IEEE Systems, Man and Cybernetics Magazine - July 2021 - 21

V
ariations in load demands, changes in system
parameters, and the continuous integration of
renewable energy sources leave scopes of
improvements in the performance of a load frequency
controller (LFC). To address such issues,
a gravitational search algorithm (GSA)-based LFC is proposed
to incorporate dynamic state variations in multiarea
power systems. It helps to minimize the frequency deviations
in individual areas and power deviations in the tie lines
maintaining the system stability. The proposed method is
validated on a two-area photovoltaic (PV) integrated thermal
power system.
In this article, different perturbed system parameters
are considered to validate performances of the controller.
Furthermore, the sturdiness of the proposed controller is
tested under time-varying load disturbances considering
nominal and perturbed system parameters. The method is
extended to four-area power systems and two-area thermal
power systems with governor deadband (GDB) nonlinearity
to consider the increased complexity of the systems
and validate the competence of the proposed controller.
The results show the effectiveness of the proposed controller
under nominal operations, parametric uncertainties
in the system, and variable load disturbances. Comparative
results with a few existing control techniques [e.g., proportional
integral (PI), proportional-integral-derivative (PID),
fuzzy-based PID, and PI-PID] and optimization algorithms
(e.g., variants of the firefly algorithm [FA]; genetic algorithm
[GA]; and ant colony optimization [ACO], particle swarm
optimization [PSO], and bacterial foraging optimization
[BFO] algorithms) are presented to demonstrate improved
performances of those systems using the proposed method.
The LFC
Maintaining reliable and uninterrupted power supplies to
consumers is a key objective of power suppliers. To
achieve this, the power system stability, operation, and
control should be robust and smooth. However, diverse
load requirements, continuous integrations of renewable
energy sources, and complex operations are the modernday
challenges for power system networks. Two control
strategies are generally used to maintain stable operations
of power systems: 1) reactive power balance or
automatic voltage regulator and 2) active power balance,
automatic generation control, or automatic load frequency
control [1]. This work is focused on the LFC.
In a balanced condition, a power system operates at a
nominal frequency, e.g., 50 or 60 Hz. A mismatch
between power generation and demand causes the frequency
to fluctuate from the nominal value. If the generated
power is less than the power demand, then the
frequency of the power system drops, and vice versa. A
sudden change in the load demand or frequency may
Digital Object Identifier 10.1109/MSMC.2021.3066149
Date of current version: 15 July 2021
damage turbine blades [2] and cause oscillations in
power systems, which can propagate through the connected
networks and affect the functioning of components,
such as ac motors [3] and transformers [4]. These
improper operations and fluctuations in frequency may
lead to the complete collapse of the power systems as
blackouts. Some cases of blackouts due to frequency
variations were already reported in the North Eastern
grid in India [5] as well as in Brazil, Canada, Italy, and
the United States [6].
The aim of the LFC is to maintain the steady-state error of
the system frequency to zero following a load disturbance.
The choice of proper controller values for the LFC is challenging,
as it refers to a multivariable, constrained control
problem due to a huge number of generators, the integration
of nonconventional energy sources, physical limitations of
different system parameters, and uncertain load demands.
Related Works
Over the years, several research studies have been done
for the LFC. In this article, a brief literature survey is presented
based on the three categories in Table 1: the controllers,
optimization techniques, and objective functions
used. The details of the literature are available in " Additional
Information. "
It is observed from Table 1 that, for the LFC problem,
the most popular controllers in the literature are the PI
and PID controllers, perhaps for their simplicity and
ease of use. In the optimization category, metaheuristic
algorithms seem to be more popular than classical techniques,
as metaheuristic algorithms use limited information
from the search space. In some problems, these
algorithms are able to produce suboptimal solutions
where classical techniques fail to perform. Different
time-domain objectives are used by researchers for the
optimization process. Among all of the objective functions
listed in Table 1, integral time absolute error (ITAE)
and integral squared error (ISE) are used extensively.
Motivation and Contribution
As discussed earlier, a sudden load disturbance affects not
only the nominal frequency but also the performance of
key components of the power system (e.g., the generator,
turbine, and so on). The motivation of this work is to
extract information of the states of the key components of
the system. The main contributions of this work can be
summarized as follows:
◆ A framework for designing an LFC incorporating information
of the dynamic variations of the system states is
proposed. A hybrid two-area LFC system is considered
where a PV module in one area is connected with a conventional
power system in another area through a tie line.
◆ To improve the performance of the system, the proposed
controller is integrated with a metaheuristic
optimization framework. The paradigm of the GSA
is followed.
July 2021 IEEE SYSTEMS, MAN, & CYBERNETICS MAGAZINE 21
IEEE Systems, Man and Cybernetics Magazine - July 2021

Table of Contents for the Digital Edition of IEEE Systems, Man and Cybernetics Magazine - July 2021
