POWER FLOW ANALYSIS

Power flow analysis assumption

steady-state
balanced single-phase network
network may contain hundreds of nodes and
branches with impedance X specified in per unit on
MVA base

Power flow equations

bus admittance matrix of node-voltage equation is

formulated
currents can be expressed in terms of voltages
resulting equation can be in terms of power in MW

Energy Conversion Lab

BUS ADMITTANCE MATRIX

Nodal solution

nodal solution is based on

the Kirchhoffs current law
impedance is converted to
admittance
yij =

1
1
=
Z ij rij + jxij

Bus admittance equations

the impedance diagram:

see Fig.6.1

Energy Conversion Lab

BUS ADMITTANCE MATRIX

Bus admittance
equations

the admittance is
based on bus-tobus: see Fig.6.2
if no connection
between bus-tobus, leave as zero
node voltage
equation is in the
form
I bus = YbusVbus

Energy Conversion Lab

BUS ADMITTANCE MATRIX

Node-voltage matrix

Ibus=YbusVbus
Y11 Y12 Y1i Y1n
V1
I1

I Y21 Y22 Y2i Y2n V

2 2
=

Yi1 Yi2 Yii Yin
I i Vi
I n
Vn
Yn1 Yn2 Yni Ynn
Ibus is the vector of injected currents
Vbus is the vector of the bus voltage from reference
node
Ybus is the bus admittance matrix

Energy Conversion Lab

BUS ADMITTANCE MATRIX

Node-voltage matrix

diagonal element Yii: sum of admittance connected to bus i

n

Yii = yij

ji

j =0

off-diagonal matrix Yij: negative of admittance between nodes I

and j

Yij = Y ji = yij

when the bus currents are known, bus voltages are unknown,
bus voltage can be solved as
1
Vbus = Ybus
I bus

inverse of bus admittance matrix is known as impedance matrix

Zbus
1

Z bus = Ybus

if matrix of Ybus is invertible, Ybus should be non-singular

Energy Conversion Lab

BUS ADMITTANCE MATRIX

Node-voltage matrix

admittance matrix is symmetric along the leading diagonal,

which result in an upper diagonal nodal admittance matrix
a typical power system network, each bus is connected by a few
nearby bus, which cause many off-diagonal elements are zero
many zero off-diagonal matrix is called sparse matrix
the bus admittance matrix in Fig.(6.2) by inspection is

Ybus

j8.5 j 2.5 j5.0

j 2.5 j8.75 j5.0
=
j5.0 j5.0 -j 22.5

0
0 j12.5

Energy Conversion Lab

j12.5

j12.5
0
0

SOLUTION OF NONLINEAR ALGEBRA EQUATIONS

Techniques for iterative solution of non-linear

equations

Gauss-Seidal
Newton-Raphson
Quasi-Newton

Gause-Seidal method

consider a nonlinear equation f(x)=0

rearrange f(x) so that x=g(x), f(x)=x-g(x) or
f(x)=g(x)-x
guess an initial estimate of x = x(k)
use iteration, obtain next x value as x(k+1) = g(x(k))
criteria for stop iteration: |x(k+1)-x(k)|
is the desired accuracy

Energy Conversion Lab

GAUSE-SEIDAL METHOD
Nature of Gause-Seidal method

see Ex.(6.2) and Fig.(6.3)

Gause-Seidal method needs many iterations to
achieve desired accuracy
no guarantee for the convergence, depend on the
location of initial x estimate

Energy Conversion Lab

GAUSE-SEIDAL METHOD

Nature of Gause-Seidal method

solution: if initial estimate x is within convergent

region, solution will converge in zigzag fashion to one
of the roots
no solution: if initial estimate x is outside convergent
region, process will diverge, no solution found
in some case, an acceleration factor is added to
improve the rate of convergence:
x(k+1) = x(k) +[g(x(k))-x(k)], where >1
acceleration factor should not too large to produce
overshoot
see Ex.(6.3) for the acceleration factor used

Energy Conversion Lab

GAUSE-SEIDAL METHOD

Extend one variable to n variable equations using GauseSeidal method

consider the system of n equations in n variables and solving for

one variable from each equation in one time of iteration

f1 ( x1 , x2 , , xn ) = c1

x1 = c1 + g1 ( x1 , x2 , , xn )

f 2 ( x1 , x2 , , xn ) = c2

x2 = c2 + g 2 ( x1 , x2 , , xn )

f n ( x1 , x2 , , xn ) = cn

xn = cn + g n ( x1 , x2 , , xn )

the updated variable x1(k+1) calculated in first equation in

Eq.(6.12) is used in the calculation of x2(k+1) in the second
equation
Ex: in the 2nd iteration x2(k+1) = c2+g2(x1(k+1)+x2(k)+x3(k)++xn(k))
at n iteration to complete n variables, the x1(k+1),,xn(k+1) is
tested against x1(k),,xn(k) for accuracy criterion

Energy Conversion Lab

POWER FLOW SOLUTION

Power Flow (Load Flow)

operating condition: balanced, single phase model

quantities used in power flow equation are: voltage
magnitude |V|, phase angle , real power P, and
reactive power Q
system bus classification:

slack bus (swing bus): taken as reference where |V| and

V are specified. It makes up the loss between generated
power and scheduled loads
load bus (PQ bus): P and Q are specified, |V| and V are
unknown
regulated bus (PV bus): P and |V| are specified, V and Q
are unknown

Energy Conversion Lab

POWER FLOW EQUATION

Power flow formulation

consider bus case in Fig.(6.7)

current flow into bus i:
n

j =0

j =1

I i = Vi yij yijV j

ji

express Ii in terms of P,Q:

Pi jQi
Ii =
Vi *

the power flow equation becomes

n
n
Pi jQi
= Vi yij yijV j
*
Vi
j =0
j =1

ji

the power flow problem results in algebraic nonlinear equations

which must be solved by iteration methods

Energy Conversion Lab

GAUSS-SEIDEL POWER FLOW EQUATION

Gauss-Seidel power flow solution

solving Vi: for PQ bus, assume P,Q are known

Vi ( k +1)

ji

solving Pi: for slack bus, assume V is known

Pi ( k +1)

Pi sch jQisch
(k )
+
y
V

ij k
Vi *( k )
=
yij

n
n

= Re Vi *( k ) Vi ( k ) yij yijV j( k )

j =0
j =1

ji

solving Qi: for PV bus, assume |V| is known

Qi( k +1)

Energy Conversion Lab

n
n
*( k ) ( k )
(k )
V
yij yijV j
= Im Vi

i
j =0
j =1

j i

ji

GAUSS-SEIDEL POWER FLOW EQUATION

Instructions for Gauss-Seidel solution

there are 2(n-1) equations to be solved for n bus

voltage magnitude of the buses are close to 1pu or
close to the magnitude of the slack bus
voltage magnitude at load buses is lower than the slack
bus value
voltage magnitude at generator buses is higher than
the slack bus value
phase angle of load buses are below the reference
angle
phase angle of generator buses are above the
reference angle

Energy Conversion Lab

INSTRUCTIONS FOR G-S SOLUTION

Instructions for PQ bus solution

real and reactive power Pisch, Qisch are known

starting with an initial estimate of voltage using Vi
equation
P sch jQ sch
i

Vi

( k +1)

Vi *( k )

+ yijV j( k )

yij

ji

Instructions for PV bus solution

Pisch, |Vi| are specified

assume Vi = |Vi|0o, solve the Qi equation as below
Qi( k +1)

Energy Conversion Lab

n
n

= Im Vi *( k ) Vi ( k ) yij yijV j( k )

j =0
j =1

j i

ji

INSTRUCTIONS FOR G-S SOLUTION

Instructions for PV bus solution

when Qi(k+1) is available, solve Vi using equation below

Vi ( k +1)

Pi sch jQi( k )
+ yijVk( k )
*( k )
Vi
=
yij

ji

since |Vi| is specified, keep imaginary part of Vi,

calculate real part of Vi
Re{Vi ( k +1) } =

solve Vi

stopping criteria

Energy Conversion Lab

Vi

(imag {Vi ( k +1) })

Vi ( k +1) = Re{Vi ( k +1) }+ j Im{Vi ( k +1) }

Re{Vi ( k +1) } Re{Vi ( k ) } ,

Im{Vi ( k +1) } Im{Vi ( k ) }

INSTRUCTIONS FOR G-S SOLUTION

Instructions for PV bus solution

to accelerate the convergence, using the following

approximation after new Vi is obtained
k)
Vi ( k +1) = Vi ( k ) + (Vi (cal
Vi ( k ) )

is in the range between 1.3 to 1.7

voltage accuracy in |Vi| and is in the range between
0.00001 to 0.00005

Energy Conversion Lab

INSTRUCTIONS FOR G-S SOLUTION

Instructions for V, slack bus solution

solve Pi

n
n
*( k ) ( k )
(k )
V
yij yijV j
= Re Vi

i
j =0
j =1

j i

ji

n
n
*( k ) ( k )
(k )
V
yij yijV j
= Im Vi

i
j =0
j =1

ji

Pi ( k +1)

solve Qi
Qi( k +1)

accuracy: the largest PQ is less than the

specified value, typically is about 0.001pu

Energy Conversion Lab

G-S Power flow Homework

For the one-line diagram shown below, using the G-S method
to determine all bus voltages (magnitude and phase) and
show the power flow solution between the buses assume the
regulated bus (#2) reactive power limits are between 0 and
600Mvar.

NEWTON RAPHSON METHOD

Newton Raphson method for solving one variable

consider the solution of one-dimensional equation f(x)=c

assume x = x(0)+x(0)
f(x)=f(x(0)+x(0))=c
use Taylors series expansion
1 d2 f
df
( 0)
( 0)
( 0)
( 0)
f ( x + x ) = f ( x ) + x + 2
2! dx
dx
( 0)

( 0)

(x )

( 0) 2

+ = c

assume x(0) is very small, higher order terms of expansion can

be neglected, Taylor series becomes
( 0)

f (x

( 0)

df
+ x ) = f ( x ) + x ( 0 ) = c
dx
( 0)

( 0)

assume f(x(0))=c-c(0), the equation becomes c(0)(df/dx)(0)x(0)

( 0)

c
(
1
)
(
0
)
the new approximation of x x = x +
( 0)
df

dx

Energy Conversion Lab

NEWTON RAPHSON METHOD

Newton Raphson method for solving one variable

the new approximation of x

x =x +
(1)

( 0)

c ( 0 )

df

dx

( 0)

Newton Raphson algorithm

c ( k ) = c f ( x ( k ) )

(k )

c ( k )
(k )

df

dx
x ( k +1) = x ( k ) + x ( k )
for more information, see Ex.(6.4)
Newtons method converges faster than Gauss-Seidal, the root
may converge to a root different from the expected one or
diverge if the starting value is not close enough to the root

Energy Conversion Lab

NEWTON RAPHSON METHOD FOR n VARIABLES

Newton Raphson method for solving n variables

( 0)

( 0)

( 0)

f1
f1
f1
( 0)
( 0)
( 0)
( 0)
( 0)
xn ( 0 ) = c1
x2 + +
f1 ( x + x ) = f1 ( x ) + x1 +
x1
x2
xn
( 0)

( 0)

( 0)

( 0)

( 0)

( 0)

f 2
f 2
f 2
( 0)
( 0)
( 0)
( 0)
( 0)
xn ( 0 ) = c2
x2 + +
f 2 ( x + x ) = f 2 ( x ) + x1 +
x1
x2
xn
f
f
f
( 0)
( 0)
( 0)
f n ( x ( 0 ) + x ( 0 ) ) = f n ( x ( 0 ) ) + n x1 + n x2 + + n xn = cn
x1
x2
xn

Energy Conversion Lab

NEWTON RAPHSON METHOD FOR n VARIABLES

Rearrange in matrix form

f ( 0 )
1
x1
( 0)
c1 f1
( 0)

f 2
( 0)
c
f

2 2 =
x1

( 0)
cn f n
( 0)
f n
x
1

f1

x2

( 0)

f 2

x2

( 0)

f n

x2

( 0)

The matrix can be written as

C(k) = J(k) X(k)

Energy Conversion Lab

f1

xn ( 0 )
x1
( 0)
f 2 ( 0 )
x2

xn

( 0)
xn
( 0)
f n

xn
( 0)

NEWTON RAPHSON METHOD FOR n VARIABLES

The Newton-Raphson algorithm for ndimensional case is

X(k+1) = X(k) +X(k) = X(k) + [J(k)]-1C(k)

where
X ( k )

C ( k )

x1( k )
(k )
x
= 2

(k )
xn

c1 f1( k )

(k )
c
f

= 2 2

(k )
cn f n

Energy Conversion Lab

J (k )

f ( k )
1
x1

(k )
f 2
= x1

(k )

n
x
1

f1

x2

(k )

f 2

x2

(k )

f n

(k )

(k )
f1

xn
(k )
f 2

xn

(k )
f n

xn

NEWTON RAPHSON METHOD FOR n VARIABLES

The Newton-Raphson algorithm

J(k) is called the Jacobian matrix

solution to X(k+1) is inefficient because it involves
inverse of J(k) , a triangular factorization is used to
facilitate the computation
in MATLAB, the operator \ (i.e., X=J\C) is used to
apply the triangular factorization
Newton-Raphson method converge to solution
quadratically when near a root
The limitation is that it does not generally converge to
a solution from an arbitrary starting point

Energy Conversion Lab

LINE FLOWS AND LOSSES

Complex power flow between bus i,j

for line model, see Fig. 6.8

current flow from bus i to bus j
I ij = I l + I i 0 = yij (Vi V j ) + yi 0Vi
current flow from bus j to bus i
I ji = I l + I j 0 = yij (V j Vi ) + y j 0V j
complex power Sij from bus i to j and Sji from j to i
Sij = Vi I ij*

S ji = V j I *ji

power loss in the line i-j

S L ( i j ) = Sij + S ji
for more Gauss-Seidel method examples, see Ex. (6.7)
and Ex. (6.8)

Energy Conversion Lab

NEWTON-RAPHSON POWER FLOW

Real power flow in terms of Vi , , and Yij

Pi =

j =1

V j Yij cos( ij i + j )

Reactive power flow

Qi = Vi V j Yij sin (ij i + j )

n

j =1

Newton-Raphson matrix form: C(k) = J(k) X(k)

J 2

J 4 V

P J 1
Q = J
3

diagonal and off-diagonal elements of J1

Pi
=
i

V
j i

V j Yij sin (ij i + j )

Pi
= Vi V j Yij sin ( ij i + j )
j
Energy Conversion Lab

ji

NEWTON-RAPHSON POWER FLOW

Newton-Raphson matrix form: C(k) = J(k) X(k)

J 2

J 4 V

P J 1
Q = J
3

diagonal and off-diagonal elements of J2

Pi
= 2 Vi Yii cos ii + V j Yij cos(ij i + j )
Vi
j i
Pi
= Vi Yij cos( ij i + j )
Vj

ji

diagonal and off-diagonal elements of J3

Qi
=
i

V
j i

V j Yij cos( ij i + j )

Qi
= Vi V j Yij cos( ij i + j )
j
Energy Conversion Lab

ji

NEWTON-RAPHSON POWER FLOW

Newton-Raphson matrix form: C(k) = J(k) X(k)

P
Q =

J1
J
3

J 2

J 4 V

diagonal and off-diagonal elements of J4

Qi
= 2 Vi Yii sin ii V j Yij sin ( ij i + j )
Vi
j i
Qi
= Vi Yij sin ( ij i + j )
Vj

power residuals Pi(k) Qi(k)

Pi ( k ) = Pi sch Pi ( k ) ,

ji

Qi( k ) = Qisch Qi( k )

new estimates for bus voltages

i( k +1) = i( k ) + i( k ) ,
Vi ( k +1) = Vi ( k ) + Vi ( k )

Energy Conversion Lab

NEWTON-RAPHSON POWER FLOW

Procedure for Newton-Raphson method:

(0)|=1.0, (0)=0.0
PQ bus: set |Vi
i
(0)=0.0
PV bus: set i
set PQ bus equation for J matrix elements:
Pi =

j=
n1

V j Yij cos( ij i + j )

Qi = Vi V j Yij sin (ij i + j )

j =1

Pi ( k ) = Pi sch Pi ( k ) ,

Qi( k ) = Qisch Qi( k )

set PV bus equation for J matrix elements:

Pi =

V
j =1

V j Yij cos( ij i + j )

Pi ( k ) = Pi sch Pi ( k )
Energy Conversion Lab

NEWTON-RAPHSON POWER FLOW

Procedure for Newton-Raphson method:

use above equation to calculate Jacobian matrix (J1, J2,
J3, J4)
solve |V| and using Newton-Raphson matrix
P
Q =

J 2

J 4 V

update |V| and by

i( k +1) = i( k ) + i( k ) ,
Vi ( k +1) = Vi ( k ) + Vi ( k )
repeat the calculation until
Pi ( k ) ,

J1
J
3

Qi( k )

for example: see Ex.(6.10)

Energy Conversion Lab

FAST DECOUPLED POWER FLOW

Fast decoupled power flow solution:

the algorithm is based on Newton-Raphson method
when transmission lines has a high X/R ratio, the
Newton-Raphson method could be further simplified

Consider the Newton-Raphson power flow

equation

P
Q =

J1
J
3

J 2

J 4 V

P are less sensitive to |V| and most sensitive to

Q is less sensitive to and most sensitive to |V|
we can reasonably eliminate J2 and J3 elements in
Jacobian matrix

Energy Conversion Lab

FAST DECOUPLED POWER FLOW

Consider the Newton-Raphson power flow equation

the power flow equation reduces to

P
Q =

J1
0

J 4 V

P = J1 = [P/], Q = J4|V| = [Q/|V|]|V|

Pi/i = -Qi - |Vi|2Bii, Bii = |Yii|sinii is the imaginary part of the
diagonal elements
since Bii >> Qi, Pi/i (diagonal elements of J1) can be further
reduced to Pi/i = - |Vi|Bii (|Vi|2 |Vi| )
off diagonal element of J1: Pi/i = - |Vi||Vj|Yijsin(ij-i+j), since j-i
is quite small, ij-i+j = ij, J1 = Pi/j = - |Vi||Vj|Bij
since |Vj|1, off diagonal elements of J1 = Pi/j = - |Vi|Bij

Energy Conversion Lab

FAST DECOUPLED POWER FLOW

Consider the Newton-Raphson power flow

equation

similarly, diagonal elements of J4: Qi/|Vi| = - |Vi|Bii

off diagonal elements of J4: Qi/|Vj| = - |Vi|Bij
therefore, P and Q has the following forms
P
= B ' ,
Vi

B and B are the imaginary part of Ybus

the updated and |V| can be obtained from
= [B']

Q
= B '' V
Vi

P
,
V

V = [B"]

Q
V

to calculate PQ bus, use simplified J1 and J4 to obtain

solution
to calculate PV bus, J4 can be further eliminated, only
J1 is used to obtain solution

Energy Conversion Lab

FAST DECOUPLED POWER FLOW

Comparison between fast decouple power flow

solution and Newton Raphson power flow solution

fast decoupled solution requires more iterations than

Newton Raphson solution
fast decoupled solution requires less time per iteration
since decoupled solution needs less time for iteration,
the overall computation time may be less than using
the Newton Raphson method
fast decoupled solution often used in fast computation
of power flow, for example, contingency analysis or online control of power flow
see Ex. 6.12

Energy Conversion Lab