100 Electrical Power Systems: Analysis, Security and Deregulation

We also have the equation similar to Eq. (2.32)

x(k*1) =xW- AX) (2.38)
The state update vector Ax) is calculatcd from Eq. (2.33) by taking the
inverse of the Jacobian matrix. Thus we get
x0-J'F (2.39)
With Eqs. (2.34) and (2.38) solved alternately in each iteration, xt-) gradually
approaches the actual solution, Convergence can be evaluated by the norm of
the correction value,
JAX4<e (2.40)

2.6.2 Load Flow Solution Using Newton-Raphson

Method
For large interconnected power systems among the numerous solution methods
available for load flow analysis, the Newton-Raphson method is considered to
be the most important. Many advantages are attributed to the Newton-Raphson
approach. Its convergence characteristics are relatively powerful comparcd to the
alternative processes, and very low computing timcs are achicved when sparse
network equations are solved by the technique of sparsity programmed ordered
climination. The geliabilityof the Newton- Raphson method is comparativcly
good, since it can solve cases that lead to divergence with the other popular
processes, but the method is by no means reliable. Failure does not occur on
some ill-conditioned problems.
The number of iterations required to obtain a solution is independerit of
the system size, but more functional evaluations are required at cach iteration.
Since in the load flow problem real power and magnitude of bus voltage are
specified for the PY'buses, the load tow cquation is fomulated in the polar
form.
The load ow equations can be rewritten as follows.
Real power

jl
N,|cos(0, +8,-&) (241)

Reactive power
, sin (0, +&, -8) (2.42)

We have two equations for cach load bus, given by Eqs. (2.41) and (2.42),
and onc cquation for cach voltage controlled bus, given by Eq. (2.41). Expanding
Eqs. (2.41) and (2.42) in Taylor's series about the initial estimate and neglecting
all higher order tems result in the following set of linear equations.
 LoadFlow Analysis 101

Ap S, (2.43)

S,
the slack bus. The Jacobian
In the above cquation, bus 1 is assumed to be angle
matrix gives the linearised relationship between small changes in voltage
A8 and voltage magnitude A| with the small changes in real and
powers A" and AQrespectively. The elements of the Jacobian
reactive
(2.42), calculated at A84)
and
matrix are the partial derivatives of Eqs. (2,41)
and Ar:
The above cquation canbe written as
(2.44)

PY buses, the voltage magnitudes are known. Therefore, if m buses

For the equations involving AQ and AV, and the
of the system are voltage controlled
the Jacobian matrix are eliminated, then there are
corresponding columns of
power constraints and (n - 1- m) reactive power constraints, and
(n-1) real m) x (2n - 2- m).
matrix is (2n - 2-
the order of the complete Jacobian
1) x (n - ).
Order of Jacobian matrix J, is (n - 1) (n - 1- m).
Order of Jacobian matrix J, is (n- x -m) x (n - l).
Order of Jacobian matrix J, is (n - 1 m) x (n -1-m).
Order of Jacobian matrix J, is (n - -
Elements of Jacobian matrix J,
(i) the diagonal elements are
(2.45)
XVlV,,sin (0,-6, +8,)
are
(ü) the off-diagonal-clements
(2.46)
l|y,| sin(0, -8, +8,) jzi
 102 Electrical Power Systems: Analysis, Security and Deregulation

Elements of Jacobian matrix J,

(i) the diagonal elements are

-2V|1X,|cos e, +V,,l cos(@, -8, +8) (2.47)

j*i
(ii) the off-diagonal elements are

=|V;llI cos (0, - 8, +8,) j:j#i (2.48)

Elements of Jacobian matrix J

() the diagonal elements are

dQ.SV|cos (0, - 6,+6)

jei
(2.49)

(ii) the off-diagonal elements are

=V,,| cos(e, - &, +6,) j#i (2.50)

Elements of Jacobian matrix J,

(i) the diagonal elements are
;=-2VlY|in ,-V|sin (0, - 8 +8 (2.51)
j#i
(ii) the off-diagonal elements are
iV| sin (0, - 8, +8,) j#i (2.52)

Difference in scheduled to calculated power (power residuals) is given by

(2.53)
(2.54)
The new estimates for the voltage magnitude and angle
(2:55)
(2.56)

Computation procedure
1. Set flat start
slack bus or 1Z0°.
" For load buses, set the voltages equal to the the slack bus or 0°.
" For generator buses, set the angles equal to
 Load Flow
Analysis 103
2. Calculate power mismatch
. For load buses, calculate pl) (Eq. (2.41)) and o4l
injections using the known and estimated system voltages. (Eq (242)
For generator buses, calculate pk1 (Eq. (2.41)
(Eq. (2.53)). and ap
3. Form the Jacobian matrix
" Use the various equations for the partial derivatives
with
to the voltage angle and magnitudes (form the Jacobian respect
" The elements of Jacobian matrix (J, J, J, and J) matrix),
from Eqs. (2.45) to (2.52). calculated
4. Find the matrix solution
. Inyerse the Jacobian matrix and multiply by the mismatch Douer
Compute AÕand A|V.
s. Difference in scheduled to calculated power

**6. Find thë new estimätes for the voltage magnitude and angle

7. Repeat the process until the mismatch (residuals) is less than the
specified accuracy
A|se
ase
8. After solving for bus voltages and angles, power flows and losses an
the network branches are calculated
Transimission lines and transformers are network branches.
" The direction of positive current flow is defined for a branch
element (demonstrated on amedium length line).
" Power flow is defined for each end of the branch.
Example: The power leaving bus i and flowing to busj as shou
below.
Bus i Busj
104 Electrical Power Systems: Analysis, Security and Deregulation
Current and power flows
i’j

j’i

Power loss
Sossi Syt S
This completes the load flow study. Finally, in Figure 2.13 all the computational
steps are summarized in the detailed flow chart.

2.6.3 Advantages and Disadvantages of

Newton-Raphson Method
Advantages less, number of
Faster, more reliable and yields accurate results, requires
iterations.
Disadvantages
Programn as well as memory is more complex.
EXAMPLE 2.6 Figure 2.14 shows the one line diagram of a simple three
of voltage at bus 1 is
magnitude
bus system with generation at bus 1. The buses 2 and 3 are given in the
adjusted to"1.05 p.u. The scheduled loads at
a 100 MVA base and the line
diagram. Line impedances are marked in p.u., on
charging susceptances are negiected.
Using the Newton-Raphson method, determine the phasor values of
(a) 3(PO buses) accurate to decimal
the voltages at the load buses 2 and
places. World Simulator.
(b) Verify the result with Power
1
2
0.02 + j0.04
+ 400 MW

0.01 +j0.03 0.0125 +j0.025 250 MVAR

Slack bus
V=1.05Z0 3

200 MW
Example 2.6.
diagram of a simple three-bus system
 Load Flow Analysis 105

Start

Read system data

Form Ybus

Initialize magnitude and phase angle of bus voltages |, 6

Set iteration count k= 0

Bust count i= I

i=i+1

Calculate,
p o
No
Is refer PV bus

Yes
sch
Calculate, AP= P-P
Calculate, AP = P. -P.

inax
0<min
Check limit

ischimax
O.imin

Yes
Ls i<n
No
Form Jacobian matrix
OP P
S...LAPI

i+a
Yes Calculate power
Check JAP, 4¢se flow and losses

No End
Set k = k+l
Newton-Raphson method.
chart for
Figure 2.13 Flow
 106 Electrical Power Systems: Analysis, Security and Deregulation

Solution:
(a) Form the Yus

0.02 + j0.04
=10- j20
Z12
1
=10- j30
13 0.01+ j0.03
=16- j32
Z23 0.0125 + j0.025
Y, =yi2 t yj3 = (10-j20) + (10-j30) = 20 -j50
Y2 =Y =-} =-(10 -j20) = -10 +j20
Y =Ys =-y =-(10-j30) =-10 +j30
Y221 t'3 = (10 -j20) + (16 -j32) = 26 - j52
Yg =Ys =- =-(16 -j32) =-16 +j32
Y33 31t32 = (10 -j30) + (16 - j32) = 26 - j62
20- j50 -10+j20 -10+j30|
=-10 + j20 26- j52 -16+ j32
Yus -10 +j30-16+ j32 26 j62
[53.85165Z-1.9029 22.3606822.03443.62278Z1.8925
Ypus 22.36068/2.0344 58.137772-1.1071 35.77709{2.0344
31.622781.8925 35.7770922.0344 67.23095Z-1.1737
[53.85165Z-68.2 22.36068Z116.6- 31.62278Z108.4

Yus 22.36068A16.6 58.137772-63.4 35.77709116.6

31.62278108.4 35.77709Z116.6 67.230952-67.2|
Initialize magnitude and angle of bus voltage
|V|=1.0S, 6, = 0.0 rad
=0.0 rad
1V0= 1.04, 6 = 0,0 rad
In the matrix form

Scheduled powers are

400
At bus.2, P2,sch Poz - Po =0 100
-=-4p.,
250
l,sch lo2-o =0 100
-=-2.5 p.u.
200
At bus 3, Pssch Pos - Pos = 100.-0= 2p.u.
 Load Flow Analysis 107
The real power at buses 2 and 3 and reactive power at bus 2 are
P,=|V|Vi|lYil cos(0,1- 6,+ 8) + |V||Yl cos 8,,
+|VI|VI|Y,;| cos (e, - 6, + 6,)
P,= (1) (1.05) (22.36068) cos(116.6 00 + 0) +(IF(s8.13777) cos(-634)
+ (1) (1.04) (35.77709) cos(116.6 - 0+ 0) = -1.1414
P= |VIVl1| cos(0,1-6+ 8) + |VIVAIYl cos(O,, -8, +&,)
+ |V||l cos ,3
P,- (1.04) (1.05) (3 1.62278) cos(108.4 -0+ 0)
+(1.04)(1) (35.77709) cos (1 16.6 - 0+0) +(1.04°(67.23095)cos(-672)
= 0.5616

e,=-VVIY,| sin (Ø,1 - 6,+ 6) - |ViIIY sin 6,

- |V,||V||Ya sin (6,3 -& + 6)
Q,=-(1) (1.05) (22.36068) sin (116.6 0 + 0) - (1 (58.13777) sin(-63.4
- (1)(1.04) (35.77709) sin (116.6 0 + 0) =-2.28
-; ros' Difference înscheduled tó calcülated power
AP'Psch Pea =-4I.141)=-2.8586:
AP = Bsch -Pcals =2-(0.5616) =1.43846
AO= 0.s6h -ele =-2.5 -(-2.28) =-0.22
The Jacobian matrix is given by

[AP

=|V|V,al sin (e;i-8, +8) +|V||V,l| sin(@, - 6, +6)

= (1)(1.05) (22.36068) sin(i16.6 -0+ 0)
+ (1)(1.04) (35.77709) sin (116.6 0 + 0) = 54.2634
VKYalsin (8,-6, +8)
=-(1)(1.04) (35.77709) sin(116.6 0 + 0) =-33.2698
aiVY,| cos (®,1 -6,+ 6) +2|V|l cos @,:
+ |V,||Y3l cos(, - , + 86)
108
Electrical Power Systems: Analysis,
= (1.05) Security and Deregulation
+ (1.04)
(22.36068) cos(116.6 0 + 0) + 2(1)
JP, (35.77709) cos(16.6 (58.13777) cos(-63.4)
0 + 0) = 24.890
=-V,Y sin(0, 6, +8,)
=-(1.04) (1) (35.77709) sin(1 16.6 -0 + 0) =
= -33.2698
|VI|V|Ylsin(®,I - 6, + 8,) +|V,||VYl sin(,; - &, + 6,)
-(1.04) (1.05) (31.62278) sin (108.4 -0 + 0)
+ (1.04) (1) (35.77709) sin (116.6 -0 + 0) =
66.0365
= |V||Y;2l cos(®,- & + &,)
= (1.04)
(35.77709) cos(116.6 0 + 0) =
-16.663
=|V||V||Y2l cos (, + 6) + |V|,||Y3| cos(e,3 - t 8)
= (1) (1.05) (22.36068) cos
(116.6 -0 + 0)
+ (1) (1.04) (35.77709) cos (1 16.6 -0 +
0) =-27.1731
=-VlVYal cos(O,; +&)
*-(1) (1.04) (35.77709) cos (116.6 0+ 0) 16.663
=-|V||,il sin (,1 - , + 6)- 2|V||Y sin (e,2)
- |KI|Ys sin (,3 - &, + 6,)
=-(1.05) (22.36068) sin (116.6 0 + 0) - 2(1) (58.13777) sin (-63.4)
- (1.04) (35.77709) sin (116.6 0 + 0) = 49.707
-2.85867 54.2634 -33.2698 24.890 480)
1.43846 = -33.2698 66.0365 -16.663| A6
-0.22 -27.1731 16.663 49.707 JAV
54.2634 -33.2698 24.890 -2,8586]
ASO) -33.2698 66.0365 -16.663 1.43846
LA|V,(0| -27.1731 16.663 49.707 -0.22

f0.0231 0.0134 -0.0071][-2.8586]

0.0137 0.0219 0.0005 1.43846
0.0081 0.0000 0.0161 -0.22:

A60) -0.04521
-0.0077
-0.0266
 Load Flow Analysis 109
New bus voltages and angles in the first iteration are

sl = 6 + A8 =0+ (-0.0452) =-0.0452

sl= 64 A6 =0 +(-0.0077) =-0.0077

V= o+ Av=1+-0.0266) =0.9734
(b) Verify the result using Power World Simulator (PWS): The one line
diagram of asimple bus system is drawn in PWS, which is shown in Figure 2.15.
194 MW 2
400 MW
Slack
250 MVAR

150 MVAR 1.05 p.u. 0.97 p.u.

0.00 deg -2.37 deg
1.04 p.u.
i0.28 deg

200 MW
144 MVAR

Figure 2.15 One line diagram of a simple three-bus system.

The first step is the formation of [Y,using the inspection method. The
calculated [Ypu values are given in Figure 2.16. Since the given problem is
a three-bus system, the size of [Yis 3 x 3 matrix.
X YBus (eus Admttance Mabic)

AI Recors Geo Set Counrs RS f)

D
Filter, Advanced Bus -Fnd.. Remove
Bus 2 Bus 3
Numbe Name Bus 1

11 20.00- j50.00 -10.00+ j20.00 -10.00 + j30.00

-10.00 + j20.00 26.00 j52.00 -16.00 + j32.00
2 2)2
3 33 -10.00 + j30.00 -16.00 + 32.00 26.00 - j62.00
Figure 2.16 Ypus result.

Newton-Raphson method
This method is cxecuted by pressing the icon Newton-Raphson power flow
iterations
available in tools ribbon, Before executing this method, the number of
and power
is to be fixed as 1in simulator options ribbon. The Jacobian values
the lst iteration.
flow results are given in Figure 2.17 and Figure 2.18 for
 Electrical Power Systems: Analysis, Security and Deregulation

Fher adavd RE
- Frd... Peove

Number Name Jacobian Equation Angle Angle Volt Mag Volt Mad
Bus 2. Bus 3Bus 2 Bus 3

1|212 Real Power S1.71 31.76 21.24 -16.80

313 Real Power -33.03 65.71 -15.32 28.99
32/2 Reactive Power -28.66 17.47 48.23 -30.53
33 Voltage Maghtde 1.00

Figure 2.17 Jacobian values.

138.0 M HVa 1.05o 0,00

GSNEPATCK 1 213.54 136.29R 253.3
TO 22 177.35 113.36 210.5
3 3 36.18 22.94 42.8
Nistoatch * 213.54 136.29
2 2 136.0 Mar 0.9741 -2.70
LOAD 1 400.00 250.0 471.7
TO -169.32-97.26 295.3
TO -229.08 -139.14 266.0
Nsatct -1.60 -13-5E
sUS 139.0. Myat 1.0400-0.45 11
GENERRTOR 200.00.. 135.63R 241.6
1 -36.01 -22.4 42.4
1. 23¬.54 156.07 286.2
Ki sta tch 199.47 .135.52

Figure 2.18 Power flow results and voltages-1st iteration.

The details of convergence are shown in Figure 2.19. The mismatches of
l powers and reactive power for each iteration are also clearly indicated.
is method takes 2 iterations to converge power flows. The converged values
Jacobian and Power flow results are given in Figure 2.20 and Figure 2.21.
OMessage Log: booksanplePWB
Startng Souton usrng Rectangdar NewtonRachson

Number: OMax P: 285.000 st bus 2Max Q: 22.000 at bus

NTber: 1Max P: 2530 at bus 3 Max Q: 13.578 at bus 2
Numte: 2Max P: 0.004 at bus 2 Max Q: 0.001 at bus
Fncedvoltage cocwdkop.teabun: 1

SoLaon Frished in 0.000 Seconds

Snlabon: Successfu Power Fow Solton

Figure 2.19 Details of convergence.

Frd. Raeo

Number Name Jacobian Equation Angde Ange Volt Mag volt Meg
Bus 2 Bus 3BUs 2Bus 3

Rea Power 51.60 -31.69 21.15 -16.73

Red Power -32.93 65.60-15.35 28.96
Reactve Power -28.55 17.40 47.95 -30.47
Voltage Magniud 1.00

Figure 2.20 Jacobian values.

 Load Flow Analysis 111

to

r
frpee
Ord

MVA 1.0500
139,0 KA MYar
140.853 259.9
218.42
179.36, 118.73 215.1
22. 12 44.9
1 39.06
33 ¿4 215.42 140.85 o.9717 -2.70 11
138.0 M var MVA
2 2 400.00 250.00 471.7
1 -170. 97 -101.35 199.1
1 1 1 -229.03 -142.05 272.7
TO 3 3
1.0400 -0.59 1 1
133.0 Myar
33 206.09 145.18R 247.7
GENERATOR 1 -38.87 -21.57 44.5
1
1 1 167.75 291.9
238.89
22 200.00 146.13
4MI smach

Figure 2.21 Converged power flow results and voltages.

PSS/E
software and it is iv
The same problem is taken and drawn in PSS/E
Figure 2.22.
1
Slack bus 2
Load bus
179.4 -171:0|0.972
-101.9400.0
218,4|118.7
140.9R 39.1 -229.1 230.0
22.1 -148.1
1.050 -38,9
-21.6
238.9
T167.7

146.2R
200.0 1.040
Generation bus

Figure 2.22 One line diagram ofa simple three-bus system.

Once the data are entered in the software it can be executed by the above
three power flow methods. Figure 2.23 shows the converged results obtained
by the Gauss-Seidel method. This window is generated from bus based repont.
The Newton-Raphson method is executed and the power flow resuls at
shown in Figure 2.23.

2.7 Fast Decoupled. Load Flow Method

The Fast Decoupled Load Flow (FDLF) method is one ofthe improved methads
which was based on the methodaad
simplification
reported by Stott and Alsac
of the Newton-Raphson m
simpliñcatikns
in 1974. This method due to its
112 Electrical Power Systems: Analysis,
1
Security and Deregulation
X---ZONE--X
2 3
FOR
NON-TRANSFORMER
BRANCHES
X--ZONE---X
X---ZONE---X

1 1
1
X---AREA--X
TRANSFORMERS X--AREA-.-X X---AREA----X

1 1 1 1
1 1
X---LOSSES---XX---LOSSES---X method.
%MVA
FOR MVAR
16.780.55 MVAR16.78 X---LOSSES---X
19.69 MVAR0.55 19.69
Newton-Raphson
%1 MW8.390.18
Report
MW 8.399.85 MW
0.189.85
RAT0NG
11:50 SET
A 0.00 0.9717PU-2.70 -0.50
MVA
%
1.050PU kV." MVA
1.0400PU.
% -kV. using
2011 kV
results
27
JUN 260
259.9 % 247.7
248 Converged
Mon, 44.9 MVA 471.7
215.1 199.0
272.7 44.5291.9

MVAR'
140.8R MW,
MVAR250.0 -148.0"MVAR
-229.0 146.2R 2.23
39.1
22.1
118.7 -171.0
-101.9 167.7
-21.6
SYSTEM
SIMULATOR--PSS®E Figure
POWER
179.3
MW218.4 400.0 -38.9
MW200.0238.9

INTERACTIVE
PTI
CKT CKT CKT
1 1 1

GENERATION
FROM FROM
GENERATION
SLACK
1BUS TO LOAD
2BUS 1
SLACK
BUS SLACK
BUSLOAD
2BUS
2LOAD
BUSGEN
3BUS BUS 3
3GEN GEN
BUS
TO
LOAD-PQ
1

BUS TO BUS TOTO BUS TO TO

 Load Flow Analysis 113

of calculations, fast convergence and reliable

results became the most widely
used method in load flow analysis. heavy loading
However, FDLF for some cases, where high RIX ratios or
conyerge well. For these
(low voltage) at some buses are present, does not
made to overcome these
cases, many efforts and developments have been convergence of systems
convergence obstacles. Some of them targeted the
However, one of the
with high RIX ratios, others those with low voltage buses.
Load Flow developed
most recent developments is a Robust Fast Decoupled
by Wang and Li; it is based on heuristic justification and general voltage
voltage
normalization methods and solves both high RIX ratios and low bus
problem simultaneously.
real power
This method exploits the property of the power system whereinflow-voltage
Alow-voltage angle (P = (VV,X;) sín 8) and reactive power
6- (V}IX) coupled.
magnitude are loosely (0- (V,VJX) cosNewton-Raphson method, we will
As the FDLF is derived from the some
and apply
start from the matrix representation of Newton-Raphson
FDLF.
simplifications and approximations to reach the equations of the
The matrix representation of the Newton-Raphson method is:
AP]
(2.51) a :

Elements of Jacobian matrix J,

(i) the diagonal elements are
dSyVX| sin (9, -6, +8;) (2.58)
j*i
(ii) the off-diagonal elements are
j #i (2.59)
=V,| sin(0., - 6, +8,)
Elements of Jacobian matrix J,
() the diagonal elements are
(2.60)
-vl cos®,+,lG|cos(8, -8,+8,)
j*i

(ii) the off-diagonal elements are

=Vlcos(0,- 8, +8,) (2.61)

Elements of Jacobian matrix J,

()) the diagonal elements are
(2.62)
-XVll; ,|cos(0, - 8, +8,)
j÷i
 114 Electrical Power Systems: Analysis, Security and Deregulation

(ii) the off-diagonal elements are

o=lV, cos(0, - 8, +8) j*i (2.63)

Elements of Jacobian matrix J,
() the diagonal elements are

=-2Vl sin 8,-V,l,sin(@, -8, +8,)

ji
(2.64)

(ii) the off-diagonal elements are

=V sin (e, - , +6,) j*i (2.65)

Now, for typical power system branches:

XIR > l and 0,, < 20° (2.66)
These two approximations will cause a weak coupling between AP and
AV, and between A0 and A, hence J, and J, entries of the initial matrix of
equation (2.57) can be ignored leading to the following decoupled equations:
AP J, 0. A.
(2.67).
aP1
[AP) -WJ[A]-A8] (2.68)

[A0] =VJIAV]=| (2.69)

Equations (2.68) and (2.69) show that the matrix equations are separated
into two decoupledequations requiring considerably less time to solve compared
to the time required for the solution of Eq. (2.57).
Furthernore, considerable simplifications can be made to eliminate the
need for recalculating J, and J, during iteration.
The elements of Jacobian matrix J, are as follows.
The diagonal elements are

Sr,, sin(@, -6, +8)-1v1% |sin(@,)

j=l

d5 =-0,-|VIY sin (0,)

=-0,- 1VB
 Load Flow Analysis 115
as
Now, the diagonal elements of J, can be written
dH_-0,-V B (2.70)

where B, = || sin , is the imaginary part of the diagonal elements of the

bus admittance matrix Ypus:
Further simplifications can be applied to Eq. (2.70), by considering
B, >2, and |V| |V|
JP (2.71)

Also, as under normal operating conditions , - is quite small, therefore

e, - 8 + 8 = , and |V)
The off-diagonal elements of J, can be written as

=-|VV)sin(0, - 6, + 8)
=-( sin(0,)
.

t(2.72),
Similárlý, the diagonal êlementsof J, may be written as

-Vl,|sin@,-,lG|
j=l
sin (e, - 6,+8,)
Multiplying the above equation by V, we get

;=VY| sinô, j=l

Vl,IX,| sin (9, -6, +8) =-V}B, +Q
Again, since B,> , 2, may be neglected
=B (2.73)
The off-diagonal elements of J, are

;=VllY,sin(0, -8, +&,)

Again assume 0,, - + &, =

=-VAY)sin 0,
(2.74)
 116 Electrical Power Systems: Analysis,
Applying these
Security and Deregulation
assumptions to Eqs. (2.68) and
(2.69), we get
J8.-)B,
--VB A8)B,
AP =-B,A;
AP
=-B'A,
Similarly, (2.75)
--7B Or
=-|V| B
=-B,A|]
=-B'A|VA (2.76)
where, B and B" are the imaginary part of the bus admittance
such that B' contains all buses admittance except those related to thematrix Yue
slack bus,
and B'" is Bdeprived from all ivoltage controlled buses related admittances.
Finally, all these approximations and. simplifications, lead to the folowing.
successive voltage magnitude and voltage angle updating equations.
A6=-[BT AP
(2.77)

AV=-[BT (2.78)
FDLF technique is very useful in contingency analysis where numerous
outages are to be simulated or a load flow solution is required for online control.
The algorithm written according to the equations derived in the previous
section is as follows:
Step 1: Create the bus admittance matrix [YpusJ
setting all bus volt
Step 2: Detect all kinds and numbers of buses -and- and the
voltage angles to 0,
ages to an initial value of 1 p.u., all
iteration counter iter to.0.
matrics B and B" according to Eqs. (2:75) and (2.76).
Step,3: Create the
Step 4: If max (AP, A0) S accuracy
AP = sch -P
A0, =0h-O.
then go to Step 6
 Load Flow Analysis 117

else
(i) Calculate J, and J, elements of Eqs. (2.71), (2.72), (2.73)
and (2.74).

(iü) Calculate the real and reactive powers at each bus, and chcck
if MVAR of generator buses are within the limits, otherwise
update the voltage magnitude at these buses by ±2 %.
If imin <,<my, calculate pl
If Qimase Drsch max
If < ,min Dsch 2min
The PV bus will act as PO bus.
(iüi) Calculate the power residuals, AP and A0.

(iv) Calculate the bus voitage and voltage angle updates AVand A&

[A8, 8T'

(v) Update the voltage magnitude Vand the voltage angle &at
each bus.
s{k*l = +sol

(vi) Increment of the iteration counter iter = iter + I

Step 5: If iter S maximum number of iteration

then go to Step 4
. else print out Solution did not converge and go to Step 6.
display of
Step 6: Print out of the power tlow solution, computation and
the line ilow and losses.
Figure 2.24 all the
This conpletes the load flow study. Finally, in chart.
computational steps are summarized in the detailed tlow
Electrical Power Systems: Analysis, Security and Deregulation

Start

Read system data

Fom Y, bus find B' and B"

Initialize magnitude and phase angle of bus voltages

Set iteration count k= 0

Buscount i 1

Calculatc, P
No
Is refer PV bus
Yes
Calculate, AP" Pschpl4
sch
Calculate, AP=P,-P" =Osch-"

Check Q limit
Oisch imin
QschOmax se"-.uch-e =Qisch'
AO"=O.-o

Yes Is iSn
No
Calculate

Yes.Calculate power
flow and iosses
Check|Ae", JAe"se
End
No
k+ 1
Set k=
for FDLF method.
Flow chart
Figure 2.24
 Load Flow Analysis 119
EXAMPLE 2.7 Figure 2.25 shows the one line diagram of a simple three
bus system with generation at bus 1. The magnitude of voltage at bus I is
adjusted to 1.05 p.u. The scheduled loads at buses 2 and 3 are given in the
diagram. Line impedances are marked as n p.u. on a 100 MVA base and the
line charging susceptances are neglected.
(a) Using the fast decoupled load flow method, determine the phasor
values of the voltages at the load buses 2 and 3(PO bus) accurate to
decimal places.
(b) Verify the result with Power World Simulator and PSS/E.
1
0.02 + j0.04 2

400 MW

Slack bus 0.01 +j0.03 0.0125 +j0.025 +250 MVAR

V=1.05
3

200 MW
Figure 2,25., One line diagram of a simple
three-bus system.
Solution: (a) Form the YpS
Y12=
Z12 0.02 + j0.04 =10- j20
1 1

Z13 0.01+ j0.03 =10- j30

1 1
Z23 0.0125 + j0.025 =16- j32
Y1 =2 t y3 = (10-j20) + (10-j30) =
20-j$0
Y12=Y1 =2 =-(10-j20) =-10 +j20
Y13 = Y;1 =-y)3 -(10 -j30) =-10 + 30
Y2 21 t '3 = (l0 -j20) + (16-j32) = 26
jS2
Y3 Y2=-3= -(16 -j32) = -16 + j32
32 = (10 -j30) + (16-j32) = 26 - j62
20- j50 -10+ j20 -10+ j30|
.
Ypus =-10+ j20 26- j52
-16+ j32
-10+ j30 -16+ j32 26- j62
Bus 1 is slack bus and the
of phase angle A®, and A6, corresponding
is
bus susceptance matrix for
evaluation
 120 Electrical Power Systems: Analysis, Security and Deregulation
-52 32
32 -62
The inverse of the above matrix is

(8T'=| -0.028182 -0.014545]

-0.014545 -0.023636|
Initialize magnitude and angle of bus voltage
|V|=1.0S, S, = 0.0 rad
|V,0= 1, -S = 0.0 rad
|V,0=1.04, S = 0.0 rad
In the matrix form

Scheduled powers are

400
at bus 2. Psch =Poz - Po =0 100 -4p.u.
250
100
-2.5 p.u.
200
at bus 3, P3sch Pos - Pos = 100 -0=2p.u.
The real power at buses 2 and 3 and reactive power at bus 2 are

P,- |V|VIlz1l cos(e,,-,+ 6) + | | | l cos e,2

+ 1V,I|VI|Y cos(e-6+ &,)
P,=(1) (1.05) (22.36068) cos (116.6 - 0 + 0) + (1' (58.13777) cos (-63.4)
+ ()(1.04) (35.77709) cos (1l16.6 - 0 + 0) = -1.1414
P,= |V;||I|Y cos(0,, - & 6)VI|V|Y cos(e,2- 8, + &)
+|||Y cos o,,
"P= (1.04) (1.05) (31.62278) cos (108.4-0 + 0)
+(1.04) (1) (35.77709) cos(116.6 -0+0) +(1.04°(67.23095) cos (-67.2)
= 0.5616

,--V|VIY,| sin (e1 -, +6,) - 1V|| sin e,

- |VVI sin(,, - +6)
--(1) (1.05) (22.36068) sin(116.6 0 + 0) - (1(58.13777) sin(-63.4)
- (1) (1.04) (35.77709) sin (1l6.6 0 + 0) =-2.28
Difference in scheduled to calculated power
AO= Ps - =-4-(-1.1414) =- 2.8586
 Load Flow Analysis 121

AP = Psch- P =2-(0.5616) =1.43846

AO= .sch -le-2.5 - (-2.28) =-0.22
The FDLF algorithm given by Eq. (2.77) becomes
[A6,1® = BT'

=B'T'

-2.8586
-0.028182 -0.014545 1.0
-0.014545 -o.023636 1.43846
1.04
-0.028182 -0.014545|-2.8586|
*00145450.023636| T:3831
-0.060483
-0.008909
Since bus 3 is a regulated bus, the corresponding row and column of B' are
eliminated and we get
B"= (-52]
-1
[B"T=-0.01923
52

(AV,j0= B"T'

[AV,0--0.01923)--00042308.
1.0
The new bus voltages and the angles in the frst iteration are

S = 8+ A6 =0+(-0.060483) =-0.060483
=8 t A =0+(-0.008909) = -0.008909