CHAPTER 3

FOUR AREA LOAD FREQUENCY CONTROL

3.1 Block Diagram of Uncontrolled Four-Area LFC

The block diagram is given in Figure 29.

x16 -a34 x16

-1

2piT 34 -a34
.05
1/3 POWER SYSTEM 3
1/R3 x16 POWER SYSTEM 4 1/3
1/R4
(9) (8) 1 (11) (12 )
(16 ) Integrator 3
X9 s
1 1 X8 (7) DF3 (10 )
DF4 X11 1 1
Tsg3.s+1 Tt 3.s+1 Kps3 X7 X10 Kps4 Tt 4.s+1 Tsg4.s+1
GOVERNOR 3 TURBINE 3 Tps 3.s+1 Tps4.s+1 TURBINE 4 GOVERNOR 4
GENERATOR 3 GENERATOR 4 W4

W3

x15

x17

(18 ) 2piT 42
2piT 31 (14 )
x14 1 x18 x18
x14 1 .05
.05 s
s
Integrator 2 Integrator 7
-1
-a42
-1
-a31 (17 ) 2piT 32
2piT 41 (15 )
1 x17 -1 -a42 x18
-1 x15 1 -a32
-a31 x14 -a41 .05
.05 s
s
Int . Integrator 9
-a41 x15 -a32 x17
x13 -a12 x13
-1

2piT 12 -a12
.05 POWER SYSTEM 2
POWER SYSTEM 1
X13

(13 ) 1
Integrator 2
s
(1) (4)
DF1 DF2
(3) (2) Kps1 x1 x4 Kps2 (5) (6)
1 x3 1 x2 Tps 1.s+1 Tps2.s+1 X5 1 X6 1
Tsg1.s+1 Tt 1.s+1 GENERATOR 1 GENERATOR 2 Tt 2.s+1 Tsg2.s+1
GOVERNOR 1 TURBINE 1 TURBINE 2 GOVERNOR 2
W2
1/R1 1/R2
1/3 1/3
W1

Fig . Block diagram of Four Area load frequency control without controller

Figure 29 Block diagram of uncontrolled four area load frequency control

3.1.1 Mathematical Model Derivation

From Figure 29 we develop the mathematical model of four area lfc as under

x1  f1
x4  f 2
x7  f 3

x10  f 4

x2  PG1

x5  PG 2

x8  PG 3

x11  PG 4

w1  PD1
w2  PD 2
w3  PD 3

w4  PD 4

For block 1
x1 Kps1

x2  W1  a31 x14  a41 x15  x13 1  Tps1s
 1 K K a Kps1 a Kps1 1
x1   x1  PS 1 x2  PS 1 x13  31 x14  41 x15  W1
TPS 1 TPS 1 TPS1 Tps1 Tps1 TPS 1
For block 2
x2 1

x3 1  Tt1s
 1 1
x2   x2  x3
Tt1 Tt1
For block 3
 x3 1

x1 Tsg1s  1
x13  a41 x15  a31 x14 
R1
 1 1 1 a a
x3   x1  x3  x13  31 x14  41 x15
R1TSg1 TSg1 TSg1 Tsg1 Tsg1

For block 4
x4 Kps2

x5  W2  a42 x18  a32 x17  a12 x13 Tps2  1
 1 K PS 2 a K a Kps2 a Kps2 K
x4   x4  x5  12 PS 2 x13  32 x17  42 x18  PS 2 w2
TPS 2 TPS 2 TPS 2 Tps2 Tps2 TPS 2
For block 5
x5 1

x6 Tt2 s  1
 1 1
x5   x5  x6
Tt 2 Tt1
For block 6
x6 1

x4 Tsg 2 s  1
a12 x13  a32 x17  a42 x18 
R2
 1 1 a a a
x6   x4  x6  12 x13  32 x17  42 x18
R2TSg 2 TSg 2 TSg 2 Tsg 2 Tsg 2

For block 7
x7 Kps3

x8  W3  x16  x17  x14 Tps3 s  1
. 1 Kps3 Kps3 Kps3 Kps3 Kps3
x7   x7  x8  x14  x16  x17  W3
Tps3 Tps3 Tps3 Tps3 Tps3 Tps3
For block 8
x8 1

x9 Tt3 s  1
 1 1
x8   x8  x9
Tt 3 Tt 3
For block 9
x9 1

x7 Tsg3 s  1
  x16  x14  x17
R3
 1 1 1 1 1
x9   x7  x9  x14  x16  x17
R3TSg 3 TSg 3 TSg 3 Tsg 3 Tsg 3

For block 10
x10 Kps4

x11  W4  a34 x16  x18  x15 Tps4 s  1
. 1 Kps4 Kps4 a Kps4 Kps4 Kps4
x10   x10  x11  x15  34 x16  x18  W4
Tps4 Tps4 Tps4 Tps4 Tps4 Tps4
For block 11
x11 1

x12 Tt4 s  1
 1 1
x11   x11  x12
Tt 4 Tt 4
For block 12
x12 1

x10 Tsg 4 s  1
  a34 x16  x18  x15
R4
 1 1 1 a 1
x12   x10  x12  x15  34 x16  x18
R4TSg 4 TSg 4 TSg 4 Tsg 4 Tsg 4
For block 13
x13 2 T12

x1  x4 S

x13  2T12 x1  2T12 x4

For block 14
x14 2T31

x7  x1 S

x14  2 T31 x1  2T31 x7

For block 15
x15 2 T41

x10  x1 S

x15  2T41 x1  2T41 x10

For block 16
x16 2 T34

x7  x10 S

x16  2T34 x7  2T34 x10

For block 17
x17 2T32

x7  x4 S

x17  2T32 x4  2T32 x7

For block 18
x18 2T42

x10  x2 S

x18  2T42 x2  2 T42 x10
The derived eighteen equations can be organized in the following vector matrix
form

x  Ax  Fw
Where

x   x1 . . . x18  = state vector

T

w   w1 w4 
T
w2 w3 = disturbance vector
. . .
Using equations x1 , x2 ,... x18 matrix A is formulated as
 -1/Tps1 Kps1/Tps1 0 0 0 0 0 0 0 0 0 0 -Kps1/Tps1 (a31*Kps1)/Tps1 (a41*Kps1)/Tps1 0 0 0 
0 -1/Tt1 1/Tt1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
 
-1/(R1*Tsg1) 0 -1/Tsg1 0 0 0 0 0 0 0 0 0 1/Tsg1 a31/Tsg1 a41/Tsg1 0 0 0 
 
0 0 0 -1/Tps2 Kps2/Tps2 0 0 0 0 0 0 0 (a12*Kps2)/Tps2 0 0 0 0 0 
0 0 0 0 -1/Tt2 1/Tt2 0 0 0 0 0 0 0 0 0 0 0 0 
 
0 0 0 -1/(R2*Tsg2) 0 -1/Tsg2 0 0 0 0 0 0 -a12/Tsg2 0 0 0 a32/Tsg2 a42/Tsg2 
0 0 0 0 0 0 -1/Tps3 Kps3/Tps3 0 0 0 0 0 -Kps3/Tps3 0 -Kps3/Tps3 -Kps3/Tps3 0 
 
0 0 0 0 0 0 0 -1/Tt3 1/Tt3 0 0 0 0 0 0 0 0 0 
0 0 0 0 0 0 -1/(R3*Tsg3) 0 -1/Tsg3 0 0 0 0 1/Tsg3 0 1/Tsg3 1/Tsg3 0 
A=  
0 0 0 0 0 0 0 0 0 -1/Tps4 Kps4/Tps4 0 0 0 -Kps4/Tps4 (a34*Kps4)/Tps4 0 -Kps4/Tps4 
 
0 0 0 0 0 0 0 0 0 0 -1/Tt4 1/Tt4 0 0 0 0 0 0 
0 0 0 0 0 0 0 0 0 -1/(R4*Tsg4) 0 -1/Tsg4 0 0 1/Tsg4 a34/Tsg4 0 1/Tsg4 
 
 2*pi*T12 0 0 -2*pi*T12 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
-2*pi*T31 0 0 0 0 0 2*pi*T31 0 0 0 0 0 0 0 0 0 0 0 
 
-2*pi*T41 0 0 0 0 0 0 0 0 2*pi*T41 0 0 0 0 0 0 0 0 
0 0 0 0 0 0 2*pi*T34 0 0 -2*pi*T34 0 0 0 0 0 0 0 0 
 
0 0 0 -2*pi*T32 0 0 2*pi*T32 0 0 0 0 0 0 0 0 0 0 0 
 
0 0 0 -2*pi*T42 0 0 0 0 0 2*pi*T42 0 0 0 0 0 0 0 0 

Disturbance matrix F is given as

  K ps1 
 - 00 0 00 0 00 0 0 0000000 
 Tps1 
 K ps 2 
 0 00 - 00 0 00 0 00000000 
Tps 2
F = 
T 

 Kps3 
0 00 0 00 - 00000000000
 Tps3 
 
 Kps4
0 00 0 00 000- 00000000 
 Tps4 

Let F1, F2,F3 and F4 are system output variables,then output matrix will be the form of

Y=Cx +Du D=0 and C & x are given as

1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
C =  
0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 
 
0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 

xT =  x1 x2 x3 x4 x5 x6 x7 x8 x9 x10 x11 x12 x13 x14 x15 x16 x17 x18 

Let F=B and w=u, so our model turns to


x  Ax  Bu

3.1.2 Matlab Programme

A Matlab programme is written using A, B, C, W & D as given below.

Kps1=100; Kps3=100; Kps2=110; Kps4=110; Tsg1=.4; Tsg2=.4; Tsg3=.4;
Tsg4=.4; Tt1=.5; Tt2=.5; Tt3=.5; Tt4=.5; Tps1=20; Tps2=20;
Tps3=20; Tps4=20; b1=.425; b2=.425; b3=.425; b4=.425; R1=3; R2=3; R3=3;
R4=3; W1=.02; W2=.04; W3=.05; W4=.06;
T12=.00795; T31=.00795; T41=.00795; T34=.00795; T32=.00795; T42=.00795;
a12=1; a31=1; a41=1; a34=1; a32=1; a42=1;
A=[-1/Tps1 Kps1/Tps1 0 0 0 0 0 0 0 0 0 0 -Kps1/Tps1 (a31*Kps1)/Tps1
(a41*Kps1)/Tps1 0 0 0;
0 -1/Tt1 1/Tt1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0;
-1/(R1*Tsg1) 0 -1/Tsg1 0 0 0 0 0 0 0 0 0 1/Tsg1 a31/Tsg1 a41/Tsg1 0 0 0;
0 0 0 -1/Tps2 Kps2/Tps2 0 0 0 0 0 0 0 (a12*Kps2)/Tps2 0 0 0 0 0;
0 0 0 0 -1/Tt2 1/Tt2 0 0 0 0 0 0 0 0 0 0 0 0;
0 0 0 -1/(R2*Tsg2) 0 -1/Tsg2 0 0 0 0 0 0 -a12/Tsg2 0 0 0 a32/Tsg2
a42/Tsg2;
0 0 0 0 0 0 -1/Tps3 Kps3/Tps3 0 0 0 0 0 -Kps3/Tps3 0 -Kps3/Tps3
-Kps3/Tps3 0;
0 0 0 0 0 0 0 -1/Tt3 1/Tt3 0 0 0 0 0 0 0 0 0;
0 0 0 0 0 0 -1/(R3*Tsg3) 0 -1/Tsg3 0 0 0 0 1/Tsg3 0 1/Tsg3 1/Tsg3 0;
0 0 0 0 0 0 0 0 0 -1/Tps4 Kps4/Tps4 0 0 0 -Kps4/Tps4 (a34*Kps4)/Tps4 0
-Kps4/Tps4;
0 0 0 0 0 0 0 0 0 0 -1/Tt4 1/Tt4 0 0 0 0 0 0;
0 0 0 0 0 0 0 0 0 -1/(R4*Tsg4) 0 -1/Tsg4 0 0 1/Tsg4 a34/Tsg4 0 1/Tsg4;
2*pi*T12 0 0 -2*pi*T12 0 0 0 0 0 0 0 0 0 0 0 0 0 0;
-2*pi*T31 0 0 0 0 0 2*pi*T31 0 0 0 0 0 0 0 0 0 0 0;
-2*pi*T41 0 0 0 0 0 0 0 0 2*pi*T41 0 0 0 0 0 0 0 0;
0 0 0 0 0 0 2*pi*T34 0 0 -2*pi*T34 0 0 0 0 0 0 0 0;
0 0 0 -2*pi*T32 0 0 2*pi*T32 0 0 0 0 0 0 0 0 0 0 0;
0 0 0 -2*pi*T42 0 0 0 0 0 2*pi*T42 0 0 0 0 0 0 0 0];
B=[-Kps1/Tps1 0 0 0; 0 0 0 0; 0 0 0 0; 0 -Kps2/Tps2 0 0; 0 0 0 0; 0 0 0
0; 0 0 -Kps3/Tps3 0; 0 0 0 0; 0 0 0 0; 0 0 0 -Kps4/Tps4; 0 0 0 0;
0 0 0 0; 0 0 0 0; 0 0 0 0; 0 0 0 0; 0 0 0 0; 0 0 0 0; 0 0 0 0];
t=0:0.2:25;
U=[W1; W2; W3; W4];
BU=B*U;
D=0;
C=[1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0; 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0
0 0;
0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0; 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0
0];
[y,x]=step(A,BU,C,D,1,t);
plot(t,y),grid,
title('Four Area ws LFC W/O Controller For Change in Demands
W1,W2,W3&W4'),
xlabel('t,sec'),ylabel('f1pu,f2pu,f3pu,f4pu')

3.1.3 Matlab Programme Results

The load frequency response is shown in Figure 30.
 Four Area ws LFC W/O Controller For Change in demands W1,W2,W3&W4
0

-0.05

-0.1
f1pu,f2pu,f3pu,f4pu

-0.15

-0.2

-0.25

-0.3

-0.35
0 5 10 15 20 25
t,sec

Figure 29 Matlab response for four area l.f.control without controller

(Kps1=100; Kps3=100; Kps2=110; Kps4=110; Tsg1=.4; Tsg2=.4; Tsg3=.4;
Tsg4=.4; Tt1=.5; Tt2=.5; Tt3=.5; Tt4=.5; Tps1=20; Tps2=20;
Tps3=20; Tps4=20; b1=.425; b2=.425; b3=.425; b4=.425; R1=3; R2=3; R3=3;
R4=3; W1=.02; W2=.04; W3=.05; W4=.06;
T12=.00795; T31=.00795; T41=.00795; T34=.00795; T32=.00795; T42=.00795;
a12=1; a31=1; a41=1; a34=1; a32=1; a42=1)

3.1.4 Simulink Model for Uncontrolled Four-Area LFC

The simulink model is given in Figure 31.
 -a34
-a34 x16
-1

2piT 34
.05
1/3 POWER SYSTEM 3
1/R3
x16 POWER SYSTEM 4 1/3
1/R4
(9) (8) 1
(16 ) Integrator 3 (11) (12 )
X9 s
1 1 X8 (7) DF3 (10 ) 1 1
DF4 X11
.4s+1 .5s+1 100 X7 X10 110 .5s+1 .4s+1
GOVERNOR 3 TURBINE 3 20 s+1 20 s+1 TURBINE 4 GOVERNOR 4
GENERATOR 3 GENERATOR 4 W4

W3

2piT 31 (14 ) (18 ) 2piT 42

1 x18
x14 1 .05
.05 s
s
Integrator 2 Integrator 7
-1 -1
-a42
-a31
(17 ) 2piT 32
2piT 41 (15 )
1 x17 -1 -a42 x18
-1 x15 1 -a32
-a31 x14 -a41 .05 .05
s s
Int . Integrator 9 -a32 x17
-a41 x15
-a12 x13
-1

2piT 12 -a12
.05 POWER SYSTEM 2
POWER SYSTEM 1
X13
(13 ) 1
Integrator 2
s
(1) (4)
DF1 DF2
(3) (2) 100 x1 x4 110 (5) (6)
x3 x2 X6
1 1 20 s+1 20 s+1 X5 1 1
.4s+1 .5s+1 GENERATOR 1 GENERATOR 2 .5s+1 .4s+1
GOVERNOR 1 TURBINE 1 TURBINE 2 GOVERNOR 2
To Workspace Scope W2
1/R1 1/R2
1/3 1/3
simout
W1
Fig . Simulink model for 4 area load frequency control without controller

Figure 31

3.1.5 Simulink Model Results

A SIMULINK model is constructed as shown in Figure 31.The file is opened and is run
in the SIMULINK window. The simulation results are shown in Figure 32.

Four area sm l.f. control without controller

0

-0.05

-0.1
Df1,Df2,Df3,&Df4

-0.15

-0.2

-0.25

-0.3

-0.35
0 5 10 15 20 25
t(sec)

Figure 32

CONCLUSION-
From the responses of the Figures 30 and 32, we observe that the both the results are
same and steady state frequency deviation is -0.14pu, and settling time is approximately
20 seconds. Maximum overshoot is -0.27.

3.2 Block Diagram of Four-Area LFC with PI Controller

The block diagram is given in Figure 33.
 -a34 x16
x16
-1

2piT 34 -a34
.05
POWER SYSTEM 3
Freq . b3 POWER SYSTEM 4 1/3
1/3 x16 1/R4 Freq . b4
Sen .3 1/R3
Sen .4
(9) (8) 1 (11) (12 )
(21 ) (16 ) Integrator 3 Gain 4 (22 )
x21 X9 s
1 Ksg3 Kt3 X8 (7) Df3 (10 )
-K- Df4 X11 Kt4 Ksg4 x22 1
s Tsg3.s+1 Tt 3.s+1 -K-
Kps3 X7 X10 Kps4 Tt 4.s+1 Tsg4.s+1 s
Integrator 3 Gain 3 GOVERNOR 3 TURBINE 3 Int .
Tps 3.s+1 Tps 4.s+1 TURBINE 4 GOVERNOR 4
GENERATOR 3 GENERATOR 4 W4

W3
x15

x17

2piT 31 (14 ) (18 ) 2piT 42

1 x18 x18
x14 x14 1 .05
.05 s
s
Integrator 2 Integrator 7
-1
-1 -a42
-a31
(17 ) 2piT 32
2piT 41 (15 )
1 x17 -1 -a42 x18
-1 x15 1 -a32
-a31 x14 -a41 .05
.05 s
s
Int . Integrator 9
-a41 x15 -a32 x17
x13 -a12 x13
-1
2piT 12 -a12
.05 POWER SYSTEM 2
POWER SYSTEM 1
X13

(13 ) 1
Integrator 2
s
(1) (4)
Df1 Df 2
(19 ) (3) (2) Kps1 x1 x4 Kps2 (5) (6) Gain 2 (20 )
x19 Ksg1 x3 Kt1
1 x2 Tps1.s+1 Tps 2.s+1 X5 Kt2 X6 Ksg2 x20
-K- 1
s Tsg1.s+1 Tt 1.s+1 -K-
GENERATOR 1 GENERATOR 2 Tt 2.s+1 Tsg2.s+1 s
Integrator 1 Gain 1 GOVERNOR 1 TURBINE 1 TURBINE 2 GOVERNOR 2 Int .
1/R1 W2 Freq .
Frequency 1/3 1/R2
1/3 b2 Sen . 2
b1 Sensor 1
W1

Fig . Block diagram of Four Area load frequency control with PI controller

Figure33. Block diagram of four-area load frequency control

with PI controller
3.2.1 Mathematical Model Derivation

From Figure 33 we develop the mathematical model of controlled four area LFC as under

x1  f1
x4  f 2
x7  f 3

x10  f 4

x2  PG1

x5  PG 2

x8  PG 3

x11  PG 4

w1  PD1
 w2  PD 2
w3  PD 3

w4  PD 4

x19   ACE1dt

x20   ACE2 dt

x21   ACE3dt

x22   ACE4 dt

u1  ki x19

u2  ki x20

u3   ki x21

u4  ki x22

For block 1
x1 Kps1

x2  W1  a31 x14  a41 x15  x13 1  Tps1s
 1 K K a Kps1 a Kps1 1
x1   x1  PS 1 x2  PS 1 x13  31 x14  41 x15  W1
TPS 1 TPS 1 TPS1 Tps1 Tps1 TPS 1
For block 2
x2 1

x3 1  Tt1s
 1 1
x2   x2  x3
Tt1 Tt1

For block 3
x3 1

x1 Tsg1s  1
u1 
R1
  1 1 1
x3   x1  x3  u1
R1TSg1 TSg1 TSg1

 1 1 k
 x3   x1  x3  i x19
R1TSg1 TSg1 TSg1

For block 4
x4 Kps2

x5  W2  a42 x18  a32 x17  a12 x13 Tps2  1
 1 K PS 2 a K a Kps2 a Kps2 K
x4   x4  x5  12 PS 2 x13  32 x17  42 x18  PS 2 w2
TPS 2 TPS 2 TPS 2 Tps2 Tps2 TPS 2
For block 5
x5 1

x6 Tt2 s  1
 1 1
x5   x5  x6
Tt 2 Tt1
For block 6
x6 1

x Tsg 2 s  1
u2  4
R2
 1 1 1
x6   x4  x6  u2
R2TSg 2 TSg 2 TSg 2
 1 1 k
 x6   x4  x6  i x20
R2TSg 2 TSg 2 TSg 2

For block 7
x7 Kps3

x8  W3  x16  x17  x14 Tps3 s  1
 . 1 Kps3 Kps3 Kps3 Kps3 Kps3
x7   x7  x8  x14  x16  x17  W3
Tps3 Tps3 Tps3 Tps3 Tps3 Tps3
For block 8
x8 1

x9 Tt3 s  1
 1 1
x8   x8  x9
Tt 3 Tt 3
For block 9
x9 1

x7 Tsg3 s  1
u3 
R3
 1 1 1
x9   x7  x9  u3
R3TSg 3 TSg 3 TSg 3
 1 1 k
 x9   x7  x9  i x21
R3TSg 3 TSg 3 TSg 3

For block 10
x10 Kps4

x11  W4  a34 x16  x18  x15 Tps4 s  1
. 1 Kps4 Kps4 a Kps4 Kps4 Kps4
x10   x10  x11  x15  34 x16  x18  W4
Tps4 Tps4 Tps4 Tps4 Tps4 Tps4
For block 11
x11 1

x12 Tt4 s  1
 1 1
x11   x11  x12
Tt 4 Tt 4

For block 12
 x12 1

x
u4  10 Tsg 4 s  1
R4
 1 1 1
x12   x10  x12  u4
R4TSg 4 TSg 4 TSg 4
 1 1 k
 x12   x10  x12  i x22
R4TSg 4 TSg 4 TSg 4

For block 13
x13 2 T12

x1  x4 S

x13  2T12 x1  2T12 x4

For block 14
x14 2T31

x7  x1 S

x14  2 T31 x1  2T31 x7

For block 15
x15 2 T41

x10  x1 S

x15  2T41 x1  2T41 x10

For block 16
x16 2 T34

x7  x10 S

x16  2T34 x7  2T34 x10

For block 17
 x17 2T32

x7  x4 S

x17  2T32 x4  2T32 x7

For block 18
x18 2T42

x10  x2 S

x18  2T42 x2  2 T42 x10

For block 19
x19 1

x1b1  a31 x14  a41 x15  x13 s
.
x19  b1 x1  x13  a31 x14  a41 x15

For block 20
x20 1

x4b2  a42 x18  a32 x17  a12 x13 s
.
x20  b2 x4  a12 x13  a32 x17  a42 x18

For block 21
x21 1

b3 x7  x16  x14  x17 s
.
x20  b3 x7  x14  x16  x17

For block 22
x22 1

b4 x10  a34 x16  x18  x15 s
.
x22  b4 x10  x15  a34 x16  x18
The above twenty two equations can be organized in the following vector matrix
form

x  Ax  Fw
Where

x   x1 . . . x 22  = state vector
T

w   w1 w4 
T
w2 w3 = disturbance vector
. . .
Using equations x1 , x2 ,... x22 matrix A is formulated as
  -1/Tps1 Kps1/Tps1 0 0 0 0 0 0 0 0 0 0 -Kps1/Tps1 (a31*Kps1)/Tps1 (a41*Kps1)/Tps1 0 0 0 0 0 0 0 
 0 -1/Tt1 1/Tt1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
 
 -1/(R1*Tsg1) 0 -1/Tsg1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 -Ki/Tsg1 0 0 0 
 
 0 0 0 -1/Tps2 Kps2/Tps2 0 0 0 0 0 0 0 (a12*Kps2)/Tps2 0 0 0 (a32*Kps2)/Tps2 (a42*Kps2)/Tps2 0 0 0 0 
 0 0 0 0 -1/Tt2 1/Tt2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
 
 0 0 0 -1/(R2*Tsg2) 0 -1/Tsg2 0 0 0 0 0 0 0 0 0 0 0 0 0 -Ki/Tsg2 0 0 
 0 0 0 0 0 0 -1/Tps3 Kps3/Tps3 0 0 0 0 0 -Kps3/Tps3 0 -Kps3/Tps3 -Kps3/Tps3 0 0 0 0 0 
 
 0 0 0 0 0 0 0 -1/Tt3 1/Tt3 0 0 0 0 0 0 0 0 0 0 0 0 0 
 0 0 0 0 0 0 -1/(R3*Tsg3) 0 -1/Tsg3 0 0 0 0 0 0 0 0 0 0 0 -Ki/Tsg3 0 
 
 0 0 0 0 0 0 0 0 0 -1/Tps4 Kps4/Tps4 0 0 0 -Kps4/Tps4 (a34*Kps4)/Tps4 0 -Kps4/Tps4 0 0 0 0 
 
0 0 0 0 0 0 0 0 0 0 -1/Tt4 1/Tt4 0 0 0 0 0 0 0 0 0 0
A=  
 0 0 0 0 0 0 0 0 0 -1/(R4*Tsg4) 0 -1/Tsg4 0 0 0 0 0 0 0 0 0 -Ki/Tsg4 
 
 2*pi*T12 0 0 -2*pi*T12 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
 -2*pi*T31 0 0 0 0 0 2*pi*T31 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
 
 -2*pi*T41 0 0 0 0 0 0 0 0 2*pi*T41 0 0 0 0 0 0 0 0 0 0 0 0 
 0 0 0 0 0 0 2*pi*T34 0 0 -2*pi*T34 0 0 0 0 0 0 0 0 0 0 0 0 
 
 0 0 0 -2*pi*T32 0 0 2*pi*T32 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
 
 0 0 0 -2*pi*T42 0 0 0 0 0 2*pi*T42 0 0 0 0 0 0 0 0 0 0 0 0 
 b1 0 0 0 0 0 0 0 0 0 0 0 1 -a31 -a41 0 0 0 0 0 0 0 
 
 0 0 0 b2 0 0 0 0 0 0 0 0 -a12 0 0 0 -a32 -a42 0 0 0 0 
 0 0 0 0 0 0 b3 0 0 0 0 0 0 1 0 1 1 0 0 0 0 0 
 
 0 0 0 0 0 0 0 0 0 b4 0 0 0 0 1 -a34 0 1 0 0 0 0 
Let F1, F2,F3 and F4 are system output variables,then output matrix will be the form of

Y=Cx +Du D=0 and C & x are given as

 Kps1 
  Tps 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
 1

 Kps2 
 0 0 0  0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
Tps2
FT =  
 Kps3 
 0 0 0 0 0 0  0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
 Tps3 
 Kps4 
 0 0 0 0 0 0 0 0 0  0 0 0 0 0 0 0 0 0 0 0 0
 Tps4 

 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
C= 
 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
 
 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0

xT =[ x1 x2 x3 . . . x21 x22 ]

Let F=B and w=u, so our model turns to


x  Ax  Bu

3.2.2 Matlab Programme

A Matlab programme is written using A, B, C, W & D as given below.

Kps1=100; Kps3=100; Kps2=110; Kps4=110; Tsg1=.4; Tsg2=.4; Tsg3=.4;

Tsg4=.4; Tt1=.5; Tt2=.5; Tt3=.5; Tt4=.5; Tps1=20; Tps2=20;
Tps3=20; Tps4=20; b1=.425; b2=.425; b3=.425; b4=.425; R1=3; R2=3; R3=3;
R4=3; W1=.02; W2=.04; W3=.05; W4=.06; Ki=.09;
T12=.00795; T31=.00795; T41=.00795; T34=.00795; T32=.00795; T42=.00795;
a12=1; a31=1; a41=1; a34=1; a32=1; a42=1;
A=[-1/Tps1 Kps1/Tps1 0 0 0 0 0 0 0 0 0 0 -Kps1/Tps1 (a31*Kps1)/Tps1
(a41*Kps1)/Tps1 0 0 0 0 0 0 0;
0 -1/Tt1 1/Tt1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0;
-1/(R1*Tsg1) 0 -1/Tsg1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 -Ki/Tsg1 0 0 0;
0 0 0 -1/Tps2 Kps2/Tps2 0 0 0 0 0 0 0 (a12*Kps2)/Tps2 0 0 0
(a32*Kps2)/Tps2 (a42*Kps2)/Tps2 0 0 0 0;
0 0 0 0 -1/Tt2 1/Tt2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0;
0 0 0 -1/(R2*Tsg2) 0 -1/Tsg2 0 0 0 0 0 0 0 0 0 0 0 0 0 -Ki/Tsg2 0 0;
0 0 0 0 0 0 -1/Tps3 Kps3/Tps3 0 0 0 0 0 -Kps3/Tps3 0 -Kps3/Tps3
-Kps3/Tps3 0 0 0 0 0;
0 0 0 0 0 0 0 -1/Tt3 1/Tt3 0 0 0 0 0 0 0 0 0 0 0 0 0;
0 0 0 0 0 0 -1/(R3*Tsg3) 0 -1/Tsg3 0 0 0 0 0 0 0 0 0 0 0 -Ki/Tsg3 0;
0 0 0 0 0 0 0 0 0 -1/Tps4 Kps4/Tps4 0 0 0 -Kps4/Tps4 (a34*Kps4)/Tps4 0
-Kps4/Tps4 0 0 0 0;
0 0 0 0 0 0 0 0 0 0 -1/Tt4 1/Tt4 0 0 0 0 0 0 0 0 0 0;
0 0 0 0 0 0 0 0 0 -1/(R4*Tsg4) 0 -1/Tsg4 0 0 0 0 0 0 0 0 0 -Ki/Tsg4;
2*pi*T12 0 0 -2*pi*T12 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0;
-2*pi*T31 0 0 0 0 0 2*pi*T31 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0;
-2*pi*T41 0 0 0 0 0 0 0 0 2*pi*T41 0 0 0 0 0 0 0 0 0 0 0 0;
0 0 0 0 0 0 2*pi*T34 0 0 -2*pi*T34 0 0 0 0 0 0 0 0 0 0 0 0;
0 0 0 -2*pi*T32 0 0 2*pi*T32 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0;
0 0 0 -2*pi*T42 0 0 0 0 0 2*pi*T42 0 0 0 0 0 0 0 0 0 0 0 0;
b1 0 0 0 0 0 0 0 0 0 0 0 1 -a31 -a41 0 0 0 0 0 0 0;
0 0 0 b2 0 0 0 0 0 0 0 0 -a12 0 0 0 -a32 -a42 0 0 0 0;
0 0 0 0 0 0 b3 0 0 0 0 0 0 1 0 1 1 0 0 0 0 0;
0 0 0 0 0 0 0 0 0 b4 0 0 0 0 1 -a34 0 1 0 0 0 0];
B=[-Kps1/Tps1 0 0 0; 0 0 0 0; 0 0 0 0; 0 -Kps2/Tps2 0 0; 0 0 0 0; 0 0 0
0; 0 0 -Kps3/Tps3 0; 0 0 0 0; 0 0 0 0; 0 0 0 -Kps4/Tps4; 0 0 0 0;
0 0 0 0; 0 0 0 0; 0 0 0 0; 0 0 0 0; 0 0 0 0; 0 0 0 0; 0 0 0 0; 0 0 0 0;
0 0 0 0; 0 0 0 0; 0 0 0 0];
t=0:0.2:45;
U=[W1; W2; W3; W4];
BU=B*U;
D=0;
C=[1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0; 0 0 0 1 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0;
0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0; 0 0 0 0 0 0 0 0 0 1 0 0 0 0
0 0 0 0 0 0 0 0];
[y,x]=step(A,BU,C,D,1,t);
plot(t,y),grid,
title('Four Area ws LFC With PI Ctrlr For Change in Demands in
W1,W2,W3&W4'),
xlabel('t,sec'),ylabel('f1pu,f2pu,f3pu,f4pu')
3.2.3 Matlab Programme Results

The load frequency response of the above programme after running in work space of
matlab is shown in Figure 34.
 Four Area ws LFC With PI Ctrlr For Change in Demands in W1,W2,W3&W4
0

-0.05

-0.1
f1pu,f2pu,f3pu,f4pu

-0.15

-0.2

-0.25

-0.3

-0.35
0 5 10 15 20 25 30 35 40 45
t,sec

Figure 34 Matlab response for four area l.f.control with PI controller

(Kps1=100; Kps3=100; Kps2=110; Kps4=110; Tsg1=.4; Tsg2=.4; Tsg3=.4;
Tsg4=.4; Tt1=.5; Tt2=.5; Tt3=.5; Tt4=.5; Tps1=20; Tps2=20;
Tps3=20; Tps4=20; b1=.425; b2=.425; b3=.425; b4=.425; R1=3; R2=3; R3=3;
R4=3; W1=.02; W2=.04; W3=.05; W4=.06; Ki=.09;
T12=.00795; T31=.00795; T41=.00795; T34=.00795; T32=.00795; T42=.00795;
a12=1; a31=1; a41=1; a34=1; a32=1; a42=1)

3.2.4 Simulink Model for Four-Area LFC with PI Controller

The simulink model is given in Figure 35.

 -a34 x16
-1

Gain 8 -a34
.05
POWER SYSTEM 3
-K- x16 POWER SYSTEM 4
b3 1/3 1/3 -K-
1/R3 1/R4 b4
(9) (8) 1
(21 ) (16 ) Integrator 3 (11) (12 ) Gain 4 (22 )
x21 X9 s
1 1 1 X8 (7) Df3 (10 )
-K - Df4 X11 1 1 x22 1
s .4s+1 .5s+1 -K -
100 X7 X10 110 .5s+1 .4s+1 s
Integrator 3 Gain 3 GOVERNOR 3 TURBINE 3 Int .
20 s+1 20 s+1 TURBINE 4 GOVERNOR 4
GENERATOR 3 GENERATOR 4 W4

W3

Gain 10 (14 ) (18 )

1 x18
x14 1 .05
.05 s
s
Integrator 2 Integrator 7 Gain 17
-1
-1 -a42
-a31
(17 )
Gain 20 (15 )
1 x17 -1 -a42 x18
-1 x15 1 -a32
-a31 x14 -a41 .05
.05 s
s
Int . Integrator 9 Gain 21
-a41 x15 -a32 x17
-a12 x13
-1

Gain 6 -a12
.05 POWER SYSTEM 2
POWER SYSTEM 1
X13

(13 ) 1
Integrator 2
s
(1) (4)
Df1 Df 2
(19 ) (3) (2) 100 x1 x4 110 (5) (6) Gain 2 (20 )
x19 1 x3 1
1 x2 20 s+1 20 s+1 X5 1 X6 1 x20 1
-K- -K-
s .4s+1 .5s+1 s
GENERATOR 1 GENERATOR 2 .5s+1 .4s+1
Integrator 1 Gain 1 GOVERNOR 1 TURBINE 1 Int .
TURBINE 2 GOVERNOR 2
Scope
1/R1 To Workspace W2
1/3 1/R2 b2
b1 1/3 -K-
-K-
W1 simout

Fig . Simulink model for four area load freqouency control with PI controller

Figure 35

3.2.5 Simulink Model Results

A SIMULINK model is constructed as shown in Figure 35.The file is opened and is run
in the SIMULINK window.The simulation results are shown in Figure 36.

Four area sm load freq. control with PI controller

0

-0.05

-0.1
Df1,Df2,Df3, & Df4

-0.15

-0.2

-0.25

-0.3

-0.35
0 5 10 15 20 25 30 35 40 45
t(sec)

Figure 36

Conclusions-
From the responses of the Figures 34 and 36, we observe that the both results match with
each other and steady state frequency deviation is zero, and the frequency returns to its
nominal value in approximately 36 seconds. Maximum overshoot is -0.26.

3.3 Matlab Programme for Optimal Control Design of Controlled Four-Area LFC
(taking Q & R as identity matrices)
The matlab programme taking Q and R as identity matrices is given as
Kps1=100; Kps3=100; Kps2=110; Kps4=110; Tsg1=.4; Tsg2=.4; Tsg3=.4;
Tsg4=.4; Tt1=.5; Tt2=.5; Tt3=.5; Tt4=.5; Tps1=20; Tps2=20;
Tps3=20; Tps4=20; b1=.425; b2=.425; b3=.425; b4=.425; R1=3; R2=3; R3=3;
R4=3; W1=.02; W2=.04; W3=.05; W4=.06; Ki=.09;
T12=.00795; T31=.00795; T41=.00795; T34=.00795; T32=.00795; T42=.00795;
a12=1; a31=1; a41=1; a34=1; a32=1; a42=1;
A=[-1/Tps1 Kps1/Tps1 0 0 0 0 0 0 0 0 0 0 -Kps1/Tps1 (a31*Kps1)/Tps1
(a41*Kps1)/Tps1 0 0 0 0 0 0 0;
0 -1/Tt1 1/Tt1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0;
-1/(R1*Tsg1) 0 -1/Tsg1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 -Ki/Tsg1 0 0 0;
0 0 0 -1/Tps2 Kps2/Tps2 0 0 0 0 0 0 0 (a12*Kps2)/Tps2 0 0 0
(a32*Kps2)/Tps2 (a42*Kps2)/Tps2 0 0 0 0;
0 0 0 0 -1/Tt2 1/Tt2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0;
0 0 0 -1/(R2*Tsg2) 0 -1/Tsg2 0 0 0 0 0 0 0 0 0 0 0 0 0 -Ki/Tsg2 0 0;
0 0 0 0 0 0 -1/Tps3 Kps3/Tps3 0 0 0 0 0 -Kps3/Tps3 0 -Kps3/Tps3
-Kps3/Tps3 0 0 0 0 0;
0 0 0 0 0 0 0 -1/Tt3 1/Tt3 0 0 0 0 0 0 0 0 0 0 0 0 0;
0 0 0 0 0 0 -1/(R3*Tsg3) 0 -1/Tsg3 0 0 0 0 0 0 0 0 0 0 0 -Ki/Tsg3 0;
0 0 0 0 0 0 0 0 0 -1/Tps4 Kps4/Tps4 0 0 0 -Kps4/Tps4 (a34*Kps4)/Tps4 0
-Kps4/Tps4 0 0 0 0;
0 0 0 0 0 0 0 0 0 0 -1/Tt4 1/Tt4 0 0 0 0 0 0 0 0 0 0;
0 0 0 0 0 0 0 0 0 -1/(R4*Tsg4) 0 -1/Tsg4 0 0 0 0 0 0 0 0 0 -Ki/Tsg4;
2*pi*T12 0 0 -2*pi*T12 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0;
-2*pi*T31 0 0 0 0 0 2*pi*T31 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0;
-2*pi*T41 0 0 0 0 0 0 0 0 2*pi*T41 0 0 0 0 0 0 0 0 0 0 0 0;
0 0 0 0 0 0 2*pi*T34 0 0 -2*pi*T34 0 0 0 0 0 0 0 0 0 0 0 0;
0 0 0 -2*pi*T32 0 0 2*pi*T32 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0;
0 0 0 -2*pi*T42 0 0 0 0 0 2*pi*T42 0 0 0 0 0 0 0 0 0 0 0 0;
b1 0 0 0 0 0 0 0 0 0 0 0 1 -a31 -a41 0 0 0 0 0 0 0;
0 0 0 b2 0 0 0 0 0 0 0 0 -a12 0 0 0 -a32 -a42 0 0 0 0;
0 0 0 0 0 0 b3 0 0 0 0 0 0 1 0 1 1 0 0 0 0 0;
0 0 0 0 0 0 0 0 0 b4 0 0 0 0 1 -a34 0 1 0 0 0 0];
B=[-Kps1/Tps1 0 0 0; 0 0 0 0; 0 0 0 0; 0 -Kps2/Tps2 0 0; 0 0 0 0; 0 0 0
0; 0 0 -Kps3/Tps3 0; 0 0 0 0; 0 0 0 0; 0 0 0 -Kps4/Tps4; 0 0 0 0; 0 0 0
0;
0 0 0 0; 0 0 0 0; 0 0 0 0; 0 0 0 0; 0 0 0 0; 0 0 0 0; 0 0 0 0; 0 0 0 0;
0 0 0 0; 0 0 0 0];
Q=[1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0;
0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0;
0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0;
0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0;
0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0;
0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0;
0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0;
0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0;
0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0;
0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0;
0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0;
0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0;
0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0;
0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0;
0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0;
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0;
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0;
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0;
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0;
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0;
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0;
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1];
R=[1 0 0 0; 0 1 0 0; 0 0 1 0; 0 0 0 1];
K=lqr(A,B,Q,R);
Af=A-B*K;
t=0:0.2:15;
U=[W1; W2; W3; W4];
BU=B*U;
D=0;
C=[1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0; 0 0 0 1 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0;
0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0; 0 0 0 0 0 0 0 0 0 1 0 0 0 0
0 0 0 0 0 0 0 0];
[y,x]=step(Af,BU,C,D,1,t);
plot(t,y),grid,
title('Four Area ws L.F. Resp. With PI Controllers For Sudden Change in
Demands W1,W2,W3&W4 by .02pu,.04pu,.05pu&.06pu resp. in Normal Optimal
Case'),
xlabel('t,sec'),ylabel('f1pu,f2pu,f3pu,f4pu')

3.3.1 Matlab Programme Results

The matlab programme when run in work space, give the results as shown in Figure 37.
 ws Optimalized Controlled Four Area LFC For Sudden Change in Demands W1,W2,W3&W4
0

-0.005

-0.01

-0.015
f1pu,f2pu,f3pu,f4pu

-0.02

-0.025

-0.03

-0.035

-0.04

-0.045

-0.05
0 5 10 15
t,sec

Figure 37

CONCLUSION-
While comparing Figures 36 and 37, we observe that the steady state frequency
deviation is zero, and the frequency returns to its nominal value in approximately 14
seconds (previously it was 36 sec.). Maximum overshoot is-0.046 (previously it was
-0.26).Hence results are improved in optimal case than using only PI controller.

3.3.2 Optimal Control Design for Controlled Four-Area LFC (deriving Q & R as
based on the problem in hand)

Let PI (performance index) in quadratic form is given as under

 
1
J   [ X 'T QX  U 'T RU ]dt
20

With the reasons given before, we can write the PI ( or J) as

1 [(b1 x1  x13  a31 x14  a41 x15 )  (b2 x4  a12 x13  a32 x17  a42 x18 )  (b3 x7  x16  x14  x17 ) 
 2 2

J 
2 0 (b4 x10  a34 x16  x18  x15 )  ( x19 2  x20 2  x212  x22 2 )  k (u12  u2 2  u3 2  u4 2 )]dt

After expanding J changes to


1
2 0
J [b12 x12  b2 2 x4 2  b32 x7 2  b4 2 x10 2  (1  a12 2 ) x132  (1  a312 ) x14 2

 (1  a412 ) x152  (1  a34 2 ) x16 2  (1  a32 2 ) x17 2  (1  a42 2 ) x18 2

2b1 x1 x13  2a31 x13 x14  2a31a41 x14 x15  2b1a41 x1 x15  2b1a31 x1 x14  2a41 x13 x15
2b2 a12 x4 x13  2a12 a32 x13 x17  2a32 a42 x17 x18  2b2 a42 x4 x18 
2b2 a32 x4 x17  2a12 a42 x13 x18  2b3 x7 x16  2 x14 x16  2 x14 x17  2b3 x7 x17  2b3 x7 x14  2 x16 x17 
2b4 a34 x10 x16  2a34 x16 x18  2 x15 x18  2b4 x10 x15  2b4 x10 x18  2a34 x15 x16
 x19 2  x202  x212  x22 2  k (u12  u2 2  u32  u4 2 )]dt

From PI, Q and R can be recognized as

  b12 0 0 0 0 0 0 0 0 0 0 0 b1 b1a31 b1a41 0 0 0 0 0 0 0
 
 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
 
 0 0 0 b2 2 0 0 0 0 0 0 0 0 b2 a12 0 0 0 b2 a32 b2 a42 0 0 0 0
 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
 
 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
 0 0 0 0 0 0 b32 0 0 0 0 0 0 b3 0 b3 b3 0 0 0 0 0
 
 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
 
 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
 0 0 0 0 0 0 0 0 0 b4 2 0 0 0 0 b4 b4 a34 0 b4 0 0 0 0
 
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
Q
 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
 
 b1 0 0 b2 a12 0 0 0 0 0 0 0 0 1  a12 2  a31  a41 0 a12 a32 a12 a42 0 0 0 0
 b a 0 0 0 0 0 b3 0 0 0 0 0  a31 1  a312 a31a41 1 1 0 0 0 0 0
 1 31 
 b1a41 0 0 0 0 0 0 0 0 b4 0 0  a41 a31a41 1  a412 a34 0 1 0 0 0 0
 
 0 0 0 0 0 0 b3 0 0 b4 a34 0 0 0 1  a34 1  a34 2 1  a34 0 0 0 0
 0 0 0 b2 a32 0 0 b3 0 0 0 0 0 a12 a32 1 0 1 1  a32 2 a32 a42 0 0 0 0
 
 0 0 0 b2 a42 0 0 0 0 0 b4 0 0  a12 a42 0 1  a34 a32 a42 1  a42 2 0 0 0 0
 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0
 
 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0
 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0
 
 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 

1 0 0 0
0 1 0 0 
R
0 0 1 0
 
0 0 0 1

3.3.3 Matlab Programme Using Derived Q & R

Matlab programme using Q and R is given as

Kps1=100; Kps3=100; Kps2=110; Kps4=110; Tsg1=.4; Tsg2=.4; Tsg3=.4;

Tsg4=.4; Tt1=.5; Tt2=.5; Tt3=.5; Tt4=.5; Tps1=20; Tps2=20;
Tps3=20; Tps4=20; b1=.425; b2=.425; b3=.425; b4=.425; R1=3; R2=3; R3=3;
R4=3; W1=.02; W2=.04; W3=.05; W4=.06; Ki=.09;
T12=.00795; T31=.00795; T41=.00795; T34=.00795; T32=.00795; T42=.00795;
a12=1; a31=1; a41=1; a34=1; a32=1; a42=1;
A=[-1/Tps1 Kps1/Tps1 0 0 0 0 0 0 0 0 0 0 -Kps1/Tps1 (a31*Kps1)/Tps1
(a41*Kps1)/Tps1 0 0 0 0 0 0 0;
0 -1/Tt1 1/Tt1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0;
-1/(R1*Tsg1) 0 -1/Tsg1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 -Ki/Tsg1 0 0 0;
0 0 0 -1/Tps2 Kps2/Tps2 0 0 0 0 0 0 0 (a12*Kps2)/Tps2 0 0 0
(a32*Kps2)/Tps2 (a42*Kps2)/Tps2 0 0 0 0;
0 0 0 0 -1/Tt2 1/Tt2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0;
0 0 0 -1/(R2*Tsg2) 0 -1/Tsg2 0 0 0 0 0 0 0 0 0 0 0 0 0 -Ki/Tsg2 0 0;
0 0 0 0 0 0 -1/Tps3 Kps3/Tps3 0 0 0 0 0 -Kps3/Tps3 0 -Kps3/Tps3
-Kps3/Tps3 0 0 0 0 0;
0 0 0 0 0 0 0 -1/Tt3 1/Tt3 0 0 0 0 0 0 0 0 0 0 0 0 0;
0 0 0 0 0 0 -1/(R3*Tsg3) 0 -1/Tsg3 0 0 0 0 0 0 0 0 0 0 0 -Ki/Tsg3 0;
0 0 0 0 0 0 0 0 0 -1/Tps4 Kps4/Tps4 0 0 0 -Kps4/Tps4 (a34*Kps4)/Tps4 0
-Kps4/Tps4 0 0 0 0;
0 0 0 0 0 0 0 0 0 0 -1/Tt4 1/Tt4 0 0 0 0 0 0 0 0 0 0;
0 0 0 0 0 0 0 0 0 -1/(R4*Tsg4) 0 -1/Tsg4 0 0 0 0 0 0 0 0 0 -Ki/Tsg4;
2*pi*T12 0 0 -2*pi*T12 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0;
-2*pi*T31 0 0 0 0 0 2*pi*T31 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0;
-2*pi*T41 0 0 0 0 0 0 0 0 2*pi*T41 0 0 0 0 0 0 0 0 0 0 0 0;
0 0 0 0 0 0 2*pi*T34 0 0 -2*pi*T34 0 0 0 0 0 0 0 0 0 0 0 0;
0 0 0 -2*pi*T32 0 0 2*pi*T32 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0;
0 0 0 -2*pi*T42 0 0 0 0 0 2*pi*T42 0 0 0 0 0 0 0 0 0 0 0 0;
b1 0 0 0 0 0 0 0 0 0 0 0 1 -a31 -a41 0 0 0 0 0 0 0;
0 0 0 b2 0 0 0 0 0 0 0 0 -a12 0 0 0 -a32 -a42 0 0 0 0;
0 0 0 0 0 0 b3 0 0 0 0 0 0 1 0 1 1 0 0 0 0 0;
0 0 0 0 0 0 0 0 0 b4 0 0 0 0 1 -a34 0 1 0 0 0 0];
B=[-Kps1/Tps1 0 0 0; 0 0 0 0; 0 0 0 0; 0 -Kps2/Tps2 0 0; 0 0 0 0; 0 0 0
0; 0 0 -Kps3/Tps3 0; 0 0 0 0; 0 0 0 0; 0 0 0 -Kps4/Tps4; 0 0 0 0; 0 0 0
0;
0 0 0 0; 0 0 0 0; 0 0 0 0; 0 0 0 0; 0 0 0 0; 0 0 0 0; 0 0 0 0; 0 0 0 0;
0 0 0 0; 0 0 0 0];
Q=[b1*b1 0 0 0 0 0 0 0 0 0 0 0 b1 -b1*a31 -b1*a41 0 0 0 0 0 0 0;
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0;
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0;
0 0 0 b2*b2 0 0 0 0 0 0 0 0 b2*a12 0 0 0 -b2*a32 -b2*a42 0 0 0 0;
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0;
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0;
0 0 0 0 0 0 b3*b3 0 0 0 0 0 0 b3 0 b3 b3 0 0 0 0 0;
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0;
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0;
0 0 0 0 0 0 0 0 0 b4*b4 0 0 0 0 b4 -b4*a34 0 b4 0 0 0 0;
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0;
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0;
b1 0 0 b2*a12 0 0 0 0 0 0 0 0 1+(a12*a12) -a31 -a41 0 a12*a32 -a12*a42 0
0 0 0;
-b1*a31 0 0 0 0 0 b3 0 0 0 0 0 -a31 1+(a31*a31) a31*a41 1 1 0 0 0 0 0;
-b1*a41 0 0 0 0 0 0 0 0 b4 0 0 -a41 a31*a41 1+(a41*a41) -a34 0 1 0 0 0
0;
0 0 0 0 0 0 b3 0 0 -b4*a34 0 0 0 1 -a34 1+(a34*a34) 1 -a34 0 0 0 0;
0 0 0 -b2*a32 0 0 b3 0 0 0 0 0 a12*a32 1 0 1 1+(a32*a32) a32*a42 0 0 0
0;
0 0 0 -b2*a42 0 0 0 0 0 b4 0 0 -a12*a42 0 1 -a34 a32*a42 1+(a42*a42) 0 0
0 0;
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0;
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0;
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0;
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1];
R=[1 0 0 0; 0 1 0 0; 0 0 1 0; 0 0 0 1];
K=lqr(A,B,Q,R);
Af=A-B*K;
t=0:.02:15;
U=[W1; W2; W3; W4];
BU=B*U;
D=0;
C=[1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0; 0 0 0 1 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0;
0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0; 0 0 0 0 0 0 0 0 0 1 0 0 0 0
0 0 0 0 0 0 0 0];
[y,x]=step(Af,BU,C,D,1,t);
plot(t,y),grid,
title('Optimalized Controlled 4 Area LFC Resp.For Sudden Change in
Demands W1,W2,W3&W4'),
xlabel('t,sec'),ylabel('f1pu,f2pu,f3pu,f4pu')

3.3.4 Matlab Programme Results

The load frequency response of the above programme after running in work space of
matlab is shown in Figure 38.

Optimalized Controlled 4 Area LFC Resp.For Sudden Change in Demands W1,W2,W3&W4

0.01

-0.01

-0.02
f1pu,f2pu,f3pu,f4pu

-0.03

-0.04

-0.05

-0.06

-0.07

-0.08
0 5 10 15
t,sec

Figure 38 Matlab response for optimalized controlled four area l.f.control

(Kps1=100; Kps3=100; Kps2=110; Kps4=110; Tsg1=.4; Tsg2=.4; Tsg3=.4;
Tsg4=.4; Tt1=.5; Tt2=.5; Tt3=.5; Tt4=.5; Tps1=20; Tps2=20;
Tps3=20; Tps4=20; b1=.425; b2=.425; b3=.425; b4=.425; R1=3; R2=3; R3=3;
R4=3; W1=.02; W2=.04; W3=.05; W4=.06; Ki=.09;
T12=.00795; T31=.00795; T41=.00795; T34=.00795; T32=.00795; T42=.00795;
a12=1; a31=1; a41=1; a34=1; a32=1; a42=1)
CONCLUSION

While comparing Figures 37 and 38, we observe that the steady state frequency deviation
is zero, and the frequency returns to its nominal value in approximately 11 seconds while
previously it was 14 seconds when Q & R were taken identity matrices.Hence results are
improved.

FINAL RESULTS

Four Area Case

Without With PI Controller Optimal Case

Controller With PI Controller
Simulink Work- Simulik Work- Q as Q as
space space (Identity (Derived)
Matrix)
Over- -0.27 -0.27 -0.26 -0.26 -0.046 -0.079
Shoots
(pu)
Freq. -0.14 -0.14 0 0 0 0
Error
Fss (pu)
Settling 20 20 36 36 14 11
Time
(sec)