IEEE International Conference on Computer, Communication and Control (IC4-2015)

Three Area Power System Load Frequency Control

Using Fuzzy Logic Controller
Sanjiv Kumar Jain1 Ashish Bhargava2 R. K. Pal3
Electrical Engineering Department Electrical Engineering Department Electrical Engineering Department
Medi-Caps Institute of Technology Medi-Caps Institute of Technology Medi-Caps Institute of Technology
And Management And Management And Management
Indore, India. Indore, India. Indore, India.
[email protected] [email protected] [email protected]

Abstract--Load frequency control has been used extensively in interchange power with neighboring areas connected through
power systems. The load frequency control has great advantage tie lines.
in terms of cost and reliability. This paper presents a load LFC is to regulate a signal called area control error (ACE),
frequency control method based on Fuzzy Logic controller which accounts for errors in the interconnection frequency as
(FLC). The main objective is to design a robust controller that
well as errors in the interchange power with frequency, as well
can ensure good performance. The paper present analysis on
dynamic performance of Load Frequency Control (LFC) of three as errors in the interchange power with neighboring areas [11]
area interconnected thermal non-reheat power system by the use over tie lines, i.e. the tie line power error. Conventional
of Fuzzy Intelligence. In the proposed scheme, control LFC uses a feedback signal that is based on the integral of the
methodology is Fuzzy Logic Control for three area ACE or is based on the ACE [6] and its integral type
interconnected thermal power system. In this proposed scheme, controller. These feedback signals are used to maneuver the
three fuzzy logic controllers for each area is implemented to get turbine governor set points of the generators so that the
zero frequency deaviations. The dynamic model of the power generated power follows the load fluctuations.
system and the controller design are elaborated in the paper. The A control strategy is needed that not only maintains
controller performances are simulated using
constancy of frequency and desired tie-power flow [14] but
MATLAB/SIMULINK simulation software. The FLC based
approach with Proportional Integral (PI) controller is shown. also be able to achieve zero steady state error and inadvertent
The simulation results are shown in view of settling time and interchange. AGC is the essential service in maintaining the
peak overshoot. system integrity by matching generation and demand in real-
Index Terms—Load frequency control (LFC), Fuzzy Logic Control time.
(FLC) , Area control error (ACE), Automatic generation control The aim of this paper is to stabilize the system frequency for
(AGC). step load disturbances with the help of Fuzzy Logic
Controller.
I. INTRODUCTION
With the restructuring of electrical power systems, Load
Frequency Control requirements gaining more attention due to
size and complexity of with increased interconnections [6].
In practice, power systems generally have more than two areas
and each area is different than others. Because of this, in the
study, the power system with three areas that consist of three
thermal units [8]. The active power and reactive power have
combined effects on the frequency and voltage, the control
problem of the frequency and voltage can be decoupled.
Load Frequency Control is one of the major requirements in
providing reliable and quality operation in multi-area power
systems. Load Frequency control has received great attention
of researchers in recent years also many control strategies
have been developed [1].
To regulate a signal known as area control error (ACE) load
frequency controller is used [4], which accounts for errors in
the interconnected system frequency (∆f) or errors in the

Authorized licensed use limited to: Mukesh Patel School of Technology & Engineering. Downloaded on November 07,2023 at 08:08:42 UTC from IEEE Xplore. Restrictions apply.
 IEEE International Conference on Computer, Communication and Control (IC4-2015)

௄೟
Nomenclature οܲீ ሺ‫ݏ‬ሻ ൌ  ο‫ݔ‬ா ሺ‫ݏ‬ሻሺ•‹ ‡ο்ܲ ൌ  οܲீ ሻ (2)
ଵା௦ఛ೟
F (s) : Nominal system frequency
xE : Valve movement 3) Generator-Load Model: The model for an isolated area is
Pc : Command power
‫ݏ݌ܭ‬
R : Speed governor droop ο‫ܨ‬ሺ‫ݏ‬ሻ ൌ  ൤ ൨ ሾοܲ‫ ܩ‬ሺ‫ݏ‬ሻ െ  οܲ‫ ܦ‬ሺ‫ݏ‬ሻሿ (3)
ͳ൅‫ݏ݌߬ݏ‬
D : Machine damping
Ksg: Speed governor gain
Гsg: Speed governor time constant ( for simplifications Ksg*Kt is considered unity)
Kt: Turbine gain
Гt: Turbine time constant
Kps: Machine gain
Гps: Machine time constant
PG : Turbine output power
Kt : Turbine gain
Гt : Turbine time constant Fig.2 Equivalent of Three areas Power System
PD : Power demand
PTL: Tie line power (steady state)
ACE: Area control error A three area interconnected power system is shown in fig.2
Tij : Synchronizing coefficient where three control areas are connected by tie-lines. In each
(i= 1,2,3; j=1,2,3; where i ≠ j) control area, all generators are assumed to form a coherent
b : Frequency bias group[12]. The three areas are considered with non-reheat type
turbine and of same parameters.

II. THREE AREA POWER SYSTEM MODEL The change in valve positions due to LFC of three areas are
given by
ଵ ଵ
A. Linearized Modeling three area LFC systems ο‫ݔ‬ாଵ ሺ‫ݏ‬ሻ ൌ  ቂοܲ஼ଵ ሺ‫ݏ‬ሻ െ 
ோభ
ο‫ܨ‬ଵ ሺ‫ݏ‬ሻቃ ൤
ଵା௦ఛೄ೒భ
൨ (4)

ଵ ଵ
ο‫ݔ‬ாଶ ሺ‫ݏ‬ሻ ൌ  ቂοܲ஼ଶ ሺ‫ݏ‬ሻ െ  ο‫ܨ‬ଶ ሺ‫ݏ‬ሻቃ ൤ ൨ (5)
ோమ ଵା௦ఛೄ೒మ

ଵ ଵ
ο‫ݔ‬ாଷ ሺ‫ݏ‬ሻ ൌ  ቂοܲ஼ଷ ሺ‫ݏ‬ሻ െ  ο‫ܨ‬ଷ ሺ‫ݏ‬ሻቃ ൤ ൨ (6)
ோయ ଵା௦ఛೄ೒య

The changes in turbine power are given by

ଵ
Fig.1 Schematic diagram of an isolated LFC. οܲீଵ ሺ‫ݏ‬ሻ ൌ  ο‫ݔ‬ாଵ ሺ‫ݏ‬ሻ (7)
ଵା௦ఛ ೟భ

An isolated single area model is shown in fig.1. The model for οܲீଶ ሺ‫ݏ‬ሻ ൌ 
ଵ
ο‫ݔ‬ாଶ ሺ‫ݏ‬ሻ (8)
Prime mover and Governor, Generator and Load is shown ଵା௦ఛ೟మ

with the help of transfer function method and combined to ଵ

form a complete block diagram for an isolated power system οܲீଷ ሺ‫ݏ‬ሻ ൌ  ο‫ݔ‬ாଷ ሺ‫ݏ‬ሻ (9)
ଵା௦ఛ೟య
[4].
The frequency deviations of each control areas are
1) Speed Governor Model: The purpose of the speed governor
is to monitor the turbine generator speed and the frequency ‫ͳݏ݌ܭ‬
response is adjusted by controlling the valves. ο‫ܨ‬ଵ ሺ‫ݏ‬ሻ ൌ  ቈ ቉ ሾοܲ‫ ͳܩ‬ሺ‫ݏ‬ሻ െ  οܲ‫ ͳܦ‬ሺ‫ݏ‬ሻ െ  οܲܶ‫ ͳܮ‬ሺ‫ݏ‬ሻሿ
ͳ ൅ ‫ͳݏ݌߬ݏ‬
ଵ ௄ೞ೒
…. (10)
ο‫ݔ‬ா ሺ‫ݏ‬ሻ ൌ  ቂοܲ஼ ሺ‫ݏ‬ሻ െ  ο‫ܨ‬ሺ‫ݏ‬ሻቃ ൤
ோ ଵା௦ఛೞ೒
൨ (1) ‫ʹݏ݌ܭ‬
ο‫ܨ‬ଶ ሺ‫ݏ‬ሻ ൌ  ቈ ቉ ሾοܲ‫ ʹܩ‬ሺ‫ݏ‬ሻ െ  οܲ‫ ʹܦ‬ሺ‫ݏ‬ሻ െ  οܲܶ‫ ʹܮ‬ሺ‫ݏ‬ሻሿ
ͳ ൅ ‫ʹݏ݌߬ݏ‬
2) Turbine Model: The model for non-reheat turbine is …. (11)

Authorized licensed use limited to: Mukesh Patel School of Technology & Engineering. Downloaded on November 07,2023 at 08:08:42 UTC from IEEE Xplore. Restrictions apply.
 IEEE International Conference on Computer, Communication and Control (IC4-2015)

‫͵ݏ݌ܭ‬
ο‫ܨ‬ଷ ሺ‫ݏ‬ሻ ൌ  ቈ ቉ ሾοܲ‫ ͵ܩ‬ሺ‫ݏ‬ሻ െ  οܲ‫ ͵ܦ‬ሺ‫ݏ‬ሻ െ  οܲܶ‫ ͵ܮ‬ሺ‫ݏ‬ሻሿ
ͳ ൅ ‫͵ݏ݌߬ݏ‬ B. Fuzzy Logic Controller
…. (12)

The tie line power deviations for three control areas are

బ బ
ଶగ்భమ ଶగ்భయ
ο்ܲ௅ଵ ሺ‫ݏ‬ሻ ൌ  ሾο‫ܨ‬ଵ ሺ‫ݏ‬ሻ െ  ο‫ܨ‬ଶ ሺ‫ݏ‬ሻሿ ൅  ሾο‫ܨ‬ଵ ሺ‫ݏ‬ሻ െ
௦ ௦
ο‫ܨ‬ଷ ሺ‫ݏ‬ሻሿ (13)
బ
ଶగ்మభ బ
ଶగ்మయ
ο்ܲ௅ଶ ሺ‫ݏ‬ሻ ൌ  ሾο‫ܨ‬ଶ ሺ‫ݏ‬ሻ െ  ο‫ܨ‬ଵ ሺ‫ݏ‬ሻሿ ൅  ሾο‫ܨ‬ଶ ሺ‫ݏ‬ሻ െ  ο‫ܨ‬ଷ ሺ‫ݏ‬ሻሿ
௦ ௦
(14)
బ బ
ଶగ்యభ ଶగ்యమ
ο்ܲ௅ଷ ሺ‫ݏ‬ሻ ൌ  ሾο‫ܨ‬ଷ ሺ‫ݏ‬ሻ െ  ο‫ܨ‬ଵ ሺ‫ݏ‬ሻሿ ൅  ሾο‫ܨ‬ଷ ሺ‫ݏ‬ሻ െ  ο‫ܨ‬ଶ ሺ‫ݏ‬ሻሿ
௦ ௦
(15)
Fig. 3. Diagram of fuzzy inference system.
Turbine reference power of each area is tried to be set to its
nominal value by an integral controller and the input of the
integral controller of each area is known as area control error Fig. 3 shows the fuzzy inference system [5], to implement
(ACE) of the same area [14]. The same for three areas are fuzzy logic technique to a real application requires the
following three steps:

‫ܧܥܣ‬ଵ ሺ‫ݏ‬ሻ ൌ  ο்ܲ௅ଵ ሺ‫ݏ‬ሻ ൅  ܾଵ ο‫ܨ‬ଵ ሺ‫ݏ‬ሻ (16) 1) Fuzzification: convert classical data or crisp data into fuzzy
data or Membership Functions (MFs) as shown in fig. 4.
‫ܧܥܣ‬ଶ ሺ‫ݏ‬ሻ ൌ  ο்ܲ௅ଶ ሺ‫ݏ‬ሻ ൅  ܾଶ ο‫ܨ‬ଶ ሺ‫ݏ‬ሻ (17) 2) Fuzzy Inference Process: combine membership functions
with the control rules to derive the fuzzy output.
‫ܧܥܣ‬ଷ ሺ‫ݏ‬ሻ ൌ  ο்ܲ௅ଷ ሺ‫ݏ‬ሻ ൅  ܾଷ ο‫ܨ‬ଷ ሺ‫ݏ‬ሻ (18) 3) Defuzzification: use different methods to calculate each
associated output and put them into a table: the lookup table. It
ACE1 = area control error of area 1, pick up the output from the lookup table based on the current
input during an application.
ACE2 = Area control error of area 2,
ACE3 = Area control error of area 3.

III. CONTROL METHODOLOGY

A. PI Controller
One of the most widely used controls with thermal and
hydro power plants governing systems is the PI type
controller. The proportional-integral controller [10] generates
an output signal consisting of two terms- one proportional to Fig. 4. Fuzzy logic controller inputs and output membership function.
error signal and the other proportional to the integral error
signal. Due to the faster transient response Proportional
controller is used to achieve the steady state condition much IV. SIMULATIONS
quicker. The proportional term of the controller produces a
control signal proportional to the error in the system, so that The proposed three area load frequency controller was
u(t) = Kp*e(t). implemented and the performance was tested with Matlab/
Typically, given a step change of load demand, low values Simulink software.
of Kp give rise to stable responses with large steady-state
errors. Higher values of Kp give better steady-state
performance, but worse transient response. Therefore, the
higher value of Kp is used to reduce the steady state error,
although increasing the gain Kp decreases the system time
constant and damping. Therefore it is evident to choose the
optimum value of Kp. The proportional action can never
eliminate the steady state error

Authorized licensed use limited to: Mukesh Patel School of Technology & Engineering. Downloaded on November 07,2023 at 08:08:42 UTC from IEEE Xplore. Restrictions apply.
 IEEE International Conference on Computer, Communication and Control (IC4-2015)

B1

-K-

1 1 Kp1
IN OUT
Ts1.s+1 Tt1.s+1 Tp1.s+1
Fuzzy Controller

1
-K-
s
Fig. 6 Fuzzy logic controller in matlab/Simulink
-K-

V. RESULTS AND ANALYSIS

B2

-K- The simulations are done on Matlab/ Simulink environment

for the three area control. A step disturbance is considered for
IN OUT
1 1 Kp2 the three same areas. With the help of fuzzy controller with
Fuzzy Controller
Ts2.s+1 Tt2.s+1 Tp2.s+1
twenty five rules, the frequency deviations are set to zero with
1
minimum settling time and overshoot.
-K-
s
Fig.7 shows the surface plot of FLC for the variations of K i
-K-
with respect to error and derivation of error.

B3

-K-

1 1 Kp3
IN OUT
Ts3.s+1 Tt3.s+1 Tp3.s+1
Fuzzy Controller

1
-K-
s

-K-

Fig..7 The variations of Ki with respect to e and de, Surface plot.

Fig.5. Three area power system simulation model

In this model shown by fig.5, the application of fuzzy

controller [3] for load-frequency control in power systems is The frequency deviations for the first area is illustrated by
used. For the same, interconnected power system having three fig.8, which clearly shows the maximum overshoot of around -
control areas including same turbine units. In the system, non- 0.02, i.e. 0.04 percent of nominal frequency of 50 Hz and a
reheat turbines are used for each area. So, same properties and good settling time of 9 to 10 seconds for a step change in load
physical constants of the areas above are considered in the after 2 seconds. The same is the case for the areas two and
simulation. In the simulation, a step load increment in the three as shown by fig.9 and fig. 10 respectively. The figures
three areas of power system is considered [7]. also shows the dynamics which are controlled by the FLC to
get the steady zero frequency error.
The area control error for each area is controlled with FLC to
From the simulation study, the fuzzy logic controller is very
optimize the integral coefficient and hence to achieve the zero
effective to suppress the frequency oscillations caused by load
frequency steady state error. The fuzzy model used is
disturbances.
mamdani and bisector is used as defuzzification method.

In the decision-making stage, total twenty five rules are

used, which are specified by a set of IF–THEN statements [9]
define the controller behaviour. The fuzzy simulation model in
Simulink is shown in fig. 6.

Authorized licensed use limited to: Mukesh Patel School of Technology & Engineering. Downloaded on November 07,2023 at 08:08:42 UTC from IEEE Xplore. Restrictions apply.
 IEEE International Conference on Computer, Communication and Control (IC4-2015)

Frequency deviation for area one advantages to the consumer and the management. The
0.01 fuzzy control machines corrosions, used in the system can be stopped by
0 lowered overshoot of the system outputs. Hence, the life of
machines and power plants are longer.
delf1

-0.01

-0.02 REFERENCES

-0.03 [1] Hassan Yousef, “Adaptive fuzzy logic load frequency control of
0 20 40 60 80 100
Time in seconds
multi-area power system”, International Journal of Electrical
Fig.8. Plot for change in frequency for Area one Power & Energy Systems Volume 68, June 2015, Pages 384–
395, January 2015.
Frequency deviation for area two [2] Anand B., Ebenezer A. Jeyakumar, “Load frequency control
0.005
Fuzzy control
with fuzzy logic controller considering non-linearities and boiler
0 dynamics”, ICGST-ACSE Journal, ISSN 1687-4811, Volume 8,
issue 111, pp 15-20, 2009.
-0.005
[3] Aravindan P., Sanavullah M.Y., “Fuzzy Logic Based Automatic
Load Frequency Control of Two Area Power System With
delf2

-0.01

-0.015
GRC”, International Journal of Computational Intelligence
Research, Volume 5, Number 1. pp. 37–44, 2009.
-0.02 [4] P. Kundur, Power System Stability and Control. New York:
-0.025
McGraw-Hill, 1994, pp. 581–623.
0 10 20 30 40 50 60
Time in seconds
70 80 90 100 [5] Mines J. N. 1997. MATLAB Supplement to Fuzzy & Neural
approach in Engineering, John Wiley NY.
Fig.9. Plot for change in frequency for Area two [6] George-Gross, Lee J.W., “Analysis of Load Frequency Control
Performance Assessment Criteria”, IEEE transaction onPower
Frequency deviation for Area three System, Vol. 16, No. 16, pp. 520-525,2001.
0.005
Fuzzy control [7] Nanda J., Kakkarum J.S., “Automatic Generation Control with
0
Fuzzy logic controllers considering generation constraints”, In
Proceeding of 6th Int Conf on Advances in Power System
-0.005 Control Operation and managements, Hong Kong, 2003.
[8] Magla A., Nanda J., “Automatic Generation Control of an
delf3

-0.01
Interconnected Hydro- Thermal System Using Conventional
-0.015
Integral and Fuzzy logic Control”, In Proc. IEEE Electric Utility
Deregulation, Restructuring and
-0.02 PowerTechnologies.(DRPT2004), Hong Kong, pp. 372-377,
2004.
-0.025
0 10 20 30 40 50 60 70 80 90 100 [9] Anand B., Ebenezer A. Jeyakumar., “ Load frequency control
Time in seconds
with fuzzy logic controller considering non-linearities and boiler
Fig.10. Plot for change in frequency for Area three
dynamics”, ICGST-ACSE Journal, ISSN 1687-4811, Volume 8,
issue 111, pp 15-20., 2009.
VI. CONCLUSION [10] Chatterjee K.. “Design of Dual Mode PI Controller for Load
Frequency Control”, International Journal of Emerging Electric
In this study, FLC application to three area generation Power System. Vol. 11, Issue 4, Article 3, 2010.
controls is considered for the frequency deviations of each [11] Demiroren A., Zeynelgil H.L., Sengor N. S., “The Application
area [7]. The transient behavior of frequency for the load of ANN Technique to Load-frequency Control for Three- area
perturbation in areas is studied. In practice, power systems Power System”, IEEE Porto Power Tech Conference, PPT001,
generally have more than three areas and each area is different 10th -13th September2001, Porto, Portugal. 2001
than others. Because of this, in the study, the power systems [12] Nasser Jalleli, Donald N. Ewart, Lester H. Fink, Louis S.
Vanslyck, “Understanding Automatic Generation Control”,
with three areas that consist of thermal units are considered.
IEEE Transactions on power systems, vol. 7, No. 3, August
The simulation results obtained shows the performance of 1992..
FLC controller against to the load perturbation at each area in [13] Gayadhar-Panda, Sidhartha-Panda, Cemal Ardil, “Automatic
the considered power system. Generation Control of Interconnected Power System with
From the analysis, the settling time of controllers was reached Generation Rate Constraints by Hybrid Neuro Fuzzy Approach”,
in the time region of 9 to 10 seconds and overshoot is very low World Academy of Science, Engineering & Technology 52, pp-
(0.04 percent of nominal frequency of 50 Hz). 543-548. 2009.
From the study, the fuzzy logic improves the system [14] Concordia C, Kirchmayer LK. Tie-line power frequency control
performance by effectively reducing the overshoot. Also the ofelectrical power systems. Part I. Trans AIEE 1953:562̢72.
stability of the system was thus maintained effectively by the
proposed controller. Decreased settling time in power systems
significantly decreases generating cost, providing economical APPENDIX

Authorized licensed use limited to: Mukesh Patel School of Technology & Engineering. Downloaded on November 07,2023 at 08:08:42 UTC from IEEE Xplore. Restrictions apply.
 IEEE International Conference on Computer, Communication and Control (IC4-2015)

Parameters used for three areas are:

R=2.4;
Гsg=0.08;
Гt=0.28;
Гps=18;
Kps=120;
T12=0.06;
T13=0.08;
Ki=.014;
T21=0.06;
T23=0.08;
T31=0.08;
T32=0.06;
Ki3=.014;
b1=0.425;
b2=0.425;
b3=0.425.
----------------------------------------

Authorized licensed use limited to: Mukesh Patel School of Technology & Engineering. Downloaded on November 07,2023 at 08:08:42 UTC from IEEE Xplore. Restrictions apply.