Bus 4
j0.8 j1.25
Admittance diagram

Applying KCL at each node (bus), then

I1  ( y01  y12  y13 )V1  y12V2  y13V3

I 2  ( y02  y12  y23 )V2  y12V1  y23V3

0  ( y31  y32  y34 )V3  y31V1  y32V2  y34V4

0  y34V4  y34V3 6
Define:  y13 and Y12  Y
Y11  y01  y 12
 y12  y23 Y13  Y31   y13
Y22  y 02
 y34 Y23  Y32   y 23
Y33  y 31  y 32
Y44  y 34 Y34  Y43  
Y42  Y24  0 y 34
Then, the Node Voltage Equation is: Y14  Y41  0y y
I1 01 02 I2

 II 1  = Y11 Y12 Y14  V1 


Y  V2 
Bus 1 Bus 2
 y12

  Y21 Y13   y13 y 23

2
 I 3  Y31 Y34  V3 
 
Bus 3
y 34
Y Y22
Y Y
 I4 Y 44  V4  Bus 4

Y23
 Or V24  [Y 1
] I [Z ] I
Ibus  [Ybus ] Vbus bus bus bus
bus
bus
32 Y
Substituting the values, then the bus admittance matrix of the network
is: Y33
j2.5 0 
Ybu   j8.5
 
 j2.5 41 42 43 0 
s j5.0
 j12.5 
 0  j8.75 j5.0
 7
 j12.5
NOTE:
I1 y01 y02 I2
The formulation of the bus admittance matrix follows
two simple rules: Bus 1
y12
Bus 2

1. The admittance of elements connected to node k y13 y 23

is added to the (k, k) entry of the admittance
matrix. Bus 3
y34

2. The negative of the admittance of elements Bus 4

connected between nodes j and k is added to the 1

2
(j, k) and (k, j) entries of the admittance matrix. 3

Linear Ii
i
Network
V
Vii
n

Reference

( (
Y11 Y12 Y13 Y14  ( y01  y12  y13 ) (0) 
 (  y21 ( y02 y12y)21  y( 13)y
Y Y Y
 23 ) 23 ( 0 ) 
21 Y 22 23 24   ()  y31 ) y (y ) ) (y y y )
Y31 Y34   32 31 32 43
(  y34 )
(0) (0) (  y43 )
Y41 Y32 3344 
YY (y ) 43 
 8
 Introduction to Power Flow
Analysis
The power flow is the backbone of the power system operation, analysis
and
design. It is extremely necessary for planning and operation.
Generator
S

Unit Transformer Generator

S
132 kV BB
Inter bus Unit Transformer
Transformer
400 kV BB

400 kV OH Line,
4 conductor ACSR Dove

400 kV BB

400/132kV SS 400/220kV SS
3x500 MVA 3x500 MVA
220/33 kV SS
33 kV BB

132 kV BB 4x40 MVA

132 kV UG Cable
HT Consumers (Bulk)

11 kV BB 11 kV BB
132 kV BB Distribution

Transformer 415 V BB

LT Consumers 9
For the network shown, there are some buses connected to generators and other
buses are connected to the loads.
The Real and Reactive power is known at each Load bus. The Generator Voltages
are also specified at the generator buses.
Generator
S
The Transmission Lines interconnecting the
buses have resistance and inductance.
Unit Transformer Generator
132 kV BB
S Therefore, the Electric Current flowing through
Inter bus
Transformer
Unit Transformer the lines results in Electrical Losses.
400 kV BB

400 kV OH Line,
The Generators in the System Must supply
4 conductor ACSR Dove the Total Electrical Loads plus the Electrical
Losses.
400 kV BB

400/132kV SS 400/220kV SS
3x500 MVA 3x500 MVA
220/33 kV SS
33 kV BB

132 kV BB 4x40 MVA

132 kV UG Cable
HT Consumers (Bulk)

11 kV BB 11 kV BB
132 kV BB Distribution

Transformer 415 V BB

LT Consumers 10
There are some constrains should be considered while running the system

1. The Generators Must Operate

Generator
S
within
Unit Transformer Generator
their Generation Capabilities.
S
132 kV BB
Inter bus Unit Transformer
2. The Generators Must Deliver
Transformer
400 kV BB the required power at the
400 kV OH Line, Voltage at the Loads.
Desired
4 conductor ACSR Dove

400 kV BB

400/132kV SS 400/220kV SS
3x500 MVA 3x500 MVA
220/33 kV SS
33 kV BB

132 kV BB 4x40 MVA

132 kV UG Cable
HT Consumers (Bulk)

11 kV BB 11 kV BB
132 kV BB Distribution

Transformer 415 V BB

LT Consumers 11
 3. There should be no bus voltage
Generator
S
either above or below the
Unit Transformer Generator specified Voltage operating limits.
S
132 kV BB
Inter bus Unit Transformer 4. There Should be no Over-Loading of
Transformer
400 kV BB
equipment, including Transmission
400 kV OH Line,
4 conductor ACSR Dove
Lines and Transformers.

400 kV BB

400/132kV SS 400/220kV SS
3x500 MVA 3x500 MVA
220/33 kV SS
33 kV BB

132 kV BB 4x40 MVA

132 kV UG Cable
HT Consumers (Bulk)

11 kV BB 11 kV BB
132 kV BB Distribution

Transformer 415 V BB

LT Consumers 12
In Case of an Equipment Over-Loaded Or Voltage-Limit Violation.
The Generation Schedule have to be adjusted and Power Flow in the
transmission lines have to be Re-routed or Capacitor Banks have to be switched
in order to bring the system into its Normal Operating Conditions.
Generator
S
To Satisfy all the previous requirement for a
Unit Transformer Generator
S
Reliable Power System Operation, Power
132 kV BB
Inter bus Unit Transformer
Flow Study is a MUST.
Transformer
400 kV BB

400 kV OH Line,
The Power flow study is an essential part in
4 conductor ACSR Dove power system Operation, Planning and
Design.
400 kV BB

400/132kV SS 400/220kV SS
3x500 MVA 3x500 MVA
220/33 kV SS
33 kV BB

132 kV BB 4x40 MVA

132 kV UG Cable
HT Consumers (Bulk)

11 kV BB 11 kV BB
132 kV BB Distribution

Transformer 415 V BB

LT Consumers 13
Power Flow
Analysis
It is the solution for the static operating condition G
G
(Normal Operation Condition) of a power
system.
G
The node voltage method is commonly used for
the power system analysis. The formulation of
G
the network equations results in complex linear
equations in terms of node currents.
In power systems, powers are known rather than currents. Thus, resulting
equations in terms of power become non-linear and must be solved by iterative
techniques.

These non-linear equations are known as power flow equations or load flow
equations.

The power flow software computes the voltage magnitude and phase angle at
each bus bar in the system under steady-state operation condition.

These programs use the bus-voltage data to compute the power flow in the
network and the power losses for all equipment and transmission lines.
14
Load Flow Equations:
For an ‘n’ bus
system

I bu
s
n
Ii   Yip Vp 
p 1
1
Y
V p  |V p |  Linear
b
Network
p
u
 | Yip |  s
Yip
ip n

V | 
I i  | Yip |  ip ||V p  I1  Y11 Y12 Y13 ....Y1n  V1 
b  ... 
=
 .... .... .... .... ....  ....
p np 1  Ii  Y i1 Yi 2 Yi ....Yin  V 
u i
 I    Y3 Yn2 Yn3 ....Ynn  
I i  p |Yip ||Vp | (s p  ip  n  n1 V
 n
1
The )current enters the 15
In power systems, power is known rather than currents. The complex
power
injected into the ith bus is:
SV   jQ 
1

I * Pi i i i
i
2
3

n Linear Ii

Vi  | Vi | I i   | Yip || Vp |   ( p  ip )
* Network i

i and
p V

Vii
1 n
n Reference

Si  Pi  jQi  | Vi | ip | Yip || V p |   ( p  

n 1
ip )

Si  |Vi |  | Yip || Vp | (i   p  

ip )
n
Si  Pi  pjQ
1 i
 | Vi | p | Yip || V p | (i   p  
n
1
)
 | Yip || V p | cos (i     ip )  Pi,cal
ip
Pi  | Vi | p1
n
p

Qi  | Vi |  | Yip || V p | sin (i     ip )  Qi,cal ELEC572,

014/605
The two equations are known as the polar form of the load flow equations. They
provide the calculated value of net real power and net reactive power entering bus
‘i’.
Let be the scheduled value of the real power generated at bus
‘i’.
Let P be the scheduled value of the real power demand at bus
‘i’.
Let P be the net scheduled power injected at bus ‘i’.
Pi , and Qi ,
g g
G1
Pi ,sch  Pi ,g 
 Pi is the
Pi ,d
mismatch Pi Bus i

 and Qi ,d
Pi ,d
Pi ,sch  Pi ,cal
 Q
PQ
i i ,sch
 Qi is the
 Qi  Qi  i (
mismatch P Q ,sch ,cal

 Qi  Qi ,d )  Qi ELEC572,

014/705
If the calculated “net” values match the scheduled “net”values, then the
mismatch is zero and the power balance condition is:
Mismatch = Scheduled values - Calculated values = 0
Pi  (P P

Qi  (Q Q
What is the Power Flow Problem?

For the network shown in the figure, there are two

power flow equations can be written for each bus. The
power flow solution is to solve these equations for
values of unknown voltages that satisfy the power
balance condition at each bus.

NOTE 1: If there is no scheduled value Pi for bus “i” then the mismatch can
not be defined and there is no requirement to satisfy the balance condition at this
bus while solving the power flow problem.
NOTE 2: Similarly, If there is no scheduled value Q for bus “i” then the
i ,sch
mismatch can not be defined and there is no requirement to satisfy the balance
condition at this bus while solving the power flow problem.
ELEC572,
0
8
1
4/
04/05
In summary, the Power Flow solution is to solve for each bus, the following
two
equations: n

Pi |Vi |  |Yip ||V

|
n

Qi |Vi |  |Yip ||V p | sin (  i   p   ip

)
p1

Note:
n
From the power flow equations, for each Pi |Vi |  | Yip ||V p | cos (  i   p   ip )  Pi
bus there are two equations and four n
,cal p1

variables,
Qi |Vi |  | Yip ||V p | sin (  i   p   ip )  Qi

Pi ,cal
p1
If there are ‘n’ buses in the system, then the number of equations are (n*2) and
the number of variables are (n*4). The number of equations is less than the
number of unknowns., Q , |Vi to be able to solve these equations, at each bus,
Ini order
two of the four variables are specified and the remaining two are calculated. The
specified quantities are
| chosen according to bus ELEC572,

014/905
 G1 G2
Example: For the network shown, the number of buses is: n=3
Bus 1 Bus 2

The number of equations is: (n*2) = 6 equations

The number of unknowns is: (n*4) = 12 variables
Write the power flow equations at each bus.
Solution: Pi , Qi , |Vi | and  for i  1 to 3 Bus 3

n
At bus 1 i
i=1
p=1:n Pi

|Vi | p1|Yip ||V p | cos (  i   p   ip
n=3 P1  |V1) | | Y11 ||V1 | cos ( 1  1  11 ) This is
the calculated
 |Y12 ||V2 | cos( 1  2  12 value of net real
) | Y ||V | cos (   ) power at bus ‘1’.
13 3 1 3 13
n  
Qi  |Vi | | Yip ||V p | sin( i   p  
Q)  |V |  |Y ||V | sin (     
ip This is
1 1 11 1 1 1
p 1
11 the calculated value
 | Y12 )||V2 | sin (  1   2   of net reactive power

 )| Y13 ||V3 | sin (  1  3   13 )

at bus ‘1’.
12

 
In a similar way, we can write the equations for P2 , Q2 , P3 and Q3 ELEC572,

024/005
Classification of buses:
PQ bus
Different buses at the network can be classified Slack

as:
1. The Load Buses (PQ bus)
PQ bus Bus PV bus

2. The Generator Bus (PV PV bus

bus) PV bus PV bus

3. The Slack or Swing Bus PV bus PQ bus

1. The Load Buses (PQ bus)

A non-generator bus. The active and reactive powers are specified at
this bus. The voltage magnitude and phase angle are unknown.
Pi,g and Qi,g
Pi and Qi areknown & |Vi |and  i are
Bus i unknown
Generators Power: Pi ,g  0 and Qi ,g 
0
Pi,d and Qi,d Delivered Power: and Qi ,d are known
Pi ,
d
In practice, the load real power is known from measurement, load
forecasting or historical record and the reactive power is assumed based on
0.85 pf.
 Pi ,d and Q
P  P 
i ,sch i ,g
Qi ,sch  Qi ,g  Qi ,d i ,d

ELEC572,

P 024/105
2. The Generator Bus (PV bus)
The bus is also known as “Voltage controlled bus” because G1
the voltage magnitude can be kept constant. At this bus the Pi,sch
net active power and the voltage magnitude are specified. Bus i
The reactive power and the voltage phase angle |V |
unknown.
are i
Pi and |Vi |are known & Qi and  i
are unknown
NOTE: There are certain buses without generators |V |
may have voltage controlled capability. At these buses
the real power generation is zero. Q
C
3. The Slack or Swing
Bus Because the system losses are not known precisely before completing the
power flow solution, it is not possible to specify the real power injected at
every bus. Hence, the real power of one of the generator buses is allowed to
swing. The swing bus supplies the slack between the scheduled real power
generation and the sum of all loads and system losses. The voltage angle of the
slack bus serves as a reference,  i  0
|Vi | and  i are known & Pi and Qi are
unknown ELEC572,

024/205
Classification of buses:

P Q V δ
P-Q bus known known unknown unknown
P-V bus known unknown known unknown
Slack bus unknown unknown known known

• Since load flow problems generate non-linear equations

that computers cannot solve quickly, numerical methods
are required.

• The following methods are commonly used in solving

these non-linear equations:
• Gauss-Seidel Method.
• Newton-Raphson Method.

23
NOTES: After solving the load flow problem

Real Power losses  Total

N N N
gen
PL   Pi,g     Pi(Net)
i1 i1 i1
Pi,d
In the load flow problem, we select the PQ bus

slack bus at which the power Pg is not Slack

Bus
PQ bus PV bus
scheduled.
After solving the load flow problem, the PV bus
difference (Slack) between the PV bus PV bus

specified power going into the system at

total PV bus PQ bus
all other buses and the total output plus
n
the losses is assigned to the slack bus. Pi |Vi
For this reason a generator bus must be | | Yip ||V |

selected as a slack bus. Qi |Vi n

| | Y ip ||V p | sin( δi  δ p  γip )
p1

Voltage of the swing bus is selected as a reference. Q1g

|V1 |
Generally, the bus of the largest generator is G1
selected as swing bus and numbered as bus 1.
P1g δ1
ELEC572,

024/405
Solution of Non-Linear Equations
The two load flow equations are:
n

Pi |Vi |  |Yip ||V

| n

Qi |Vi |  |Yip ||V p | sin(  i   p   ip )

p1
reactive
These
power equations provide
entering bus the calculated
‘i’. The equationsvalue of net real and only a numerical
are non-linear
power
solutionand net
is possible. There are different methods could be implemented to solve
these equations. Among those is the Gauss-Seidel and Newton-Raphson methods.

Gauss-Seidel Iteration
Consider a system of non-linear equations having “n” unknowns x1 , x 2 , .........., xn

f 1 ( x1 , x 2 , .........., x n )
f 2 ( x1 , x 2 , .........., x n )
.. .. .. .. .. ..
f n ( x1 , x 2 , .........., xn ) 25
Rearranging, then
Eq. 1

1i
Assuming initial values,
x xo ,o x1 o , x2o ..........,
3 n

Start the first iteration

Substituting the initial values in Eq. 1, then All values are initial
valuesx o ,o x o , ..........,
first iteration x 1 2 n

first variable
x11  f11 ( x1o , x2o , .........., xn o
) 1 x1  x11 from previous step
x  f 1 ( x 1 , x o , .........., and all other values
are initial values
xo ) x o , ..........,
xo
2 2 1 2 n
x  f ( x , x , x ..........,
1 1 1 1 o 2
n
x1  x 1 & x  x21
xo ) 1
and 2
3 3 1 2 3 n x o , ..........,
Or in general xo
3
x  f ( x , x , .....,x ,.....,x )
1 1 1 1 o o
n
i i 1 2 i n 26
Where xi1 is the first approximation xi using the initial assumed values.
of
After K iteration, The k approximation of
th th x i is:

K th iteration
xi  fi ( x1 , x2 ,
k 1 k k , x k  1 , x k  1 .....,x k  1 )
i i i 1 n
i th
var
.....x
1 k

The changes in the magnitude of each variable from its value at

the previous iteration is:
x ki x ik 1

 x  xk
i i
x ik 1
If  x i  then the solution has converged.


Where,  is a small value ( for exmple :   0.001

)

27
EXAMPLE:
For the following equation, find an accurate value for x up to 5 decimal
places.

2x  log( x )  7

SOLUTION:
Using Gauss-Seidel x  0.5(7 

xo  x 1  0.5(7  log 1 )  3.5 1st iteration

x1  x 2  0.5(7  log 3.5 )  3.772034 2nd ite

x  3.772034
2
x  0.5(7  log 3.772034 ) 
3 3 rd i

3.788287
x 3  3.788287 x 4  0.5(7  log 3.788287 )  3.789221

x 5  3.789274
x 6  3.789278   0.000004 28
EXAMPLE:
For the following equations, find an x and y after 4
iterations.
x  0.7 sin x  0.2 cos y & y  0.7  0.2 sin
cos x y
Using Gauss-Seidel, assuming initial values
SOLUTION:
All values are initial
x o  y o  0.
valuesx o ,o x o , ..........,
x 1 2 n

x 1  0.7 sin x o 
x 1  0.7 sin 0.5  0.2 c
x 1  0.5 x1  x11 from previous step
and all other values
are initial values
y 1  0.7 cos 0.51111  0.2 sin x o , ..........,
xo
0.5
y 1  0.51465
2
n

x 2  0.5 y 2  0.510241
x 3  0.520211 y 3  0.509722
x 4  0.522520 y 4  0.509007 29
 Gauss-Seidel Advantages
• Each iteration is relatively fast (computational
order is proportional to number of branches +
number of buses in the system
• Relatively easy to program

30
 Gauss-Seidel Disadvantages
• Tends to converge relatively slowly, although
this can be improved with acceleration
• Sometimes it misses solutions, particularly on
large systems

31
Gauss-Seidel Summary

32
Gauss-Seidel Summary

33
Gauss Seidel Method when PV buses are absent
Assuming a power system in which the voltage controlled buses are absent. If the
system has n buses, then; one bus will be considered as a slack bus and the other n-1
buses are load buses (PQ-buses).

For the Slack or Swing Bus:

|Vi | and   0 are known & Pi and Qi are unknown

The swing
i
bus voltage is taken as a reference. Its voltage magnitude is known and
its phase shift angle is set to zero.

For (n-1) Load Buses (PQ bus):

Pi and Qi are known & |Vi |and  i are unknown

Using Gauss-Seidel method, we assume the initial values for the magnitude
and phase shift angle of (n-1) buses. These values are updated at each iteration.

34
For an ‘n’ bus
system
I bus  Ybus Vbus ….. Eq. 1

For the ith bus of an ‘n’ bus system, the

current
entering this bus is:

I i Yi1 V1  Yi 2 V2  ...... Yii  ...Yin Vn ….. Eq. 2

Vi n
I i  Yii Vi  Y
p1
ip V p ….. Eq. 3
pi

 
n

Vi  1  I i  p  ….. Eq. 4

ip
Yii  Y 
 
V
In power systems, power is known rather p1
than
p currents. The complex
power injected into the ith bus is: i
35
 ….. Eq. 5

S i*  Vi * I i ….. Eq. 6

OR
P
 ij ….. Eq. 7
Ii 

Substituting in Eq. 4

Rearranging in GS
  method
n

Vi  1 
 P i  jQ i
  Y ip V p  ….. Eq. 8 “Vi” is moved to
Yii  V i* p1 
the left hand side
 pi  of the equation.

Since bus 1 is the slack bus “reference”, then Vi represents n-1 set of
equations for i= 2, 3, …., n. These equations will be solved using G-S
method for the unknowns V2, V3, …..Vn.
36
NOTES:
1. Eq. 8 can be written
as: 1 P  jQ n
NOTE Vi  Yip V ….. Eq. 9
The values for P and
Q are the scheduled Vi*
i
Yii
i
  Y
p1 ii
p

pi
values for PQ Bus.
n
Ki
Vi   L ip p ….. Eq. 10
V i*
V
K i  Pi  jQi p1
pi
Lip  Y
Yii
The values for Ki and Lip are computed onceandin the beginning and used in every
iteration.
2.The voltages at all the buses in a power system are close to 1.0 pu. Therefore,
we can start the G-S iteration process assuming initial values for the voltages
equal to 1.0 and making zero angle.
V ( 0)  V ( 0)  ...... V ( 0)  10
2 3 n

3.At each step in the iteration process use the most updated values for the
voltages to compute the new values for the bus voltages.
37
 n
Ki
Vi  * 
Vi
L ip ….. Eq. 11
Vp p1
pi
i n
K 1 ….. Eq. 12
Vi  i L
V* The most updated voltage
Therefore, for the (kth+1) iteration, V values are from the
previous iteration
n
Ki i

V i (k 1)   L
1
ip V p(k 1)   Lip Vp ….. Eq. 13
(Vi (k ) )* p
1
p i 1 (k )

The most updated

voltage values are from
for i  the same iteration
1,2 ,........ .n
The iteration process is continuous till the convergence occurs,
i.e.;
| Vi (k1) |  | (k
|  | Vi (k) |  ….. Eq. 14
i 1)
V 
for i 
1,2,........ .n
38
4. The current and complex power at ith bus are:
1
2
3
I i Yi1 V1  Yi2 V2  ...... Yii Vi  ...Yin Linear Ii
i
Network
Vn
And * V
Vii
Si  Pi  jQi  Vi I i n
Reference

S*  P  Q  V * I
i i i i
Or i

P  Re{V * (Y V  V 2  ...... ii V i  ...Yin Vn )}

i2
Y
i i i1 1 Y
Q   Im{ V * (Y V  Y V  ...... Y V  ...Y V )}
i i i1 1 i2 2 ii i in n

The two equations are known as the rectangular form of the load flow equations.
They provide the calculated value of net real power and net reactive power
injected to bus ‘i’.
39
EXAMPLE 1:
For the system shown in the figure, the line impedances are as indicated in
per unit on 100MVA base.
A. Using Gauss-Seidel method find the bus voltages after 7 iterations.
B. Using the bus voltages find the Slack bus real and reactive power.

1 2
256.6
0.02  j0.04
MW
pu
G
0.01  j0.03 0.0125  j0.025 pu
pu 110.2
V1  1.050 pu
o
M

3
138.6 45.2
MW M

40
 A. Using Gauss-Seidel method find the bus voltages after 7
iterations.
Formulation of the Bus Admittance Matrix 1 2
256.6
1 0.02  j0.04
y12   10  j20  y 21 pu
MW
0.02 
G Dr. A.M. Gaouda
j0.04 UAE University
 j0.03
0.01 ELEC572, 0.0125  j0.025 pu
1 V1  1.050 o pu
pu 04/05 110.2

y13   10  j30  y 31 MVA

0.01  j0.03 R

1
y 23   16  j32  y 32
0.0125  j0.025 3
138.6 45.2
D
MW MVA
r
R .

Y11 Y13
 y12  y13 
A
Y12  y12  y13 .

Ybus  
M

   y 21 y 21  y 23  y 23 
.

Y21 Y23  G
Y22   y 31  y 32 y 31  y 32 
a

Y Y33 
o
u
d
31 a

Y32 U
A
 10   10  E
 20  j50
j20
Ybus 
U
 10   16  j32
j30 n
i

  10  26  j52 26  j62
v

j20 e
r
j30  16   s 41
i
Classification of buses: 1 2
256.6
0.02  j0.04
Bus 1: Slack Bus pu
MW

G
0.01  j0.03 0.0125  j0.025 pu
V1  1.050 pu o
V  1.050o pu
pu
Dr. A.M. Gaouda
110.2
1 MVAR
UAE University

Buses 2 and 3: Load Buses (PQ bus)

ELEC572,
04/05

3
P2 , P3 ,Q2 and Q3
138.6 45.2

V are
,V known
2 , and are unknown
3 2 3
MW MVAR


 256.6  110.2MVAR
P2 ,d Q 2 ,d
MW
 45.2MVAR
P3 ,d  138.6 MW Q3 ,d Pig and Qig

Pi ,sch  Pgi  Pdi G

&
Qi ,sch  Q gi  Qdi Bus i

 P 
S i ,sch
Pid and Qid
 ( P2 , g  P2 ,d )   Q2 ,d )
S 2 ,sch 42
 Reminder
 P2 ,d )  j( Q2 ,g  Q2 The bus admittance matrix
 ( P2 ,g pu is
S 2 ,sch ,d ) Base MVA
 20  j50  10  j20  10  j30 
( 0  256.6 )  j( 0  
10  j20 26  j52   
S2  pu   10   16  j32
16 j32
26 - j62
,sch
110.2 ) 100 j30 

MVA
Pi ,sch  Pgi  Pdi
  2.566 
S 2 ,sch
Qi ,sch  Q gi  Qdi
  1.386  j0.452
S 3 ,sch
pu S i ,sch  Pi ,sch  jQ i ,sch
Note: Scheduled Power is the same as Injected Power
Using GS method, select the initial values for the unknowns
as: V (0)  V (0)
2 3  10
Start the first iteration pu

1 Y 
P2,sch  jQ2,sch  Y21  23
V (0)
3 
V 2(1)  Y22  V Y22
(V 2(0) )* Y 22 
1

1 2.566
26 j52
j1.102  10  j20 16  j32 
V 2(1)  (1.0)*  26  j52 1.05  26  1.0
j52 
 43
OR, to simplify the calculations, we
have: n Y
Reminder
The bus admittance matrix
Vi  1 Pi  jQi is

Vi* Yii
  Y
ip
p1 ii
V p
 20  j50  10  j20  10 
pi 
10  j20 26  j52  16  j32
n   10 
j30  16  j32 26 - j62
V2  K 2  j30 


2p p
*
V2 p1 L
p
K2
2 V
(0)

V 2(1)    L21 V1  L23 3
(V 2(0) )*
V
The values for Ki and Lip are computed once in the beginning and used in every
iteration.

K 2  P2  j L21  Y and L23  Y23

Y22

K2 = -0.0367 - j0.031 L21 = -0.3846 L23 = -0.6154

V1  1.050 pu o V3( 0)  10

Dr. A.M. Gaouda pu
and
V 2(1)  0.9825  j0.031
UAE University
ELEC572,
04/05 44
 K3
V 3(1)  (0) *
 L
(V

K 3  P3  j and L31  Y and L32  Y32

Y33
K3 = -0.0142 - j0.0164 L31 = - 0.4690 + j0.0354 L32 = -0.5310 - j0.0354

V 1  1.050 o pu V 2(1)  0.9825  j0.031

and V3(1)  1.0011 j0.0353

Start the second iteration K 2 , K 3 , L21 , L23 , L31 , L32 constants willbethe
same.

(2) K2 V2( 2 )  0.9816  j0.0520

V2  (1) *
 L
(V

K3

(2)
 L V3( 2 )  1.0008 
V 3(2)  31 1 V L32 2
(V3(1) )* j0.0459
V
45
Start the third iteration K 2 , K 3 , L21 , L23 , L31 , L32 constants willbethe
same.
K2 = 0.9808 - j0.0578
V 2(3)  (2) *
 L
(V

K3 = 1.0004 - j0.0488
V 3(3)   L
(V (2) *

Start the fourth K 2 , K 3 , L21 , L23 , L31 , L32 constants willbethe

iteration same.
K2 = 0.9803 - j0.0594
 L
(3)

V 2(4)  21 1 V L
23 3
(V2(3) )*
V
K3
V 3(4)   L = 1.0002 - j0.0497
(3) *
(V

After 7 iterations,
V2( 7 )  0.9800  j0.0600  0.98183  3.5035o
pu
V3 ( 7 )  1.0000  j0.0500  1.00125  2.8624o 46
B. Using the bus voltages find the Slack bus real and reactive power.

1 2
256.6
0.02  j0.04
MW
pu
G
0.01  j0.03 0.0125  j0.025 pu
pu 110.2
V1  1.050 o pu
M

3
138.6 45.2
MW M

V1  1.05  j0.0o
V2  0.9800pu j0.0600  0.98183  3.5035 pu
o

V3  1.0000  j0.0500  1.00125 

2.8624 o 47
Using the rectangular form of the load flow
Reminder
equations, then the net active and reactive powers at The bus admittance matrix
1st bus are: is

 20  j50  10  j20  10 
  16  j32
10  j20 26  j52
  10 
j30  16  j32 26 - j62

Q   Im{ V * (Y V  Y12 V2  Y13 V3 )} j30 

i 1 11
1

S i ,sch  S i ,cal

Pi ,sch  Pgi  Pdi

P1  jQ1  4.0938 - j1.8894
Q i ,sch  Q gi  Qdi
The net schedule value:
1 2
P1 ,sch  4.0938 pu Q1 ,sch  1.8894 pu 0.02  j0.04
256.6
MW
pu
Dr. A.M. Gaouda
G UAE University
0.01  j0.03 0.0125  j0.025 pu
Pi ,sch  Pgi  Q i ,sch  Q gi  Qdi ELEC572,
pu
04/05 110.2
V1  1.050o pu MVAR
Pdi

Pg1 MVA=100
Base 409.38MW Qg1  188.94MVAR 3
138.6 45.2
D
MW MVAR r
.
Gauss Seidel Method when PV buses are present
Assuming a power system has n buses, then; one bus will be considered as a
slack bus and the other buses are load buses (PQ-buses) and voltage controlled
buses (PV-buses). Let the system buses be numbered as:

i1 Slack bus

i  2 , 3, .....,m PV  buses
i  m  1,m  2 , ......,n PQ  buses

For the voltage controlled buses,

Pi and |Vi |are known & Qi and  i are unknown

|Vi |  |Vi ….. Eq. 23

| Specified
 Qi ….. Eq. 24
Qi ,min
 Qi ,max
The second requirement for the voltage controlled bus may be violated if the
bus voltage becomes too high or too small. It is to be noted that we can
the bus voltage by controlling the bus reactive power.
control 49
Therefore, during any iteration, if the PV-bus reactive power
Qi ,min  Qi  Qi ,max
violates its limits then set it according to the following rule.

set
IF Qi  Qi ,max  Qi
Qi ,max NOTE
For PQ  bus
IF Qi  Qi ,min set Qi Pi and Qi are known
And treat this bus as PQ-bus.  Qi ,min & |Vi |and δi are
unknown

Load flow solution when PV buses are present Reminder

For P Q 
For PV bus “ i ” : a. Calculate Qi Calc

In the polar form, n

Qi
|Vi |  |Yip ||V p | sin (  i   p  
For the (kth+1) ip ) i
iteration, Q ( k 1 )  | V 1 p1 || V ( k  1 ) | sin (  ( k )   ( k  1 ) 
)
 |Y
i speci ip ip
| p1 
i p i p
n
)
 |Y | sin (  
i speci ip p( k ) i (k ) i (k ) ip
 |V | p ||V  50
i
In rectangular form,

Q   Im{ V * (Y V  Y V  ...... Y V  ...Y V )}

i i i1 1 i2 2 in n
ii i

For each iteration:

For p  1 to ( i  1 ), use |V | &

For p  i to n, use |V | & 

Set |Vi ||Vi | speci

b. Check Qik+1 to see if it is within the limits

 Q
Qi ,min
Case 1: If the reactive power limits are not
violated,
• Use the most updated value of Qi
calculate Vi ( k 1) to calculate Ki .

Ki i n
Vi (k 1)
 L
1
ip V
(k 1)
p  L ip V (k )
p |Vi (k1) | (k
i1)
K i  Pi  jQi
Yii
 (Vi (k ) )* p
1
p i
1  51
 (k
Use | Vi | speci and i1) For the PV-bus voltage.

Voltage magnitude is known for PV
Reset the magnitude
|V i(k 1) |  | bus, therefore the new calculated
magnitude will not be used.

V ( k 1)  |V |
i i 
Speci  (k 1) Only the calculated angle
will be updated and
i
used.
Case 2: If the reactive power limits are violated,

IF Q i(k 1)  Qi,max set Qi(k  Qi,max

1)
Or

IF Q i(k 1)  Qi, set (k 1)

Consider this bus as a PQ-Bus, calculate bus voltage V ( k 1) i

Ki i n K i  Pi  jQi
Vi
(k
1)

 L
1
ip V (k
p1)  L ip V (k
p)
Yii
(Vi (k ) ) p
1
p i
1
* The PV-bus becomes PQ-bus
and both Voltage magnitude
Vi (k 1) |Vi (k 1) | i
(k 1) and angle are calculated and
used in coming iterations.52

EXAMPLE 2:

The line impedances are as indicated in per unit on 100MVA base. The line
charging susceptances are neglected. Using Gauss-Seidel method find the
power flow solution of the system. Assume the lower and upper limits of Q3 are
60 MVAR and 120 MVAR, respectively.

1 2
400
0.02  j0.04
MW
pu
G
0.01  j0.03 0.0125  j0.025 pu
pu 250
V1  1.050 pu
o
M

3 |V3 |  1.04
200
MW G

53
 1 2
1 400
0.02  j0.04
y12   10  j20  y21 pu
MW
0.02  j0.04
G
0.01  j0.03 0.0125  j0.025 pu
1 pu 250
y13   10  j30  y31 V1  1.050 o pu
MVA
0.01  R

j0.03
y 23   16  j32  y 32 3
1 j0.025
0.0125  |V 3 | 
1.04
200
MW G

Y11 Y13
 y12  y13 
Y12  y12  y13 
   y21  y23 
Ybus  Y21 Y Y23 
y 21  y 23
Y31
22
Y33    y 31  y 32 y 31  y 32 
Y32

 10  j20  10 
 20  j50
  10  j20 j30
Ybus 26  j52  16  
  10   16  j32 26  j62
j32 
j30 
54
Classification of buses: 1 2
400
0.02  j0.04
MW
pu
Bus 1: Slack Bus G
0.0125  j0.025 pu
0.01  j0.03
pu 250
V1  1.050 pu
o

V1  1.050o pu MVAR

3
Bus 2: Load Bus (PQ bus) |V3 |  1.04
200
MW G

P2 and Q 2 are known V2 and  2 are

unknown
 P2 ,d )  j( Q2 ,g  Q2
S2  pu
,sch ,d ) Base MVA

( P2 ,g
S2 ( 0  400 )  j( 0  pu
 100
,sch
250 )
  4  j2.5 pu
S 2 ,sch

Bus 3: Voltage Controlled Bus (PV bus)

|V3 | and P Q3 ,sch and  3 are

a unknown
P3 ,sch  2.0 pu S3, sch  2.0  j Q3, sch pu
55
Using GS method, select the initial values for the unknowns
as:
V1  1.050o pu V2( 0)  1 2
400
10 0.02  j0.04
pu
MW

G
| V3 |  (0)
3 0 o
V  1.050 pu
o
0.01  j0.03
pu
0.0125  j0.025 pu
250

1.04
1 MVAR

Start the first iteration

3 |V3 |  1.04
Bus 2 is PQ Bus
200
G
n MW
K2
V2   10   10

V 2*
 L2 p V p  20  j50
 10  j20 j20 
p1   10 26 
p 2 

Y23 (0)  K2
P2  jQ2  Y21 V     L
V 2(1)   V1 Y
3 21 1V 23 V 3(0) 
(0)
Y22 (V2 ) *  Y 22 22  (V 2(0) )*
L
K2 = - 0.0692 - j0.0423 L21 = - 0.3846 L23 = -0.6154

V2(1)  0.9746 - j0.0423 56

Bus 3 is PV Bus
Calculate and Check Q3 is within the limits  Q
Q3 ,min
Q (1)   Im{V * (Y V  Y V (1)  Y (0)
3 3
3
31 1 32 2 V33 )}

3  j1.1600
Q(1)
S3, sch  2.0  j1.16 pu

K3 = 0.0274 + L31 = -0.4690 + j0.0354 L32 = -0.5310 - j0.0354

j0.0208
V3 (1)  1.0378 - j0.0052  1.0378 -
0.2854o
Reset the magnitude |V3(1) ||V
Voltage magnitude is
|
fixed for a PV bus,
V3(1)  1.04-0.2854o therefore the new
calculated magnitude
V3 (1)  1.0400 - j0.0052 will not be used.
57
  20  j50  10  j 20  10  j 30 
 4 Y bus    10  j20 26  j52  16  
S 2 ,sch P3 ,sch  2   10  j30  16  26  j62
 j32  j32 1 
From 1st Iteration: 0.02  j0.04
pu
2
400
MW

V2  0.974
(1)
V1  1.050o pu
0.01  j0.03 pu 0.0125  j0.025 pu
250
MVAR

V3(1)  1.040 3 |V3 | 

1.04
200

MW G

K2 = - 0.0692 - j0.0423 K3 = 0.0274 +

L21 = - 0.3846 j0.0208 L31 =-0.4690 +

L23 = -0.6154 j0.0354

L32 = -0.5310 - j0.0354

Y 
P2  jQ2  Y21 1  23 V3(1)   K 2   L
V 2(2)  Y  V Y22
21 V 23 V 3(1) 
1
22 (V 2(1) )*  Y 22  (V 2(1) )*
L

Q 3(2)   Im{(V3 (1) )* (Y31 V1  32 V 2(2)  33 V 3(1) )}

Y Y

P3  jQ3  Y31 V1  32 V2(2)   K3   L
V 3(2)   Y33
31 1 V 32 V 2(2) 
Y Y (V )  Y33  (V3(1) )*
33 3 (1) *
58
L
Start the second iteration

Bus 2 is PQ Bus K
V 2( 2 )  2
(1) *
 L
(V

V2( 2 )  0.9711 - j0.0434

Bus 3 is PV Bus
Calculate and check Q3 is within the limits  Q
Q3 ,min
Q ( 2)   Im{V * (Y V  Y
3 3 31 1 32 (2) 2
V 33 (1) )}
V
3
Y
Q 3(2)  j1.3881(reset to the upper limit)  S  2.0  j1.2
3,sch
K3 pu
(2) (2)
 and L32 are constants and will be the
V3   L31 V1  L V
32 2 L31
same.
(V3 (1) )*
 K 3 is changed as Q3 change
P  jQ(2)
K3  3 3
Y
33

V3 ( 2 )  1.0391 - j0.0073  1.0391-

Reset the magnitude 0.4028o

V3 ( 2 )  1.04-0.4028o  1.0400 - j0.0073 59

 Newton-Raphson Algorithm
• The second major power flow solution
method is the Newton-Raphson algorithm
• Key idea behind Newton-Raphson is to use
sequential linearization
General form of problem: Find an x
f (xˆ) 

60
Newton-Raphson Method (scalar)
1. For each guess of xˆ, x (k ) , define
x (k )  xˆ - x (k )
2. Represent f (xˆ) by a Taylor series about f
(x) df (x (k ) )
(k )
f (xˆ)  f (x(k ) )  
x dx
1 d 2(kf) (x )

2 dx 2  x )(k 2
 higher order
terms

61
Newton-Raphson Method, cont’d
3. Approximate f (xˆ) by neglecting all
terms except the first two
df (x (k ) )
f (xˆ)  0  f (x(k )
(k )
)  x
dx
4. Use this linear approximation to solve for x(k )
1
   df (x )  f (x (k ) )
(k )
x (k ) 
 dx 

5. Solve for a new estimate of
xˆ
(k 1) (k ) (k )
 x  x
62
x
Example
Use Newton-Raphson to solve f (x)  x2 - 2
 0 The equation we must iteratively solve is
1
x (k )   df (x )  f (x (k ) )
(k )

 dx 

1
x (k )    ((x (k ) ) 2 -
2x (k )
2)
x (k 1)  x (k)  x (k )
1
x (k )    ((x (k ) ) 2 -
x (k 1)  2x (k )
2) 63
Cont’d Example
1
x (k 1)  x (k )   ((x (k ) ) 2 -
2x (k ) 
 2)
)  1. Iteratively solving we x (k )
(0)
Guess x(k
k x f (x (k )
)
get
0 1 1 0.5
1 1.5 0.25 0.08333
2 1.41667 6.953103 2.454 103
3 1.41422 6.024 106

64
Multi-Variable Newton-Raphson
Next we generalize to the case where x is an n-
dimension vector, and f (x) is an n-dimension
function
 x1   f1 (x) 
 x   
x  2 f (x)   f 2 (x) 
 M  M 
x 
 f (x)
 n  n 
Again define the solution xˆ so f (xˆ)  0
and
x  xˆ  x
65
 Multi-Variable Case, cont’d
The Taylor series expansion is written for each fi
(x) f1(x) f1(x)
f1 (xˆ )  f1 (x) x1  x2 K
x1 x2

f1 (x) xn  higher order terms
xn
M
f n (x) f n (x)
f n (xˆ) f n (x) x1  x2  K
 x1 x2
f n (x) x n  higher order terms
xn 66
 Multi-Variable Case, cont’d
This can be written more compactly in matrix form

 f1(x) f1(x) L f1 (x) 

 x1 x2 xn 
 1
 f1 (x)   f (x) f f 2 (x) x
2 2
 f (x)    (x) L   x 
f (xˆ )  2 x1 x2 xn  2 

 M     M
  M O O
M 
 f n (x) xn 
 f (x) f n (x) L f n (x) 
n
 1 x2 xn 

 higher x
order terms
67
 Jacobian Matrix
The n by n matrix of partial derivatives is known
as the Jacobian matrix, J(x)

 f1(x) f1(x) L f1 (x) 

x2 xn 
 
f 2 (x)
 f12 (x) f 2 (x) L
x
J(x)   x x2 xn 
 1  
 M O
O M 
 f (x) f n (x) 
n f n (x) L
 1 x2  xn 
x 68
Example 2
 x1 
Solve for x =   such that f (x)  0
x 2 
where
f1 (x)  2x 2  x 2  8  0
1 2
f 2 (x)  x 2  x 2  x x 40
1 2 1 2
First symbolically determine the Jacobian matrix

 f1 (x)
f1 (x) 

 x1x2 
J(x) =
 f 2 (x) f2 (x)

x2 

x1  69
Cont’d Example
 4x1 2x2 
J(x)
 2x  x1  2x2
 1
= x2 


x
 1
 4x1 2x2  1  f1 (x) 
Then
 x2     2x x1  2x2   f 2
   1  x2
 
1(x)
Arbitrarily guess x (0)
 
1 

 1  2.1
2 1    
(1)
x
1   43 1 3
   1.3 70
Cont’d Example

71
 Newton-Raphson Power Flow
• Advantages
– Fast convergence as long as initial guess is close to
solution
– Large region of convergence
• Disadvantages
– Each iteration takes much longer than a Gauss-Seidel
iteration
– More complicated to code, particularly when
implementing sparse matrix algorithms
• Newton-Raphson algorithm is very common in
power flow analysis

72
Newton-Raphson Summary

73
Newton-Raphson Summary

74
Newton-Raphson Summary

75
Newton-Raphson (NR) Power Flow
• Advantages
– fast convergence as long as initial guess is close to
solution
– large region of convergence
• Disadvantages
– each iteration takes much longer than a Gauss-Seidel
iteration
– more complicated to code, particularly when
implementing the Jacobian matrix
• NR algorithm is very common in power flow
analysis

76
NR Application to Power Flow
We first need to rewrite complex power equations
as equations with real coefficients
 n * n
i  ik*Vk
Si  Vi Ii   Y *

 Vi  k YikVk 
* V k
1 1
These can be derived by defining
Yik @ Gik  jBik
Vi @ Vi e ji  Vi i
ik @ i   k
Recall e j  cos  j sin  77
Real Power Balance Equations
n
Si  Pi  jQi   ik* k
nY *
V   Vi Vk e ik j
 (Gik  jBik )
k k
Vi 1 1
n
(cosik  j sin ik )(Gik  jBik )
 k Vi Vk
1
Resolving into the real and imaginary parts
n

Pi  k Vi Vk (Gik cosik  Bik sinik )  PGi  PDi

1
n

Qi   Vi Vk (Gik sin ik  Bik cosik )  QGi  QDi

k 1
78
 NR Power Flow
In the NR power flow we use Newton's
method to determine the voltage magnitude and angle
at each bus in the power system.
We need to solve the power balance equations
n

Pi   Vi Vk (Gik cosik  Bik sinik )  PGi  PDi

k n1

Qi   Vi Vk (Gik sin ik  Bik cosik )  QGi  QDi

k 1

79
 Power Flow Variables
Assume the slack bus is the first bus (with a fixed
voltage angle/magnitude). We then need to
determine the voltage angle/magnitude at the other
buses.
 2   P2 (x)  PG 2  PD2 
 
 M 
M
 
x   n  f (x)   Pn (x)  PGn  PDn 
 V 2  Q (x)  Q  Q 
2 G2 D2
 M 
 M 

 V n   Q (x)  Q 
n Gn Dn 
 Q 80
 NR Power Flow Solution

The power flow is solved using the following procedure:

Set k  0; make an initial guess of x, x(k )
While f   Do
(x(k ) )
x (k 1) x (k )  J(x (k ) ) 1 f

(x (k ) )

k1
k
81
 Power Flow Jacobian Matrix
The most difficult part of the algorithm is determining
and inverting the n by n Jacobian matrix, J(x)

 f1(x) f1(x) L f1 (x) 

 x2 xn 

f 2 (x)
 f12 (x) f 2 (x) L
x
J(x)   x x2 xn 
 1  
 M O
O M 
 f (x) f n (x) 
n f n (x) L
 1 x2  xn 
x 82
Power Flow Jacobian Matrix, cont’d
Jacobian elements are calculated by differentiating
each function, fi (x), with respect to each variable.
For example, if fi (x) is the bus i real power equation
n
fi (x)  Bik sinik )  PGi  PDi

 Vi Vk (Gik cosik
kn1
 Vi Vk (Gik sinik  Bikcos ik
fi(x)
i

k
1
)
k i
fi (x)  Vi V j (Gik ik  Bikcos ik ) (j
 j sin  i) 83
 Two Bus NR Example
For the two bus power system shown below, use the
NR power flow to determine the voltage magnitude
and angle at bus two. Assume that bus one is the
slack and SBase = 100 MVA.
Line Z = 0.1j

One 1.000 pu Two 1.000 pu

0 MW 200 MW
0 MVR 100 MVR

x    2  Ybus    j10 j10 

 V2  
 j10 84
 Two Bus Example, cont’d
General power balance equations
n

Pi  k Vi Vk (Gik cosik  Bik sinik )  PGi  PDi

1
n

Qi  k1 Vi Vk (Gik sin ik  Bik cos ik )  QGi  QDi

Bus two power balance equations
V2 V1 (10 sin2 )  2.0  0
V2 V1 (10 cos 2 )  V2 2 (10)
1.0  0
85
 Two Bus Example, cont’d
P2 (x) V2 (10sin2 )  2.0 
 0
V2 (10cos2 )  V2 2 (10) 1.0 
(x)  0 the Jacobian matrix:
Q2calculate
Now

 P2 (x)
  V2 
2
J (x)   
PQ (x)
(x) Q 2 (x)
 
2 2
 2 
 2

V
10sin2 
 10 V2 cos2
10 V2 sin 10cos 2  20 V2 
2 86
Two Bus Example, First Iteration
0
Set k  0, guess x (0)

1
Calculate

V2 (10sin 2 )  

f(x )   V
(0) 2.0 2   2.0
2 (10cos )  (10) 
2 
V2  1.0 
1.0  
10 V2 10sin2 
(0)
J(x )  
10 V2 sin 10cos 2  20 V2  10 0 
cos2 2
10 0 1  0
0.2
 
0  
      

Solve x(1)  
0  10 1.0 
1 2.0 10 
 0.9   87
Two Bus Example, Next Iterations
0.9 (10 sin(0.2)) 

  0.212
f(x )  0.9(10
(1) 2.0 2 
 cos(0.2))  0.9 10   0.279
1.0
 8.82 1.986 
J(x )  
(1)

1.788 8.199
0.2  1.9861 0.212
(2)    8.82  0.233
x 
    
 0.9  1.788 8.199  0.279  0.8586
  
 0.0145  (3)  0.236
f(x ( 2) )  x
 0.0190
  0.8554

(3)  0.0000906
  Done! V2  0.8554
f(x ) 
 0.0001175 13.52
 
88
 Two Bus Solved Values
Once the voltage angle and magnitude at bus 2 are
known we can calculate all the other system values,
such as the line flows and the generator reactive
power output
200.0 MW -200.0 MW
168.3 MVR Line Z = 0.1j -100.0 MVR

One 1.000 pu Two 0.855 pu -13.522 Deg

200.0 MW 200 MW
168.3 MVR 100 MVR

89
 Three Bus PV Case Example
For this three bus case we have

 2   P2 (x)  PG 2  PD 2
x    f   P3 (x)  PG3  PD3  
 3 
(x)
 V2 0  Q2 (x)  QD 2 
 
Line Z = 0.1j

0.941 pu
One 1.000 pu Two -7.469 Deg

170.0 MW 200 MW
68.2 MVR 100 MVR
Line Z = 0.1j Line Z = 0.1j

Three 1.000 pu

30 MW
63 MVR
90
 NR 5-bus Example

T2
800 MVA
1 T1 5 4 345/15 3 520 MVA
Line 3
345 kV kV
50 mi
400 MVA 800 MVA
15 kV 15 kV
400 MVA 345 kV 40 Mvar 80 MW
Line 2

Line 1
345 kV
15/345 kV 100 mi 200 mi

2
280 Mvar 800 MW

Single-line diagram

91
 NR 5-bus Example
V  PG QG PL QL QGmax QGmin
Bus Type per degrees per per per per per per
unit unit
unit
unit unit unit unit
Table 1.
1 Swing 1.0 0   0 0  
Bus input
data 2 Load   0 0 8.0 2.8  
3 Constant 1.05  5.2  0.8 0.4 4.0 -2.8
voltage
4 Load   0 0 0 0  
5 Load   0 0 0 0  

Maximum
R’ X’ G’ B’ MVA
Bus-to- per unit per unit per unit per unit per unit
Table 2.
Bus
Line input data
2-4 0.0090 0.100 0 1.72 12.0
2-5 0.0045 0.050 0 0.88 12.0
4-5 0.00225 0.025 0 0.44 12.0
92
 NR 5-bus Example

Maximum
R X Gc Bm Maximum TAP
per per per per MVA Setting
Table 3. Bus-to- unit unit per unit per unit
Transformer Bus unit unit
input data
1-5 0.00150 0.02 0 0 6.0 —
3-4 0.00075 0.01 0 0 10.0 —

Bus Input Data Unknowns

1 V1 = 1.0, 1 = 0 P1, Q1

2 P2 = PG2-PL2 = -8 V2, 2
Table 4. Input data Q2 = QG2-QL2 = -2.8
and unknowns 3 V3 = 1.05 Q3, 3
P3 = PG3-PL3 = 4.4
4 P4 = 0, Q4 = 0 V4, 4
5 P5 = 0, Q5 = 0 V5, 5
93
 NR 5-bus Example

One Five Four Three

A A

MVA MVA

395 MW A
520 MW
MVA

114 Mvar 337 Mvar

slack

1.000 pu 0.974 pu A A
1.019 pu 80 MW
0.000 Deg -4.548 Deg MVA MVA
-2.834 Deg 40 Mvar
1.050 pu
-0.597 Deg

0.834 pu Two
-22.406 Deg

800 MW
280 Mvar

94
 Good Power System Operation
• Good power system operation shouldn’t have reliability
violations for either the current condition or in the event
of statistically likely contingencies
• Reliability requires as a minimum that there be no
transmission line/transformer limit violations and that
bus voltages be within acceptable limits (perhaps 0.95 to
1.08)
• North American Electric Reliability Corporation now has
legal authority to enforce reliability standards (and there
are now lots of them). See
http://www.nerc.com for details (click on Standards)

95
37 Bus Example Design Case
Metropolis Light and Power Electric Design Case 2
A
SLA CK3 4
MVA 5
A

MVA

2 2 0 MW
1 .0 3 RA Y 3 4
5 2 Mv ar
pu 5 sla ck

System Losses: 10.70 MW A A A

1 .0 2 SLA CK1 3 8
T IM 3 4
M VA MVA

pu
MVA

5 1 .0 2 RA Y 1 3
A A
pu 8 A
1 .0 3
MVA
pu
MVA

T IM 1 3 3 3 MW A MVA

1 .0 0 1 .0 3
8 1 3 Mvar MV A
pu 1 5 .9 1 8 MW pu
A 1 .0 2 pu RA Y 6 9
A Mvar 5 Mvar 3 7 MW
MVA
A
1 7 MW A
MVA
P A I6 9 1 3 Mvar
1 .0 1 MVA 3 Mvar MVA
1 .0 2 T IM 6
pu
pu 9 A 1 .0 1 GRO SS6 9 A

2 3 MW pu
MVA
M VA
FERNA 6 9
7 Mvar A
1 .0 1 WO LEN6 9
1 2 MW
A A pu
HISKY 6 9 3 Mvar
MV A

MVA M VA
P ET E6 9 A
A
A
4 .9 Mvar
MO RO 1 3 MVA
5 8 MW A M VA

8 3 9 MW M VA
4 0 Mvar 1 .0 1 M VA

1 3 Mv ar pu 1 .0 0 BOB 1 3
1 2 MW
A

H A NNA H 6 9 2 8 .9 DEMA R6 9 pu 8
5 Mvar
A A

6 0 MW Mvar MVA

1 9 Mvar
MVA MVA

1 .0 0 2 0 MW
1 .0 0
A
1 .0 2 BOB 6
pu 1 2 Mvar
pu 0 .9 9 1 4 .2 UIUC 6 M VA pu 9
1 .0 0 pu
pu Mvar 9
1 2 .8 1 2 4 MW 5 6 MW
KYLE69
A A

Mvar 4 5 Mvar
A
M VA 1 3 Mv ar LYNN1 3 8
1 6 MW
MVA

M VA

A -1 4 Mv ar
2 5 MW A

1 4 MW
A
MVA
3 6 Mvar BL T 1 3
A MA NDA 6
MVA
1 .0 0 M VA 4 Mvar
0 .9 9 8
A
9
A pu
pu
M VA MVA
A
MVA SH IMKO 6 9 1 .0 2
HO MER6 9 2 5 MW 1 .0 1 7 .4 Mvar pu
A

1 0 Mvar pu BLT 6 9 M VA
A
1 .0 1 MVA

A
pu 1 5 MW
2 0 MW
MVA

H A LE6 9 5 5 MW 5 Mvar
3 Mvar M VA A

2 5 Mvar A

1 .0 0 MVA

pu 3 6 MW 1 .0 1
A A M VA

6 0 MW MVA 1 0 Mvar 7 .3 Mvar MVA pu A

A

MVA
1 2 Mvar
1 .0 0 1 .0 0 P A T T EN6 M VA

0 .0 Mvar M VA
A
pu pu
9 4 5 MW 1 4 MW ROGER6 9
MVA
1 .0 1 WEBER6 9 0 Mvar
LA UF6 9 2 Mvar
1 .0 2 pu
2 3 MW
pu 2 2 MW 1 0 MW
A A
6 Mvar 1 4 MW A

2 0 MW 1 5 Mv ar 5 Mvar
3 Mvar
2 8 Mv ar
MVA MVA MV A

JO 1 3 8 JO 3 4
LA UF1 3 1 .0 2
1 .0 0 SA V OY 6 9 1 .0 2 pu 3 8 MW 5
8 pu 3 Mvar
pu 1 .0 1 BUCKY 1 3 A

A pu 8 M VA A

150
MVA 1 .0 1 SA V OY 1 3 M VA
A A
MW
pu 8
MVA MVA 0 Mvar
150
A
MW
MVA 0 Mvar
1 .0 3
1 .0 2 pu A

pu
MV A

96
Looking at the Impact of Line Outages
Metropolis Light and Power Electric Design Case 2
A
SLA CK3 4
M VA 5
A

MV A

2 2 7 MW
1 .0 3 pu RA Y 3 4
4 3 Mvar
5 sla ck

System Losses: 17.61 MW A A A

1 .0 2 SLA C K1 3 8
T IM 3 4
M VA M VA

pu
MVA

5 1 .0 2 RA Y 1 3
A A
pu 8 A
1 .0 3
MVA
pu
M VA

T IM 1 3 3 3 MW A MVA

1 .0 1 1 .0 3
8 1 3 Mvar
pu
MV A
1 6 .0 1 8 MW pu
A 1 .0 2 pu RA Y 6
A Mvar 5 Mvar 37
9
M VA
MVA
A
1 7 MW A

P A I6 9 3 Mvar
1 .0 1 pu
MVA MVA
1 .0 2 T IM 6
pu 9 A 1 .0 1 GRO SS6 9 A

2 3 MW pu
MVA
MVA
FERNA 6 9
7 Mvar A

A A
1 2 MW
HISKY 6 9 3 Mvar
MV A

MVA MVA
P ET E6 9
A
4 .9 Mv
MORO 1 3 5 8 MW
8 3 9 MW M VA
4 0 Mvar 1 .0 1
1 3 Mvar pu
1 2 MW HA NNA H 6 9 2 8 .9 DEMA R6 9
5 Mvar
A

6 0 MW Mvar
1 9 Mvar
MVA

1 .0 0 2 0 MW
1 .0 0 pu
pu 0 .9 0 1 1 .6 UIUC 6 91 2 Mvar
1 .0 0
pu Mvar
pu
1 2 .8
KYLE69
A

A Mvar
MVA

M VA

2 5 MW
MVA
3 6 Mvar
A 0 .9 0 pu A MA NDA 6
110% 9

HO MER6 9 MVA
2 5 MW
1 0 Mvar
A
2 0 MW 135% H
3 Mvar MV A
0 .9 4
pu A

6 0 MW MVA

MVA
1 2 Mvar
0 .0 Mvar

1 .0 1 pu

2 0 MW
40 M

Opening one line (Tim69-Hannah69) causes an overload.

This would not be allowed 97
 Power Flow And Design
• One common usage of the power flow is to
determine how the system should be modified to
remove contingencies problems or serve new load
• In an operational context this requires working with the
existing electric grid
• In a planning context additions to the grid can be
considered
• In the next example we look at how to remove the
existing contingency violations while serving new
load.

98
 An Unreliable Solution
Metropolis Light and Power Electric Design Case 2
A
SLA CK3 4
M VA 5
A

MV A

2 6 9 MW
1 .0 2 RA Y 3 4
6 7 Mvar
pu 5 sla ck

System Losses: 14.49 MW A A A

1 .0 2 SLA C K1 3 8
T IM 3 4
M VA M VA

pu
MVA

5 1 .0 1 RA Y 1 3
A A
pu 8 A
1 .0 3
MVA
pu
M VA

T IM 1 3 3 3 MW A MVA

0 .9 9 1 .0 2
8 1 3 Mvar
pu
MV A
1 5 .9 1 8 MW pu
A 1 .0 2 pu RA Y 6
A Mvar 5 Mvar 37
9
MVA
A
1 7 MW A
M VA
P A I6 9
1 .0 1 MVA 3 Mvar MVA
1 .0 2 T IM 6
pu
pu 9 A 1 .0 1 GRO SS6 9 A

2 3 MW pu
MVA
MVA
FERNA 6 9
7 Mvar A

A A
1 2 MW
HISKY 6 9 3 Mvar
MV A

MVA A MVA
P ET E6 9
A
4 .9 Mv
MORO 1 3 96% 5 8 MW
8
MV A
3 9 MW M VA
4 0 Mvar 1 .0 1
1 3 Mvar pu
1 2 MW HA NNA H 6 9 2 8 .9 DEMA R6 9
5 Mvar
A

6 0 MW Mvar
1 9 Mvar
MVA

1 .0 0 2 0 MW
1 .0 0 pu
pu 0 .9 7 1 3 .6 UIUC 6 91 2 Mvar
1 .0 0
pu Mvar
pu
1 2 .8
KYLE69
A

A Mvar
MVA

M VA

2 5 MW
MVA
3 6 Mvar
A
0 .9 7 pu
2 5 MW
HO MER6 9 1 0 Mvar
MV A

A MA NDA 6
A 9
2 0 MW
H
3 Mvar MVA

0 .9 9
pu A

6 0 MW MVA

MVA
1 2 Mvar
0 .0 Mvar

1 .0 2 pu

2 0 MW
40 M

99
 A Reliable Solution
Metropolis Light and Power Electric Design Case 2
A
SLA CK3 4
MVA 5
A

MVA

2 6 6 MW
1 .0 2 RA Y 3 4
5 9 Mvar
pu 5 sla ck

System Losses: 11.66 MW A A A

1 .0 2 SLA C K1 3 8
T IM 3 4
MVA MVA

pu
M VA

5 1 .0 1 RA Y 1 3
A A
pu 8 A
1 .0 3
M VA
pu
M VA

T IM 1 3 3 3 MW A MVA

1 .0 0 1 .0 3
8 1 3 Mvar MV A
pu 1 5 .8 1 8 MW pu
A 1 .0 2 pu RA Y 6 9
A Mvar 5 Mvar 3 7 MW
M VA
A
1 7 MW A
MVA
P A I6 9 1 3 Mvar
1 .0 1 MVA 3 Mvar MVA
1 .0 2 T IM 6
pu
pu 9 A 1 .0 1 GRO SS6 9 A

2 3 MW pu
MVA
M VA
FERNA 6 9
7 Mvar A
1 .0 1 WO LEN6 9
1 2 MW
A A pu
H ISKY 6 9 3 Mvar
MVA

MVA MVA
P ET E6 9 A
A
A
4 .9 Mvar
MORO 1 3 M VA
5 8 MW A M VA

8 3 9 MW MVA
4 0 Mvar 1 .0 1 M VA

1 3 Mvar pu 1 .0 0 BOB 1 3
1 2 MW
A

HA NNA H 6 9 2 8 .9 DEMA R6 9 pu 8
5 Mvar
A A

6 0 MW Mvar MVA

1 9 Mvar
M VA M VA

Kyle138 2 0 MW
0 .9 9
A
1 .0 2 BO B 6
pu 0 .9 9 1 4 .1 UIUC 6 91 2 Mvar
1 .0 0
MVA pu 9
pu Mvar
A

M
VA
pu
1 2 .8 1 2 4 MW 5 6 MW
KYLE69
A A

Mvar 4 5 Mvar
A
M VA 1 3 Mvar LYNN1 3 8
1 6 MW
MVA

MVA

A -1 4 Mvar
2 5 MW A

1 4 MW
A
MVA
3 6 Mvar BLT 1 3
M VA
1 .0 0 MVA 4 Mvar
0 .9 9 8
A A
pu
pu
MVA 2 5 MW MVA
A
MVA SH IMKO 6 9 1 .0 2
H OMER6 9 1 0 Mvar 1 .0 1 7 .4 Mvar pu
A

A MA NDA 6 pu BL T 6 9 M VA
A
1 .0 1 M VA

A 9 pu 1 5 MW
2 0 MW
M VA

HA LE6 9 5 5 MW 5 Mvar
3 Mvar MVA A

2 9 Mvar A

1 .0 0 MVA

pu 3 6 MW 1 .0 1
A A MVA

6 0 MW MVA 1 0 Mvar 7 .3 Mvar M VA pu A

A

MVA
1 2 Mvar
1 .0 0 1 .0 0 P A T T EN6 M VA

0 .0 Mvar M VA
A
pu pu
9 4 5 MW 1 4 MW RO GER6 9
1 .0 1 WEBER6 9 0 Mvar
MVA

LA UF6 9 2 Mvar
1 .0 2 pu
2 3 MW
pu 2 2 MW 1 0 MW
A A
6 Mvar 1 4 MW A

2 0 MW 1 5 Mvar 5 Mvar
3 Mvar
3 8 Mvar
MVA MV A MV A

JO 1 3 8 JO 3 4
LA UF1 3 1 .0 2
1 .0 0 SA VO Y 6 9 1 .0 2 pu 3 8 MW 5
8 pu 4 Mvar
pu 1 .0 1 BUC KY 1 3 A

A pu 8 MVA A

150
MVA 1 .0 1 SA VO Y 1 3 MVA
A A
MW
pu 8
MVA MVA 1 Mvar
150
A
MW
M VA 1 Mvar
1 .0 3
1 .0 2 pu A

pu
MV A

Previous case was augmented with the addition of a

138 kV Transmission Line 100