Volume 8, Issue 5, May – 2023 International Journal of Innovative Science and Research Technology

ISSN No:-2456-2165

Optimization of Load Frequency Control of Different

Area System Based on Genetic Algorithm
Ashirwad Sharma1 Aman Singh2
Dept. of EEE Dept. of EEE
Galgotias College of Engineering and Technology Galgotias College of Engineering and Technology
Greater Noida, India Greater Noida, India

Anshuman Raj3
Dept. of EEE
Galgotias College of Engineering and Technology
Greater Noida, India

Abstract:- Load frequency regulation (LFC) is an of power systems and adjust control settings for greater
essential part of a power system's operation in order to performance [1].
maintain the balance between generation and load
demand. This article proposes two unique models for a GA is a type of optimisation algorithm that is impacted
single area and two area power system that uses a by natural selection. GAs are particularly well-suited to
Genetic Algorithm (GA) and a Proportional-integral- dealing with challenging optimisation problems, such as the
derivative (PID) controller. A comparison between the LFC problem in power systems. The main concept behind
proposed GA-PID controller and the conventional PID using GAs for LFC is to use the GA to find the best
controller is made. The simulation results show that the controller settings that minimize frequency deviation. In the
GA-PID controller significantly improves LFC GA, the controller parameters may be represented as a
performance by reducing overshoots and settling time chromosome, and the fitness function can be expressed as
while maintaining frequency variation within acceptable the frequency deviation. The GA repeatedly generates a
boundaries. The recommended approach is further population of potential solutions and evolves the population
tested under various operational conditions and using selection, crossover, and mutation operators until the
disturbances, demonstrating its robustness and optimal solution is determined [2-3].
effectiveness in maximizing LFC. The results
demonstrate that the GA-PID controller outperforms the II. LITERATURE REVIEW
conventional PID controller and can maintain the
stability and reliability of the power system. Overall, the An overview of the most current studies on GA-based
proposed GA-PID controller provides a useful and controllers for LFC is the objective of this literature review.
efficient way to optimize the LFC of a power system. The The authors of [4] use local optimisation and a particle
findings of this study can be used to develop LFC swarm technique to examine the optimum of semi-active
strategies that are more advanced and improve the suspension lqr parameters. Wang used the generalized
power system's stability and reliability. gradient technique to improve the air springs. People
provide a multi-objective optimisation technique for
Keywords:- Genetic Algorithm, Load Frequency Control, information control timing that employs a genetic algorithm
PID Controller. to solve it and takes into consideration components such as
delay, passing capacity, and parking frequency to design a
I. INTRODUCTION model [5]. The authors of [6] proposed a distributed and
coordinated voltage and frequency (V-f) control system for
The LFC is a vital part of operating a connected power island microgrids that takes into consideration demand that
system because it maintains the balance between the supply is dependent on V-f and the restricted capacity of distributed
and demand of electricity. LFC requires a balance between energy resources (DERs). The Non-dominated Sorting
power generation and load demand in an interconnected Genetic approach II (NSGA-II) methodology and costs are
power system. The power output of the generators may be used to evaluate communication system latency, ultra-
changed to control the frequency deviation of the LFC capacitor size, and secondary controller parameters [7]. For
system. In order to regulate frequency fluctuation within the LFC applications, a supervisory adaptive model predictive
constraints of the power system, the LFC problem may be control (AMPC) technique based on extremely low rating
presented as an optimization problem. Researchers have SMES is presented [8]. An optimisation approach based on
examined the use of optimisation techniques such as GA to the CS algorithm is created and used in a nonlinear magnetic
design controllers for LFC. The widespread use of GA- levitation system to adjust all FLC parameters, including the
based controllers for LFC has sparked a lot of attention due RB [9]. An islanded microgrid (MG) employs sliding mode
to their ability to manage the nonlinear and dynamic nature control (SMC), also known as fractional order SMC

IJISRT23MAY318 www.ijisrt.com 80
Volume 8, Issue 5, May – 2023 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
(FOSMC), for load frequency control (LFC), which is based area power system with high penetration of wind energy.
on the fractional calculus (FC) concept [10]. The problem of spectral PRB allocation and spatial stacking
in a heterogeneous network is still unresolved. The authors
The major objective of [11] was to develop a control of [16] look at a power consumption model for a multi-cell
strategy for HVAC systems that incorporated fan coil unit multi-user massive MIMO 5G network to account for the
(FCU) temperature management for energy savings in combined effects of both dimensions.
chilled water systems in order to increase the efficiency of
HVAC system operation. The monitoring platform's Hence there is an urgent need for optimizing the load
improved multi-population genetic algorithm (IMPGA), frequency control, and genetic algorithms are one of the
which has many goals, is used to determine which switching most promising techniques for this.
technique is the most efficient [12]. The authors of [13]
examine power quality issues at an offshore wind-powered  Following are the major contributions of this work:
oil and gas facility using an island configuration. In [14], the
authors present a self-contained, interconnected hybrid  Proposed a novel electrical model
microgrid system that uses a sea wave generator, a biodiesel  Analyzed the electrical model in terms of frequency
generator, and a battery energy storage system in microgrid deviation with and without PID controller
1, and an organic Rankine cycle-based solar thermal power  Analyzed the effect of genetic algorithms on load
plant, a biodiesel generator, and superconducting magnetic frequency control.
energy storage in microgrid 2. [15] illustrates a practical
control method for improving frequency stability in a three-

III. PROPOSED SYSTEM MODEL

The proposed system model is shown in Fig. 1 and Fig. 2. Fig. 1 presents a single area power system, on the other hand Fig. 2
presents two area power systems. Further the LFC of this model is analyzed after incorporating PID controller and GA.

Fig 1 Single Area Power System

IJISRT23MAY318 www.ijisrt.com 81
Volume 8, Issue 5, May – 2023 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
A. Genetic Algorithm
Genetic algorithms are becoming a key technique for resolving optimization problems in electrical engineering. They have
the ability to address complex optimisation problems that are challenging for traditional optimisation techniques. The many uses
of genetic algorithms in electrical engineering include power systems, control systems, and circuit design. By accurately
simulating the processes of natural selection and genetic inheritance, genetic algorithms can be used to optimize the parameters in
electrical systems to provide higher performance and cheaper costs.

Fig 2 Two Area-Interconnected Power System

Before using a genetic algorithm, it is necessary to

define the problem and the objective function. The fitness
function is defined next, and it is then used to evaluate each
solution's performance. The genetic algorithm then carries
on with selection, crossover, and mutation in order to
produce a new population of answers. The best solutions are
selected during the selection process from the current
population based on their fitness scores. The crossover
approach involves selecting two solutions at random from
the population, exchanging a portion of their genetic
makeup to create two new offspring, and then repeating the
process twice. The process of mutation results in arbitrary
changes to the chromosomes of the progeny. The steps of
selection, crossover, and mutation are carried out repeatedly
until an appropriate response is found. Each stage of the GA
is presented in Fig. 3, and explained below.

Prior to using a genetic algorithm, the task and

objective of the function must be determined. This requires
choosing which variables will be optimized as well as
outlining the conditions that must be satisfied. After the
problem is specified, the evolutionary algorithm produces an
initial population of solutions, each represented by a series
of binary digits. The fitness function, which is discussed
next, is then used to evaluate each solution's performance.
Fig 3 Flowchart of Genetic Algorithm The fitness function is frequently expressed as a function of

IJISRT23MAY318 www.ijisrt.com 82
Volume 8, Issue 5, May – 2023 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
the objective function, which evaluates the efficacy of each
solution. The fitness function assigns a fitness value to each
solution based on how well it operates, and the solutions
with higher fitness values are selected for further
optimisation.

The genetic algorithm then carries on with selection,

crossover, and mutation in order to produce a new
population of answers. Throughout the selection process, the
best candidates are selected from the current population
based on their levels of fitness. The crossover process
involves randomly selecting two solutions from the
population and exchanging a portion of their genetic
material to create two new offspring. The mutation process
involves randomly changing some of the bits in the
chromosomes of the offspring. The process of selection,
crossover, and mutation is repeated until a satisfactory Fig 4 Frequency deviation subjected to load change without
solution is found. In electrical engineering applications, the PID controller
genetic algorithm can be used to optimize parameters such
as circuit design, power system operation, and control
system design. For example, in circuit design, the genetic
algorithm can be used to optimize the values of resistors,
capacitors, and other components to minimize power
consumption or improve performance.

B. PID Controller
Proportional-integral-derivative (PID) controller is
widely used in electrical applications, a well-liked feedback
control mechanism, to regulate and control a variety of
systems, including power systems, motor speed control,
temperature management, and many others. The PID
controller algorithm generates an error signal by comparing
the desired set point value with the actual output value, and
then it sends a control signal in an effort to lessen the error
signal.

The PID controller computes the control signal using Fig 5 Frequency deviation subjected to load change with
proportional, integral, and derivative control actions. The PID controller
proportional action, which is proportionate to the error
signal, is used to correct the error in real-time. The integral
action, which is proportionate to the error over time, is used
to correct steady-state errors. The derivative action is
proportionate to the pace at which the error is evolving and
forecasts how the system will behave in the future.

IV. RESULT & DISCUSSION

The performance of the system model proposed in Fig.

1 is analyzed under various conditions, in terms of
frequency deviation Vs time plot. The frequency is taken in
hertz (Hz) and time in seconds.

Fig. 6 Effect of GA on Frequency deviation subjected to

load change with PID controller

IJISRT23MAY318 www.ijisrt.com 83
Volume 8, Issue 5, May – 2023 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
REFERENCES

[1] Ram Babu, N., Bhagat, S. K., Saikia, L. C.,

Chiranjeevi, T., Devarapalli, R., & García Márquez,
F. P. (2023). A comprehensive review of recent
strategies on automatic generation control/load
frequency control in power systems. Archives of
Computational Methods in Engineering, 30(1), 543-
572.
[2] Mac, T. T., Iba, D., Matsushita, Y., Mukai, S., Inoue,
T., Fukushima, A., ... & Moriwaki, I. (2022).
Application of genetic algorithms for parameters
identification in a developing smart gear system.
Forschung im Ingenieurwesen, 86(3), 433-450.
[3] Zitar, R. A. (2022, November). A Review of the
Genetic Algorithm and JAYA Algorithm
Fig 7 Comparative analysis of Frequency deviation Applications. In 2022 15th International Congress on
subjected to load change with and without GA for Image and Signal Processing, BioMedical
one area network Engineering and Informatics (CISP-BMEI) (pp. 1-7).
IEEE.
Fig. 4 and Fig. 5 illustrate the frequency deviation [4] Orka, N. A., Muhaimin, S. S., Shahi, M. N. S., &
graph of the proposed model with and without PID Ahmed, A. (2022). An Enhanced Gradient Based
controller. Severe inconsistency and noise has been Optimized Controller for Load Frequency Control of
observed in models without PID controllers, however after a Two Area Automatic Generation Control System.
using the PID controller a significant change is observed in In Engineering Applications of Modern
the frequency output. But still, this output is not up to the Metaheuristics (pp. 79-107). Cham: Springer
mark and cannot be utilized in any physical application due International Publishing.
to high risk of system failure and energy loss. Thus, in Fig. 6 [5] Yuan, C. Y., Li, K. T., Zang, G. R., & Wang, X. C.
GA is utilized. GA is a popular technique used to improve (2022, September). Optimization of semi-active
the system performance by utilizing the basic biological suspension LQR parameters based on local
concepts. After incorporating GA and PID controllers into optimization with a skipping out particle swarm
the proposed system model, a straight line is obtained which algorithm. In International Conference on Mechanical
shows very less or no noise in the frequency output. Design and Simulation (MDS 2022) (Vol. 12261, pp.
1111-1116). SPIE.
Further, Fig. 7 and Fig. 8 presents the comparative [6] She, B., Li, F., Cui, H., Wang, J., Min, L., Oboreh-
study between different models. In Fig. 7 the model Snapps, O., & Bo, R. (2023). Decentralized and
frequency output with PID controller and GA is compared. coordinated vf control for islanded microgrids
It is very clear that utilizing GA and PID together considering der inadequacy and demand control.
completely outperformed the output of PID alone. IEEE Transactions on Energy Conversion.
[7] Alizadeh, G. A., Rahimi, T., Babayi Nozadian, M. H.,
V. CONCLUSION Padmanaban, S., & Leonowicz, Z. (2019). Improving
microgrid frequency regulation based on the virtual
In this paper the load frequency controller of a single inertia concept while considering communication
area and two area power system is analyzed. Because of system delay. Energies, 12(10), 2016.
high power requirements and with growing need of energy, [8] Mir, A. S., & Senroy, N. (2017). Adaptive model
it is of significant importance to optimize the load frequency predictive control scheme for application of SMES
control of the electrical system. Novel power models for one for load frequency control. IEEE Transactions on
area and two area power systems are produced in this work, Power Systems.
and its performance is analyzed in terms of frequency [9] García-Gutiérrez, G., Arcos-Aviles, D., Carrera, E.
deviation response. The proposed models are analyzed with V., Guinjoan, F., Motoasca, E., Ayala, P., & Ibarra,
different combinations including individual model, model A. (2019). Fuzzy logic controller parameter
with PID controller, and model with PID controller and optimization using metaheuristic cuckoo search
genetic algorithm. It is observed that response with genetic algorithm for a magnetic levitation system. Applied
algorithms showed most promising results, and it is Sciences, 9(12), 2458.
appropriate for industry. All the simulations are performed [10] Esfahani, Z., Roohi, M., Gheisarnejad, M.,
on Matlab and Simulink. Moreover, in future the proposed Dragičević, T., & Khooban, M. H. (2019). Optimal
model can be analyzed with Thyristor Controlled Phase non-integer sliding mode control for frequency
Shifter and other controllers. regulation in stand-alone modern power grids.
Applied Sciences, 9(16), 3411.

IJISRT23MAY318 www.ijisrt.com 84
Volume 8, Issue 5, May – 2023 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
[11] Lin, C. M., Liu, H. Y., Tseng, K. Y., & Lin, S. F.
(2019). Heating, ventilation, and air conditioning
system optimization control strategy involving fan
coil unit temperature control. Applied Sciences,
9(11), 2391.
[12] Bao, G., & Ke, S. (2019). Load transfer device for
solving a three-phase unbalance problem under a
low-voltage distribution network. Energies, 12(15),
2842.
[13] Alves, E., Sanchez, S., Brandao, D., & Tedeschi, E.
(2019). Smart load management with energy storage
for power quality enhancement in wind-powered oil
and gas applications. Energies, 12(15), 2985.
[14] Bhuyan, M., Barik, A. K., & Das, D. C. (2020). GOA
optimised frequency control of solar-thermal/sea-
wave/biodiesel generator based interconnected hybrid
microgrids with DC link. International Journal of
Sustainable Energy, 39(7), 615-633.
[15] Khamies, M., Magdy, G., Hussein, M. E., Banakhr, F.
A., & Kamel, S. (2020). An efficient control strategy
for enhancing frequency stability of multi-area power
system considering high wind energy penetration.
IEEE Access, 8, 140062-140078.
[16] Marwaha, S., Jorswieck, E. A., Jassim, M., Kuerner,
T., Perez, D. L., Geng, X., & Bao, H. (2023). Energy
Efficient Operation of Adaptive Massive MIMO 5G
HetNets. arXiv preprint arXiv:2302.08223.

IJISRT23MAY318 www.ijisrt.com 85