Journal of Applied Engineering, Technology and Management (JAETM)

LOAD FREQUENCY CONTROL OF INTERCONNECTED POWER

SYSTEM GRID INVOLVING WIND AND HYDROPOWER PLANT AND
COMPARATIVE ANALYSIS
Pardeep Rai1, Bikram Chhetri2, Chimi Dem3, Khalishore Chhetri 4, Phuntsho Tashi5, Sanjita Lepcha6
Department of Electrical Engineering, Jigme Namgyel Engineering College, Dewathang
, [email protected] 2, [email protected], [email protected],
[email protected], [email protected], [email protected]

Abstract— Power demand is expanding day by day. The greatest test is to give continuous and top-
notch power to clients in factor conditions particularly when we interconnect the two areas utilizing
power tie-line. To achieve this, the two boundaries should consistently be checked for each condition,
these boundaries are Load Distribution and Load Frequency Control (LFC).The primary work of the
load frequency control is to regulate the power output of the generator within a specified area with
respect to change in system frequency and tie-line power, such as to maintain the scheduled system
frequency and power interchange with other areas in a prescribed limit [1][2]. In this paper, the study
of LFC system for two areas consisting of Hydropower Plant and Wind Power Plant are carried out. The
fuzzy gain scheduled proportional-integral (FGSPI) and fuzzy gain scheduled proportional integral
derivative (FGSPID) controllers are designed for load frequency control (LFC) of a two-area interconnected
power system. The proposed FGSPI and FGSPID controllers are compared against conventional
proportional-integral (PI) controllers and Proportional integral derivation (PID) controllers concerning
settling times and peak overshoots of the tie-line power and frequency deviations as performance indices.
Comparative analysis indicates that the proposed intelligent controller gives better performance than
conventional controllers. Simulations have been performed using Simulink toolbox in MATLAB.

Keywords— load frequency control, wind power plant, hydropower plant, two area network, area control
error, PI, PID, fuzzy gain scheduling

I. INTRODUCTION
The objective of the load frequency control (LFC) is to maintain the scheduled frequency and scheduled
tie-line power in a normal mode of operation, during the small perturbation in operating conditions. The
input mechanical power is used to control the frequency of the generators and the change in the frequency
and tie-line power are sensed, which is a measure of the change in rotor angle. A well-designed power
system should be able to provide acceptable levels of power quality by keeping the frequency and voltage
magnitude within tolerable limits [3]. The enormous interconnected power systems are made out of control
areas that represent different units of generators. The different areas are interconnected through tie-lines.
The tie-lines are utilized for exchanging the power between the consecutive two areas and provides inter-
area support in case of abnormal conditions of the power system.
number of conventional controllers like Proportional(P), Integral(I), Proportional Integral (PI) and
Proportional Integral Derivatives (PID) are utilized in a control system for controlling frequency deviation

JAETM, VOLUME I, ISSUE I, JUNE 2021 ISSN (PRINT): 2707-4978 & (ONLINE): 2789-0848 45
 Journal of Applied Engineering, Technology and Management (JAETM)

and tie-line power, as these controllers are simple to implement, easy to understand and have low cost. The
nature of their control technique is dependable and announced as vigorous for some working conditions [4].
However, the response of a system with these controllers is slow and poor in comparison to the intelligent
controller. Zadeh presented a fuzzy set hypothesis and the first fuzzy logic control algorithm was
implemented by Mamdani on a steam motor. The enormous, complex and interconnected power systems
suffer with countless nonlinear properties subsequently, the fuzzy logic controller is one of the better
controllers for the systems [5]. In this paper, Chhukha Hydro Power Plant and Rubesa Wind Power Plant
are taken as the two-area network for frequency and power deviation control. The FGSPI and FGSPID
controllers and conventional controllers (PI and PID) are used for LFC and they are compared base on their
settling time and peak overshoot.

II. SYSTEM MODELLING

In this paper, Hydropower Plant and Wind Power Plant were taken as Area-1 and Area-2 respectively. Area-
1 consists of four units and Area-2 with two units. The system model consists of hydro-governor, wind
turbine, hydro turbine and generators of both wind and hydropower plants and each component is
represented in the transfer function. The speed regulation constant and frequency bias factor are the
feedback to the system. The frequency deviation ∆f1 is for Area-1 and ∆f2 for Area-2. ∆PL1 and ∆PL2 are the
power demand increment for Area-1and Area-2 respectively and it is given in step load form. Area-1 and
Area-2 are interconnected by the tie-line power. The area control error (ACE) for the two areas is given to
the two controllers [19]. ACE is given in equation (1).

ACEi = ∆P , + B ∆f (1)

Where, Bi = Di + 1/Ri (2)

Where:
 ACEi = area control error of the ith area
 ∆fi = frequency error of ith area
 ∆Ptie,ij = tie-line power flow error between ith and jth area
 Bi = frequency bias coefficient of ith area

JAETM, VOLUME I, ISSUE I, JUNE 2021 ISSN (PRINT): 2707-4978 & (ONLINE): 2789-0848 46
 Journal of Applied Engineering, Technology and Management (JAETM)

Fig. 1. Two area multi-unit Hydro and Wind power system

III. CONTROLLERS
In this study, PI, PID, FGSPI and FGSPID controllers are used to control the load frequency and tie-line
power of the two-area network. The conventional controller PI and PID are compared with the proposed
hybrid FGSPI and FGSPID controllers.
A. PI Controller
The proportional-integral controller produces an output, which is the combination of outputs of the
proportional and integral controllers. PI controller will eliminate forced oscillations and steady-state error
resulting in the operation of the on-off controller. PI controllers are very often used in industry, especially
when the speed of the response is not an issue [23].
The value of the controller output u(t) is fed into the system as the manipulated variable input. Output power
equals the sum of proportion and integration coefficients[24].

u(t) = K e(t) + K ∫ e(t)dt (3)

Apply Laplace transform on both sides

U(s) = (K + )E(s) (4)

U(s)/E(s) = K + S (5)
Therefore, the transfer function of proportional integral controller is KP + KIs.

JAETM, VOLUME I, ISSUE I, JUNE 2021 ISSN (PRINT): 2707-4978 & (ONLINE): 2789-0848 47
 Journal of Applied Engineering, Technology and Management (JAETM)

The block diagram of the unity negative feedback closed loop control system along with the proportional-
integral controller is shown in the figure 2.

Fig. 2. Schematic diagram of PI controller

The proportional-integral controller is used to decrease the steady-state error without affecting the stability
of the control system.
B. PID Controller
The PID algorithm is the most commonly used feedback controller. It is a robust easily understood algorithm
that can provide excellent control performance despite the varied dynamic characteristic of the process. As
the name suggests, the PID consists of three basic modes namely the proportional, integral and derivatives
mode. When utilizing the PID algorithm it is necessary to decide which mode to be used (P, I & D) and
then specify the parameter for each mode used. Generally, three basic algorithms P, PI and PID are used to
automatically adjust some variables to hold a measurement to the desired variable [25][26].

Fig. 3. Schematic diagram of PID controller

The output of PID controller in time domain

u(t) = K e(t) + K ∫ e(t)dt + Kd d/dt e(t)

Taking the Laplace on both the side:

U(s) = (K + + K S)E(s) (6)
( )
= (K + + K S) (7)
( )
( )
= (8)
( )
Whereas, E(s) and U(s) are the input and output of the plant respectively.

JAETM, VOLUME I, ISSUE I, JUNE 2021 ISSN (PRINT): 2707-4978 & (ONLINE): 2789-0848 48
 Journal of Applied Engineering, Technology and Management (JAETM)

a. Ziegler-Nichol’s Tuning Method

Ziegler and Nichols created two strategies for controller tuning during the 1940s. The idea was to tune the
controller dependent on the following idea: Make a simple experiment, extract some features of process
dynamics from the experimental data, and determine controller parameters from the features [27][28][29].
The following processes are followed:
 Increase the gain until the loop starts oscillating. Note that linear oscillation is required and that it
should be detected at the controller output.
 Record the controller critical gain Kp = Kc and the oscillation period of the controller output, Pc.
 Adjust the controller parameters according to the Table given below.

TABLE 1. ZiZIEGLER-NICHOLS TUNING RULE BASED ON CRITICAL GAIN Kc AND CRITICAL PERIOD Pc

Kp Ti Td

P Controller 0.5Kc ∝ 0

PI Controller 0.45Kc Pc/1.2 0

PID Controller 0.6Kc 0.5Pc Pc/8

Simultaneously, the value of Ki and Kd can be calculated using the following formula:
Ki = Kp/Ti (9)
Kd = Kp*Td (10)

Whereas, Ti and Td are the Integral time and Derivative time respectively.

C. Fuzzy Gain schedule PI and PID controllers

Gain scheduling is the common technique used in the system whose dynamic changes non-linearly with
operating conditions. Here, the fuzzy logic set supervises and modifies the operations, i.e., gain scheduling
of the conventional controllers. FGSPI and FGSPID controllers are designed and used to control the
frequency deviation of two area networks [30]. The design of FGSPI and FGSPID controllers can be divided
into three parts:
 The allocation of the proper inputs
 The determination of the rules associated with inputs
 The defuzzification of the output

JAETM, VOLUME I, ISSUE I, JUNE 2021 ISSN (PRINT): 2707-4978 & (ONLINE): 2789-0848 49
 Journal of Applied Engineering, Technology and Management (JAETM)

Fig. 4. Schematic diagram showing the working processes of FGSPI and FGSPID Controller
a. Fuzzification
In this process, the precise numerical values obtained by measurement are converted to membership values
of the various linguistics. For this controller, the two inputs are the Area Control Error (ACE) and the change
in error [ d/dt (ACE)] [31].
b. Fuzzy Rule Base
For the proposed controller, the Mamdani method was selected and realized by five triangular membership
functions for each of the three linguistic variables (ACE d/dt(ACE), K) with a suitable choice of intervals
of the membership functions, where ACE and d/dt(ACE) act as the inputs of the controller and K is the
output of the controller. In table II below, NB, NS, Z, PS, PB represent negative big, negative small, zero,
positive small, and positive big respectively. The rule base has been formed in such a manner. For example;
If ACE is NB and d(ACE)/dt is NB then the controller action is PB [33]. The appropriate rules used in the
study are given in table 2.

TABLE 2. FUZZY LOGIC RULE FOR FGSPI AND FGSPID CONTROLLER

ACE
NB NS Z PS PB
NB PB PB PB PB Z
d/dt(ACE) NS PB PB PS Z Z
Z PS PS Z NS NS
PS Z Z NS Z NB
PB Z NS NB NB NB

c. Defuzzification

The transformation of a fuzzy set into a numeric value is called defuzzification. Before feeding the data to
the system it is very much important to do the defuzzification of the fuzzy set. The data needs to be

JAETM, VOLUME I, ISSUE I, JUNE 2021 ISSN (PRINT): 2707-4978 & (ONLINE): 2789-0848 50
 Journal of Applied Engineering, Technology and Management (JAETM)

converted into the numerical value from the membership function before feeding to the system[35]. The
output from the fuzzy logic controller is used to schedule the gain of PI and PID controller.

(a)

(b)

(c)

Fig. 5. Membership Function for FGSPI and FGSPID Controller of (a) ACE, (b) ∆ACE and (c) output

JAETM, VOLUME I, ISSUE I, JUNE 2021 ISSN (PRINT): 2707-4978 & (ONLINE): 2789-0848 51
 Journal of Applied Engineering, Technology and Management (JAETM)

IV. SIMULATED RESULT AND DISCUSSION

The load frequency control of the two area network was carried out using different controllers. The step
load disturbance of 0.1 p.u. was applied in area-1 and the deviations in frequency and tie-line power flows
were investigated. Conventional controllers like PI and PID controllers are used. The conventional
controllers are compared with the proposed hybrid-type controllers like FGSPI and FGSPID controllers.
The simulation of the two-area model with different controllers was carried out using the simulation toolbox
in MATLAB. The simulated results were compared on the peak overshoot and the settling time of the output

frequency.
Fig. 6. Simulated result with PI controller

Fig. 7. Simulated result with PID controller

JAETM, VOLUME I, ISSUE I, JUNE 2021 ISSN (PRINT): 2707-4978 & (ONLINE): 2789-0848 52
 Journal of Applied Engineering, Technology and Management (JAETM)

Fig. 8. Simulated result with FGSPI controller

Fig. 9. Simulated result with FGSPID controller

The quantitative comparative analysis of the results with respect to frequency and tie-line power deviations
for different controllers is given in Table III. The performance indices used are peak overshoot and settling
time. As it is clear from Table III, that the response with FGSPID controller is the best among all the
controllers because the values of peak overshoot and the settling time are minimum in the case of FGSPID
controller.
TABLE 3. COMPARATIVE ANALYSIS OF RESULTS
Controllers Areas Peak Overshoot (p.u) Settling Time (Sec)
Area-1 0.58 115
PI Controller Area-2 0.35 100
Tie-lie power 0.15 30

JAETM, VOLUME I, ISSUE I, JUNE 2021 ISSN (PRINT): 2707-4978 & (ONLINE): 2789-0848 53
 Journal of Applied Engineering, Technology and Management (JAETM)

Area-1 0.40 110

PID Controller Area-2 0.26 40
Tie-lie power 0.12 20
Area-1 0.08 18
FGSPI Controller Area-2 0.10 28
Tie-lie power 0.09 12
Area-1 0.07 10
FGSPID Controller Area-2 0.09 12
Tie-lie power 0.08 11.5

V. CONCLUSION
In this research, the conventional controllers (PI and PID) and propose hybrid types controllers (FGSPI and
FGSPID) approach are employed for load frequency control of an interconnected power system involving
Chhukha Hydropower plant and Rubesa Wind power plant. The system was simulated using the Simulink
toolbox in MATLAB. The proposed hybrid types of controllers like FGSPI and FGSPID are reported as
with better performance (dynamic response improvement) in comparison to conventional controllers like
PI and PID controllers. The system response is compared in terms of the peak overshoot (p.u) value and the
settling time (sec). The proposed controllers are not only simple in design but also easy to implement.
Moreover, the online adaptation of supplementary controller gain makes the proposed controllers more
effective and it is expected that the controller will perform effectively under different operating conditions.
Simulation results obtained demonstrate the usefulness of the proposed controllers. Taking PI and FGSPID
controllers into consideration the peak overshoot in Area-1, Area-2 and tie-line power was reduced by 87%,
74% and 46% respectively. Similarly, the settling time in Area-1, Area-2 and tie-line power was reduced
by 91.3%, 88% and 61.6% respectively. From the research that has been carried out it is possible to conclude
that the proposed hybrid type FGSPID controller had a faster response to the system feedback error and it
can be implemented in the system where a fast response is required.

VI. APPENDIX
Parameter Description Value Unit

Tg Time constant of governor of Area-1 5 Second

TR Reset time of governor of Area-1 10 Second

Tw Water starting time of Hydro turbine 15.8 Second

in Area-1

H1 Inertia constant of generator of Area-1 3.04 MW.sec/MVA

D1 Load damping constant Area-1 1

JAETM, VOLUME I, ISSUE I, JUNE 2021 ISSN (PRINT): 2707-4978 & (ONLINE): 2789-0848 54
 Journal of Applied Engineering, Technology and Management (JAETM)

R Droop Characteristic of Area-1 10 Hz/pu MW

β1 Frequency bias factor of Area-1 1.1 pu MW/Hz

ρ Air density of Area-2 1.20735 Kg/m3

V Wind velocity of Area-2 Variable m/s

Tt Wind Turbine time constant 7.5 sec

Cp Power coefficient of Area-2 0.53

λ Tip speed ratio 5.79

θ Pitch Angle 35.5 Degree

β2 Frequency bias factor of Area-2 0.56 pu MW/Hz

G Gearbox Ratio 1:120

H2 Inertia constant of wind generator of Area-2 3.79 MW.sec/MVA

D2 Load Damping constant of Area-2 1

r Blade length 16.5 meter

T Synchronizing Coefficient for Tie Line for Two Area 0.08 MW/radian
Systems

REFERENCES
[1] Sathans and A. Swarup, “Intelligent load frequency control of two-area interconnected power system and comparative
analysis,” Proc. - 2011 Int. Conf. Commun. Syst. Netw. Technol. CSNT 2011, no. December, pp. 360–365, 2011, doi:
10.1109/CSNT.2011.81.
[2] M. R. Tur, M. Wadi, A. Shobole, and S. Ay, “Load Frequency Control of Two Area Interconnected Power System Using
Fuzzy Logic Control and PID Controller,” 7th Int. IEEE Conf. Renew. Energy Res. Appl. ICRERA 2018, vol. 5, pp. 1253–
1258, 2018, doi: 10.1109/ICRERA.2018.8566890.
[3] D. Sambariya and V. Nath, “Load Frequency Control Using Fuzzy Logic Based Controller for Multi-area Power System,”
Br. J. Math. Comput. Sci., vol. 13, no. 5, pp. 1–19, 2016, doi: 10.9734/bjmcs/2016/22899.
[4] A. Sreenath, Y. R. Atre, and D. R. Patil, “Two area load frequency control with fuzzy gain scheduling of PI controller,”
Proc. - 1st Int. Conf. Emerg. Trends Eng. Technol. ICETET 2008, pp. 899–904, 2008, doi: 10.1109/ICETET.2008.255.
[5] B. A. Reddy, Y. Srinivas, and E. Venkata Ramesh, “Analytical structures of gain-scheduled fuzzy PI controllers,” 2010 Int.
Conf. Ind. Electron. Control Robot. IECR 2010, pp. 122–128, 2010, doi: 10.1109/IECR.2010.5720138.
[6] O. R. Hydraulic, S. E. T. P. Adjustmnets, F. B. Transmitter, S. F. Pt, and G. Or, “CHAPTER-6 HYDRO-TURBINE
GOVERNING SYSTEM ( Reviewed by Dr . R . Thapar ),” vol. II, pp. 147–173.
[7] N. K. Kumar and I. E. S. Naidu, “Load Frequency Control for A Multi Area Power System Involving Wind , Hydro and
Thermal Plants,” vol. 3, no. 1, pp. 1008–1013, 2014.
[8] Woodward, “Speed Droop and Power Generation,” Power, vol. 01302, 1991.
[9] O. Beltran, R. Peña, J. Segundo, A. Esparza, E. Muljadi, and D. Wenzhong, “Inertia estimation of wind power plants based
on the swing equation and Phasor Measurement Units,” Appl. Sci., vol. 8, no. 12, 2018, doi: 10.3390/app8122413.
[10] A. K. Sahoo, V. Chitra, and T. Kanagapriya, “Load Frequency Control for a Distributed Grid System Involving Wind and
Hydro Power Generation,” 2012.

JAETM, VOLUME I, ISSUE I, JUNE 2021 ISSN (PRINT): 2707-4978 & (ONLINE): 2789-0848 55
 Journal of Applied Engineering, Technology and Management (JAETM)

[11] P. S. Kundur, “(P.Kundur-1)Power System Stability and Control Volume I,” McGraw-Hill, Inc. pp. 1–600, 2004.
[12] N. Nguyen and J. Mitra, “Effect of wind power on load frequency control,” IEEE Power Energy Soc. Gen. Meet., vol. 2016-
Novem, 2016, doi: 10.1109/PESGM.2016.7741814.
[13] A. J. Mahdi, W. H. Tang, and Q. H. Wu, “Derivation of a complete transfer function for a wind turbine generator system by
experiments,” PEAM 2011 - Proc. 2011 IEEE Power Eng. Autom. Conf., vol. 1, no. September, pp. 35–38, 2011, doi:
10.1109/PEAM.2011.6134789.
[14] P. P. Komarnicki, “Topic : Wind Power - Modeling • Control System Modeling Maximizing of Energy Yield,” 2019.
[15] A. Mohanty, M. Viswavandya, D. Mishra, and S. Sthitapragyan, “Modeling and optimization of remote solar - Wind hybrid
station for stability & reactive power management,” Proc. - 2015 Int. Conf. Man Mach. Interfacing, MAMI 2015, pp. 1–6,
2016, doi: 10.1109/MAMI.2015.7456590.
[16] A. Kalmikov, “Wind Power Fundamentals,” Wind Energy Eng. A Handb. Onshore Offshore Wind Turbines, pp. 17–24,
2017, doi: 10.1016/B978-0-12-809451-8.00002-3.
[17] Y. Zhang, X. Liu, and Y. Yan, “Model predictive control for load frequency control with wind turbines,” J. Control Sci.
Eng., vol. 2015, 2015, doi: 10.1155/2015/282740.
[18] G. Singh and S. K. Sinha, “Frequency and tie-line power control of two area network with multi sources of generators,”
India Int. Conf. Power Electron. IICPE, vol. 2016-Novem, 2016, doi: 10.1109/IICPE.2016.8079333.

[19] Y. Mi et al., “Sliding mode load frequency control for multi-area time-delay power system with wind power integration,”
IET Gener. Transm. Distrib., vol. 11, no. 18, pp. 4644–4653, 2017.
[20] S. K. Vavilala, R. S. Srinivas, and M. Suman, “Load Frequency Control of Two Area Interconnected Power System Using
Conventional and Intelligent Controllers,” vol. 4, no. 1, pp. 156–160, 2014.
[21] K. Bedoud, M. Ali-Rachedi, T. Bahi, and R. Lakel, “Adaptive Fuzzy Gain Scheduling of PI Controller for control of the
Wind Energy Conversion Systems,” Energy Procedia, vol. 74, pp. 211–225, 2015, doi: 10.1016/j.egypro.2015.07.580.
[22] P. Gopi and G. Suman, “A new approach for Tuning of PID Load Frequency Controller of an Interconnected Power System,”
Int. J. Mod. Trends Eng. Res., 2015.
[23] R. R. Bajya and M. R. Taparia, “Modeling of a nonlinear hydro power plant and analysis with PID controllers,” ICCCCM
2016 - 2nd IEEE Int. Conf. Control Comput. Commun. Mater., no. Iccccm, pp. 16–18, 2017, doi:
10.1109/ICCCCM.2016.7918262.
[24] S. R. Krishna, P. Singh, and T. I. T. Excellence, “International Journal of Current Trends in Engineering & Technology
Optimal Design & Analysis of Load Frequency Control for two Interconnected Area , Using PSO as Smart Tool to Tune
PID,” vol. 02, pp. 2–5, 2016.
[25] A. T. Hammid, M. Hojabri, M. H. Bin Sulaiman, A. N. Abdalla, and A. A. Kadhim, “Load Frequency Control for
hydropower plants using PID controller,” J. Telecommun. Electron. Comput. Eng., vol. 8, no. 10, pp. 47–51, 2016.
[26] A. I. Dounis, P. Kofinas, C. Alafodimos, and D. Tseles, “Adaptive fuzzy gain scheduling PID controller for maximum power
point tracking of photovoltaic system,” Renew. Energy, vol. 60, pp. 202–214, 2013, doi: 10.1016/j.renene.2013.04.014.
[27] M. R. Nayeri, S. Khosravi, and M. Aliyari Sh., “Design and implementation double input fuzzy PI and fuzzy gain scheduling
PI controller on hydraulic plant,” 13th Iran. Conf. Fuzzy Syst. IFSC 2013, 2013, doi: 10.1109/IFSC.2013.6675648.
[28] H. D. Mathur and H. V. Manjunath, “Extended fuzzy logic based integral controller for three area power system with
generation rate constraint,” Proc. IEEE Int. Conf. Ind. Technol., pp. 917–921, 2006, doi: 10.1109/ICIT.2006.372279.
[29] K. K. Swamy and K. C. Sekhar, “Comparative Analysis of Different Controllers in Two – Area Hydrothermal Power
System,” pp. 493–501, 2017.
[30] F. U. Syed, M. L. Kuang, M. Smith, S. Okubo, and H. Ying, “Fuzzy gain-scheduling proportional-integral control for
improving engine power and speed behavior in a hybrid electric vehicle,” IEEE Trans. Veh. Technol., vol. 58, no. 1, pp. 69–
84, 2009, doi: 10.1109/TVT.2008.923690.
[31] M. S. Anower, M. R. I. Sheikh, M. F. Hossain, M. G. Rabbani, and A. B. M. Nasiruzzaman, “Fuzzy gain scheduling of an
AGC in a single area power system,” Proc. 4th Int. Conf. Electr. Comput. Eng. ICECE 2006, no. December, pp. 9–12, 2007,
doi: 10.1109/ICECE.2006.355277.
[32] S. Baghya Shree and N. Kamaraj, “Hybrid Neuro Fuzzy approach for automatic generation control in restructured power
system,” Int. J. Electr. Power Energy Syst., vol. 74, no. 3, pp. 274–285, 2016, doi: 10.1016/j.ijepes.2015.05.029.

JAETM, VOLUME I, ISSUE I, JUNE 2021 ISSN (PRINT): 2707-4978 & (ONLINE): 2789-0848 56