7.

TWO-AREA SYSTEM

A control area is characterized by the same frequency throughout. This

tantamount to saying that the area network is “rigid” or “strong”. In the single-
area case we could thus represent the frequency deviation by the single variable
Δf. In the present case we assume each area individually “strong”. Having
interconnected them with a “weak” tie-line therefore leads us to the assumption
that the frequency deviations in the two areas can be represented by two
variables Δf1 and Δf2 respectively.

7.1 MODELING THE TIE-LINE and BLOCK DIAGRAM FOR TWO-AREA SYSTEM

In normal operation the power flows in the tie-line connecting the areas 1 and 2 is
given by

0 0
V1 V2
sin (δ 1  δ 2 )
0 0
P012 = (47)
X

where δ10 and δ20 are the angles of end voltages V1 and V2 respectively. The order
of the subscripts indicates that the tie-line power is defined in direction 1 to 2.
Knowing dy/dx = Δy/Δx, for small deviations in angles δ 1 and δ2 the tie-line power
changes by an amount

0 0
V1 V2
ΔP12  cos (δ 1  δ 2 ) (Δδ1 – Δδ2)
0 0
(48)
X

We now define the “synchronizing coefficient” of a line as

0 0
V1 V2
T0 = cos (δ 1  δ 2 )
0 0
MW / rad. (49)
X

Then the tie-line power deviation is ΔP12 = T 0 (Δδ1 – Δδ2) MW (50)

We like to have ΔP12 in terms of frequency deviations Δf 1 and Δf2. We know

t
ω 1 dδ
f=  ; i.e. dδ  2 Π f dt Thus δ  2 π  f dt (51)
2 π 2 π dt 0

t
and hence Δδ = 2 π  Δf dt (52)

Expressing tie-line power in terms of Δf1 and Δf2 we get

t t
ΔP12 = 2 π T (  Δf1 dt -
0
 Δf 2 dt ) (53)
 t t
ΔP12 = 2 π T 0 (  Δf1 dt -  Δf2 dt ) (53)

Taking Laplace transformation of the above eq. we get

2 π T0
ΔP12(s) = [ Δf1 (s) – Δf2 (s) ] (54)
s
Representing this equation in terms of block diagram symbols yields the diagram
in Fig. 13.
Δf1 (s)

+ ΔP12 (s)
1
2 π T0
s
_

Δf2 (s)
Fig. 13 Representation of Tie-lone power flow

Tie-line power ΔP12 shall be treated as load in area 1. Similar to power balance eq.
Δf (s) = Gp (s) [ ΔPT(s) – ΔPD(s) ] (22) we can write

Δf1 (s) = Gp1 (s) [ ΔPT1(s) – ΔPD1(s) – ΔP12(s)] (54 a)

Block diagram for two-area uncontrolled system is shown in Fig. 14.

Δf1 (s) = Gp1 (s) [ ΔPT1(s) – ΔPD1(s) – ΔP12(s)] (54 a)

1
Δf1 (s)
R
1 ΔPD1 (s)

ΔPref 1 (s) _ ΔPg1 (s) ΔPT1 (s) _

+ GHT1 + Δf1 (s)
Gp1
_
+
ΔP12 (s) 1 0
2πT
s
_
-1

ΔPref 2 (s) ΔPg2 (s) ΔP21 (s)

ΔPT2 (s) _
+ GHT2 Gp2 Δf2 (s)
_ + _

ΔPD2 (s)
1
R
2 Δf2 (s)

Fig. 14 Block diagram representation of two area system

Similarly we need to add ΔP21 in area 2. Defining the tie-line power in direction 2
to 1 as ΔP21.

ΔP21 = - ΔP12 (55)

For this reason, transfer function of -1 is introduced in the block diagram.

Further, we remember that the powers in the single-area diagram were expressed
in per unit of area rating. The parameters R, D and H, likewise were based on the
same base power. When two or more areas of different ratings, are involved, we
must refer all powers and parameters to the one chosen base power.
7.2 STATIC RESPONSE OF UNCONTROLLED TWO-AREA SYSTEM

We shall first investigate the static response of the two-area system with fixed
speed changer position; i.e. ΔPref 1 = ΔPref 2 = 0

We assume that the loads in each area are suddenly increased by the constant
incremental steps ΔPD1= M1 and ΔPD2 = M2. At steady state, as seen from Fig. 15,
Δf1 0 = Δf2 0 = Δf0. We shall presently limit our analysis to finding the static
changes in frequency and tie-line power denoted by Δf0 and ΔP12 0 respectively.

1
Δf 0
R
1 ΔPD1 0
Δf 0
_ ΔPT1
ΔPg1 0 0 _
GHT1 +
Gp1
_ D1 Δf 0
+
ΔP12 0

1 1
Note that ΔPT1 0 = - Δf0 and ΔPT2 0 = - Δf0 (56)
R1 R2

Knowing the output of Gp1 as Δf0 input to Gp1 = D1 Δf0 . Similarly, input to Gp2 = D2
Δf0.
Consider the summing point with ΔPT1 0 , ΔPD1 0, and ΔP12 0.

1
Δf 0
R
1 ΔPD1 0
Δf 0
_ ΔPT1
ΔPg1 0 0 _
GHT1 +
Gp1
_ D1 Δf 0
+
ΔP12 0

1 1
- Δf0 - M1 - ΔP12 0 = D1 Δf0 and - Δf0 – M2 + ΔP12 0 = D2 Δf0 i.e.
R1 R2

1
D1 Δf0 + Δf0 + ΔP12 0 = – M1; i.e. β1 Δf0 + ΔP12 0 = – M1 (57)
R1

1
D2 Δf0 + Δf0 - ΔP12 0 = – M2; i.e. β2 Δf0 - ΔP12 0 = – M2 (58)
R2

where area frequency response characteristic (AFRC) of each area are defined as

1 1
β1 = D1 + β2 = D2 + (59)
R1 R2
 β1 Δf0 + ΔP12 0 = – M1 (57)

β2 Δf0 - ΔP12 0 = – M2 (58)

Solving the above equations (57) and (58) for Δf 0 and ΔP12 0 we get

M1  M 2
Δf0 = - (60)
β1  β 2

β 1 M 2  β 2 M1
ΔP12 0 = - ΔP21 0 = (61)
β1  β 2

Eqns. (60) and (61) become simple if we assume identical area parameters; i.e.

R1 = R2 = R; D1 = D2 = D; Then β1 = β2 = β
We then get

M1  M 2
Δf0 = - (62)
2β

M 2  M1
ΔP12 0 = - ΔP21 0 = (63)
2
For example, if a step load change occurs only in area 2, we get

M2
Δf0 = - Hz (64)
2β

M2
ΔP12 0 = - ΔP21 0 = p.u. MW (65)
2
The above two equations tell us, in a nutshell, the advantages of pool operation:
1. The frequency drop will be only half that would be experienced if the areas
were operating alone.
2. 50% of the added load in area 2 will be supplied by area 1 via. the tie-line.
Example 8
A 2000 MW control area 1 is interconnected with a 10000 MW area 2. The 2000 MW
area has the system parameters as

R = 2.0 Hz / p.u. MW; D = 0.01 p.u. MW / Hz

Area 2 has the same parameters, but on a base of 10000 MW. A 20 MW load
increase takes place in area 1. Find static frequency drop and tie-line power
change.

Solution
Let us take 2000 MW as base power.
1
β1 = 0.01 + = 0.51 p.u. MW / Hz
2
β2 = 0.51 p.u. MW / Hz on a base of 10000 MW
10000
= 0.51 x = 2.55 p.u. MW / Hz on a base of 2000 MW
2000
 20
M1 = = 0.01 p.u. MW
2000

0.01
Δf0 = - = - 0.003268 Hz
0.51  2.55

2.55 x 0.01
ΔP12 0 = - ΔP21 0 = - = - 0.008333 p.u. MW = - 16.67 MW
0.51  2.55

Note that the frequency drop is only one sixth of that experienced by area 1
operating alone (0.01961 Hz compare Example 6).

Also note that this frequency support is accomplished by an added delivery of

16.67 MW from the larger area.
7.3 DYNAMIC RESPONSE OF UNCONTROLLED TWO-AREA SYSTEM

With the very simple turbine model that we have used, the two area system in Fig.
14 is of eighth order for a step load change. It would therefore meaningless, to
attempt a direct analytic approach for finding the dynamic response of the
system. Typical simulation results for a two-area system are shown in Fig. 15.

Fig. 15 Simulation results of a two-area system

 Fig. 15 Simulation results of a two-area system

It can be seen that the solution for Δf1(t) and Δf2(t) are oscillatory. However,
because of system damping, finally they settle at a steady state value. Similarly,
the solution of ΔP21(t) has oscillation at beginning and finally settle at a steady
state value.
7.4 TIE-LINE BIAS CONTROL FOR TWO-AREA SYSTEM

The persistent static frequency error is intolerable. Also, a persistent static error
in tie-line power flow would mean that one area would have to support the other
on a steady state basis. To circumvent this, some form of reset integral control
must be added to the two-area system.

The control strategy of “tie-line bias control” is based upon the principle that all
operating pool members must contribute their share to frequency control in
addition to taking care of their own net interchange. This means that for two-area
system, at steady state, both Δf0 and ΔP12 0 must be zero.

To achieve these objectives, the Area Control Error (ACE) for each area consists
of a linear combination of frequency and tie-line error. Thus

ACE1 = ΔP12 + B1 Δf1 (66)

ACE2 = ΔP21 + B2 Δf2 (67)
The speed changer commands will thus be of the form

ΔPref 1 = - KI1  ( ΔP12 + B1 Δf1 ) dt (68)

ΔPref 2 = - KI2  ( ΔP21 + B2 Δf2 ) dt (69)

The constants KI 1 and KI 2 are integrator gains and the constants B1 and B2 are
the frequency bias parameters. The minus sign must be included to ensure that,
if there is positive frequency deviation or tie-line power deviation, then each area
should decrease its generation

7.5 STATIC SYSTEM RESPONSE WITH TIE-LINE BIAS CONTROL

The chosen strategy will eliminate the steady-state frequency and tie-line
deviations for the following reasons.

Following a step load change in either area, a new static equilibrium can be
achieved only after the speed-changer commands have reached constant values.
But this evidently requires that both the integrants in eqns (68) and (69) be zero;
i.e.

ΔP12 0 + B1 Δf 0 = 0 (70)

and ΔP21 0 + B2 Δf 0 = 0 i.e. - ΔP12 0 + B2 Δf 0 = 0 (71)

The above two conditions can be met with only if

Δf 0 = 0 and ΔP21 0 = - ΔP12 0 = 0 (72)

Note that this result is independent of the values of B1 and B2. In fact, one of the
bias parameters can be zero, and we still have a guarantee that eq. (72) is
satisfied. Having checked use of the integral controller, let us see how they can be
included in the block diagram.

Laplace transform of Equations (68) and (69) gives

K K
ΔPref 1 (s) =  I1 [ΔP12 (s) + B1 Δf1 (s)]; ΔPref 2 (s) =  I2 [ΔP21 (s) + B2 Δf2 (s)]
s s
 K K
ΔPref 1 (s) =  I1 [ΔP12 (s) + B1 Δf1 (s)]; ΔPref 2 (s) =  I2 [ΔP21 (s) + B2 Δf2 (s)]
s s

Δf1 (s) 1
Δf1 (s)
R
1 ΔPD1 (s)
B1
_ ΔPT1 (s)
+ ΔPg1 (s) _
K
 I1 GHT1 +
Gp1 Δf1 (s)
s + _
+ ΔPref 1 (s) +
ΔP12 (s) ΔP12 (s) 1 0
2πT
s
ΔP12 (s) _

-1 -1

ΔPref 2 (s) ΔPg2 (s) ΔP21 (s)

ΔP21 (s) ΔPT2 (s) _
+ KI 2
 + GHT2 Gp2 Δf2 (s)
s _ + _
+
ΔPD2 (s)
1
B2 R Δf2 (s)
2
Δf2 (s)

Fig. 16 Block diagram representation of two area system with tie-line bias control