2012 IEEE Global Humanitarian Technology Conference

Simulation Study of Load Frequency Control of Single and Two

Area Systems
*Adil Usman and BP Divakar
Department of Electrical and Electronics Engineering
Reva Institute of Technology and Management
[email protected]*

Abstract—The load variation on a power system being unpredictable decoupled; the frequency is dependent on the active power
causes drifts in frequency and voltage from their nominal values while the voltage is highly dependent on the reactive power.
resulting in loss of generation due to tripping of lines and even Thus the control issue in power systems can be decoupled into
blackouts. Frequency and voltage drifts can be minimized and kept
within tolerable limits by automatic generation control. Automatic
two independent problems; one is about the active power and
generation control of a system consists of two parts; load frequency frequency control while the other is about the reactive power
control (LFC) and automatic voltage regulator (AVR). In this paper, and voltage control. As the load on a system is not constant
a simulation study is carried out to understand the operation of load there is a need for Automatic Frequency and Voltage
frequency controller by developing models in SIMULINK. The regulators at the generating stations.
simulation study helps students to understand the principle and As said above any mismatch between generation and
challenges behind load frequency controller. demand causes the system frequency to deviate from
scheduled value. Thus high frequency deviation may lead to
Index Terms –Automatic generation and control (AGC), Tie-line system collapse. This necessitates an accurate and fast acting
power, automatic load frequency control (ALFC), frequency bias
controller to maintain constant nominal frequency. The
tie-line control.
limitations of the conventional controls viz., Integral, PI, and
I. INTRODUCTION PID are slow and lack of efficiency in handling system non
linearities. Much research is ongoing to develop Artificial

A Power system is a combination of generation,

Intelligence based control technique [2]. For AGC Genetic
algorithm has been implemented for optimization purposes in
[3].The performance of the intelligent controllers has been
transmission and distribution networks and loads. The active
and reactive power demands from different loads vary compared with the conventional PI and PID controllers for the
continuously. The change in real power demand affects the single area system as well as two-area interconnected power
frequency while the change in the reactive power affects the system [4]. The result shows that artificial intelligence
voltage. The main constraint on a system is to keep the techniques have better dynamic response compared to PI
deviations of the frequency and the voltage constant. In an controller.
interconnected power system, as a power load demand varies
randomly, both area frequency and tie-line power interchange A. Motivation: (Global Humanitarian Benefit)
also vary. The objectives of load frequency control (LFC) are The syllabus of Power system operation and control in an
to minimize the transient deviations in theses variables (area undergraduate course in Electrical Engineering consists of
theoretical discussions on the automatic generation system and its
frequency and tie-line power interchange) and to ensure their
model. The primary focus in a typical class environment is to
steady state errors to be zeros. When dealing with the LFC
make students understand the function of an automatic generation
problem of power systems, unexpected external disturbances, control through the aid of models and mathematical equations.
parameter uncertainties and the model uncertainties of the The model consists of several individual blocks with specific
power system pose big challenges for controller designer. It is function. It is vital for students to understand the outputs of each
well known that three-phase alternating current (AC) is block to appreciate the control philosophy of AGC. However
adopted for generation and distribution of electricity. For students tend to focus on the result rather than on the process
efficient transmission and distribution proper balance of active which generates the desired result thereby failing to grasp the
and reactive power is essential otherwise the power system fundamentals of the control logic. On the other hand if the
may enter into instability mode resulting in possible outages students understand the principle underlying such a control they
[1]. As a result of the imbalance, the frequency and voltage will be in a position to appreciate the complexity of the problem
levels will oscillate around the steady state values and settle and thereby develop a liking towards this field. Hence the main
down to a finite value after a long delay which is undesirable objective of the present paper is to adopt the models developed in
from stability point of view. Thus a control system is essential [1] for simulation in order to enhance the learning capability of
students The main advantage of the simulation models is that
to cancel the effects of the random load changes and to keep
students can modify certain parameters of the models and thereby
the frequency and voltage at
gain firsthand experience in the operation of an AGC through
the standard values. Although the active power and reactive simulations. Many critical thinking questions set up by the tutor
power have combined effects on the frequency and voltage, will make students to focus on the functioning of individual
the control problem of the frequency and voltage can be blocks rather than focusing on the end result. The simulation
models can effectively be used as a case study to supplement

978-0-7695-4849-4/12 $26.00 © 2012 IEEE 214

DOI 10.1109/GHTC.2012.38
classroom lectures and thereby students can be motivated to carry Where ∆Po is the load disturbance and ∆Pref is the incremental
out further research work in this interesting area to design a speed reference setting.
robust power system. The recent blackout in the northern India
once again stresses the importance of having a robust system that
The Automatic Load Frequency Control (ALFC) loop shown
prevents blackout. This paper makes an attempt to educate future
engineers about the complexity of the problem that the humanity in fig.2 (a) is called the primary ALFC loop. It achieves the
is facing today primary goal of real power balance by adjusting the turbine
output ΔPm to match the change in load demand ΔPO.

Organization The transfer equation of the system can be given as:

The present paper is organized as follows: section II of this 1 1

KGs H s  
paper shows the typical block diagram of AGC .The AGC  
R 2 Hs  D  1   g s 1   T s 
(1)
model is explained in detail in this section with and without
the secondary loop. The section III describes simulation
results followed by the conclusion in section IV.
 s 

1   g s 1   T s  (2)
 PL s  2 Hs  D 1   g s 1   T s   1
II. AUTOMATIC GENERATION AND CONTROL
R

The schematic diagram of AGC is shown below. It consists of s   Pos T s  (3)
speed valve controller, turbine, generator and governor. Any
change in load is reflected on the frequency. The change in For the case with no frequency-sensitive load (D=0),
frequency is compared with a reference speed setting. A steam
valve controller is used to regulate the steam valve thereby ss   Po R (4)
increasing the power output from the generators resulting in
matching of generation and demand. As a result the frequency
is restored to the original value. The generators are grouped Where ∆ω is the frequency drift for the step change in load
together to form a coherent network. Based on this the ∆PL. The above equation gives the steady state value of the
network can be classified as single area or two area systems as frequency drift following a load disturbance and it can be
discussed below. inferred that the system’s new operating frequency will be less
than the nominal value due to the load disturbance. However
from the stability point of view, the frequency drift should be
brought down to zero or to a level acceptable for stable
operation and this is done with the help of a secondary loop
shown below.

B. Single Area (with secondary loop)

Fig. 1 Schematic of AGC [ 2 ].

A. Model of AGC of a single Area

Fig. 2 (b).Single Area with secondary Loop[ 2 ]

The ALFC loop shown in Fig. 2(a) achieves the primary goal
of real power balance by adjusting the turbine output ΔPm to
match the change in load demand ΔPo. But a change in load
Fig. 2 (a) Model of a Single Area without secondary Loop [ 2 ] results in a steady state frequency deviation  . The

215
restoration of the frequency to the nominal value requires an A typical control area consists of number of generators
additional control loop called the supplementary loop. This coupled together and operating in unison. The combination of
objective is met by using an integral controller which makes turbine-generators with AFC can be grouped under a system
the frequency deviation zero. The ALFC with the referred to as a Control Area. The two areas are connected by
supplementary loop is generally called the AGC. The block means of a tie line with a reactance X12.Then the power
diagram of an AGC is shown figure 2(b). delivered out of area 1 to area 2 will be given as P12. This
power is known as tie line power and is given in (6)
The main requirement in the frequency control is to make-
Δω=0. So the speed changer setting is changed in response to E1 E2
Δω(s) through an integrator. For this purpose the signal from P12  sin 12 (6)
X 12
Δω(s) is fed back through an integrator block (1/s) to adjust
ΔPref so as to bring the frequency to the steady state value. Where X12  X1  X tie  X 2 and 12  1   2 Equation
Because of the secondary loop the steady state value of
can be linearized
Δω(s)=0. Thus the integral action results in automatic
adjustment of ΔPref so as to make Δω=0. This action is rightly
dP12
called Automatic Generation Control. P12  12  Ps 12 (7)
d12 
120
The transfer function with the integral group is given below
[2], The tie-line power derivation

P12  Ps 1   2 

1
1
P ref  Po 
D
R (5) Consider a change in load PO1 in area1. The steady state
frequency deviation  is the same for both the areas. That
B. Two Area system and tie line flow
is   1  2 . Thus, for area1, we have [2]
1) With primary loop
Pm1  P12  Po1  D1 (8)

Pm1  Pm2  P12  D2 (9)

where, P12 is the tie line power flow from Area1to Area 2;
and for Area 2


Pm1  (10)
R1


Pm2  (11)
R2

Substituting these equations, yields

 1 
Fig. 2 (c). Two Area System (without secondary Loop) [ 2 ]   D1   P12  PD1 (12)
R
 1 
The model is similar to the Single Area System but with the  1  D   P (13)
 R 1 12
 1 
additional input of P12 . Suppose there is a change in load
( Po ) in area1, the frequencies of the two area systems
settle to a steady state value. Solving for  , we get
  1  2

216
  PO1  PO1 (14) 3) Two Area system (with secondary loop)

 1  D    1  D  1   2
 R 1  2
 1   R2 

 P011
P12  (15)
1   2
where, 1 and  2 are the composite frequency response
characteristic of Area1 and Area 2 respectively [2]. An
increase of load in area1 by PO1 results in a frequency
reduction in both areas and a tie-line flow of P12 . A
positive P12 is indicative of flow from Area1 to Area 2
while a negative P12 means flow from Area 2 to Area1.
Similarly, for a change in Area 2 load by PO 2 , we have
 as

 PO 2
  (16)
1   2
 PO 2 1
and P12  P21  (17) Fig. 2(d).Two Area System (with secondary Loop) [ 2 ]
1   2
ACEs are used as actuating signals to activate changes in the
2) Frequency bias tie line control reference power set points, and when steady-state is reached,

The tie line deviation reflects the contribution of regulation P12 and  will be zero. The integrator gain constant
characteristic of one area to another. The basic objective of must be chosen small enough so as not to cause the area to go
supplementary control is to restore balance between each area into a change mode. The block diagram of a simple AGC for a
load generation. This objective is met when the control action two-area system is shown in fig 2(d). Conventional LFC is
maintains frequency at the scheduled value. The
based upon tie-line bias control, where each area tends to
supplementary control should ideally correct only for changes
in that area. In other words, if there is a load change in Area1, reduce the area control error (ACE) to zero. The control error
there should be supplementary control only in Area1 and not for each area tends to consists of linear combination of
in Area 2. For this purpose the area control error (ACE) is frequency and tie-line error [2].
used [2]. The ACE of the two areas are given by
For area 1: ACE1 = P12  1 ACEi  nj  Pij  Ki (20)
For area 2: ACE2 = P21   2 
In an interconnected (multi area) system, there will be one
ALFC loop for each control area (located at the ECC of that
area). They are combined as shown in Fig 2(d). for the
interconnected system operation. For a total change in load of
PD the steady state deviation in frequency in the two areas
is given by [2]
PL1 PL1
   (18)
 1   1  1   2
  D1     D2 
 R1   R2 
Where
1 1
1   D1 (15) and  2   D2 (19)
R1 R2

217
 III. SIMULATION RESULTS

A. Single area system [without sec loop] Pm

1
1

del PL
10s+0.8
gen +load del ws  1

1 1
-30
0.5s+1 0.2s+1
turbine governor 1/R
Pref
Fig.3 (a) Single area system without secondary loop

Fig.3 (d) Simulation results of single area system with secondary loop

Fig. 3c shows the single area system with a secondary loop

where an integral controller with a gain Ki is adopted to adjust
the speed reference signal Pref so that  s returns to
zero.Fig.3 (d) shows the simulation results of single area with
the secondary loop and thus students can be motivated to think
how the frequency drift has been made zero due to the integral
loop.

C. Two area system[without secondary loop]

20 .6

B 1=(D1 + 1/R1)
20

1/R

Fig.3 (b) Simulation results of single area system without secondary loop
del PL

1 1 1

The models discussed in the previous sections are simulated in ACE

0.2s+1
governor
del Pv 0.5s+1
turbine
del Pm1 10 s+0.6
gen +load
del w1

SIMULINK. As seen from the plot above, the change in load

1
0.3
s
Gain 2 Integrator 1 del w1
del Pm1

brings about the change in speed which causes the variation in del P12 2
1
s del ws

the frequency  .The unsettled oscillations can be seen from

Ki Integrator

del Pm2

the graph. From the above plot students will be able to ACE
0.3

Gain 3
1
s
Integrator 2
1 1 1 del w2

understand that the frequency drift will settle down to a finite 0.3s+1 del Pv 0.6s+1 del Pm2 8s+0.9 del w2

governor 2 turbine 2 gen +load 2

value and that the new operating frequency will be lower than 16

the nominal value. 16 .9

1/R1

B 2= (D2+1/R2)

B. Single area system [with secondary loop]

Fig.3 (e) Two area system without secondary loop
del Pref

1 1 1
0.2s+1 del Pv 0.5s+1 del Pm 10s+0.8 del w

governor turbine gen +load

 2
del PL

del PL Scope

1
del PL

20

1/R

1
7 del w
s
Ki Integrator

Pm1
del Pref

Fig.3(c) Single area system with secondary loop

P12 Pm 2

Fig.3 (f) Simulation results of two area system without secondary loop

218
Fig.3 (e) shows the schematic of LFC of 2-area system As seen from the figures the secondary loop causes the return
without the secondary loop while fig.3 (f) shows the of frequency drifts to zero. Students will be able to play with
simulation results. As the two systems are interconnected, the the gain of the secondary loop and observe the change in the
frequency drifts of the two will settle down to equal value dynamic response of the system.
after some oscillations. The mechanical inputs of the two vary
to reduce the mismatch power between the electrical load in Observations: From the above simulation plots it can be
area 1 and the mechanical inputs. It can also be observed that observed that the system experiences frequency drift
area 2 will generate excess power to share the load change in following a load disturbance and it is mainly due to the
area 1. Students can observe the tie-line power flow following mismatch between the electrical load and the mechanical input
a load disturbance in area 1. Compared to the same result with to the turbine. The system oscillation is serious in single area
single area system, students will appreciate the stability system compared to two area system because all the load
improvement with interconnection. change in load is to be met by only one area. Also, using the
secondary loop in both the single area as well as the two area
C. Two area system[with secondary loop] system the change in frequency is brought to zero.

IV. CONCLUDING REMARKS

20 .6 A simulation study of single and two area system with

automatic generation and control is carried out with models
B1=(D1 + 1/R1)
20

1/R

developed in SIMULINK. The simulation of these systems

del PL has been carried out and results analyzed. The operation of
1
del Pv
1
del Pm1
1
del w1
single area and two area systems with and without secondary
0.2s+1 0.5s+1 10 s+0.6
ACE

0.3
1
governor turbine gen +load
loops are very well depicted through simulation models. The
s

del Pm1
Gain 2 Integrator 1 del w1 advantage of interconnection is best understood by comparing
del P12 2
1
s del ws the results of single and two area systems. It can be seen that
Ki Integrator

del Pm2 the oscillations due to change in load in any area is damped
ACE 1
down quickly because of tie line power flow. It can also be
0.3
s
Gain 3 Integrator 2
1
0.3s+1 del Pv
1
0.6s+1 del Pm2
1
8s+0.9 del w2
del w2
observed that the dynamic response is mainly governed by the
governor 2 turbine 2 gen +load 2
secondary loop and hence design criteria of which is
16
extremely vital for efficient implementation.
1/R1
16 .9

B2= (D2+1/R2)
It is envisaged that through these simulations the
Fig.3 (g) Two area system with secondary loop topic of automatic generation and control can be made very
interesting to students thus fulfilling the basic purpose of this
paper.
ACKNOWLEDGEMENT

 2 The authors acknowledge the support from Principal and

management of REVA ITM for this work.
1
REFERENCES
1 [1] Yao Zang, Tsinghua “Load Frequency Control Of Multiple-Area Power
Pm1 Systems”University July, 2007 Master of science in Electrical Engineering.

[2] I. J Nagrath and D. P Kothari Modern power system analysis- TMH 1993.
Pm 2
P12 [3] Sandeep Bhongade, H.O. Gupta, Dr. Barjeev Tyagi ,“Genetic Algorithm
based PID controller for Frequency Regulation Ancillary services”,
International Journal of Engineering Science and Technology
Vol. 2(12), 2010, 6902-6908.

[4] Mohamed. M .Ismail M. A. Mustafa Hassan “Load Frequency Control

Fig.3 (h) Simulation results of two area system with secondary loop Adaptation Using Artificial Intelligent Techniques for One and Two Different
Areas Power System”, International Journal Of Control, Automation And
Systems Vol. 1, No. 1, January 2012.

219