An optimal design for an automatic voltage regulation system using a multivariable PID controller based on hybrid simulated annealing – white shark optimization Authors: Ahmed K. Ali Published: 2024 ABSTRACT This paper combines the simulated annealing (SA) and white shark optimisation (WSO) algorithms to create a novel hybrid metaheuristic algorithm named SA–WSO. The proposed algorithm finds the best approach to tuning the parameters of three types of regulators. Three types of proportional–integral–derivative (PID) controllers are utilized for an automatic voltage regulator (AVR) system, including fractional–order (FOPID) –controllers. A novel filtered fractional–order PID controller is also presented to enhance the response of the AVR system. The simulation results demonstrate the highly effective optimisation of three controllers, including the rarely used filtrated–FOPID controller, outperforming other controller approaches in the state of the art. The controller performance analyses are also conducted to demonstrate the superiority of SA–WSO-optimized for tuning the controller parameters in different conditions for disturbance rejection and robustness. A comprehensive comparative analysis is performed to determine the most appropriate controller for the AVR system, and it is found that hybrid SA–WSO algorithm provides significant advantages in terms of improved convergence speed. Introduction Power-generating units always strive to minimise power losses to improve overall efficiency and performance, a significant issue in control engineering. An effective approach to mitigating these losses is implementation of an optimal design of automatic voltage regulation (AVR) systems. The controller in AVR systems is critical for minimizing power losses in the generator section. These systems play a crucial role in maintaining the desired voltage levels, thereby optimizing power generation1–3. AVR systems have emerged as a crucial component in power generation units, offering effective voltage regulation and control. The core task of an AVR system is to continuously maintain the output voltages of the generator at the standard level of voltage3,4. A closed-loop configuration of an AVR system consists of five components: regulator, amplifier, exciter, generator, and sensor . In an AVR system, sensors measure voltage across the generator terminal, and amplifiers amplify control signals after regulator adjustment5. In the same context, the exciter continuously controls the output voltage from the generator terminals to improve power generation efficiency, reliability of generator output voltages. This regulation of generator output voltage ensures the desired level, which can improve system performance by reducing power losses in the generation system4. Typically, the AVR, implemented using a microprocessor unit, utilises the error signal and the control law to generate control signal3. Thus, the development of new control techniques to enhance the efficiency of AVR systems remains a critical issue in the field of power systems2. This paper discusses the enhancement of AVR system, focusing on recent developments in control techniques and their effectiveness in improving operation efficiency. Even with the emergence of modern control techniques and control theory, the proportional-integral-derivative (PID) controller remains the most widely employed controller in AVR systems. The control law of a PID controller is a combination of three parameters; proportional gain \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\text{K}}_{\text{P}}$$\end{document}, integral gain \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\text{K}}_{\text{i}}$$\end{document}, and differential gain \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\text{K}}_{\text{d}}$$\end{document}1,5–18. Based to these parameters, the PID control law combines three control strategies. However, some studies consider extending typical PID control by fractional calculus into the integral and derivative actions. The control law of the fractional-order PID (FOPID) regulator incorporates fractional calculus into integral and derivative, as discussed in references3,6–8, and the control law is defined with five parameters9. Therefore, to achieve the best performance for the FOPID controller, five parameters in the control law equation must be tuned. Generally, the execution of fractional calculus is more complicated than the execution of integer-order derivatives and integrals. On the other hand, it provides greater flexibility and has the potential to deliver superior performance8. In previous studies, researchers focused mainly on typical PID controllers with three tuning parameters, which, despite their extensive application, often lack the flexibility necessary for intricate system dynamics. Fractional calculus has led to the development of fractional-order PID (FOPID) controllers, which provide flexibility performance using the fractional integration and differentiation. This paper presents an optimal design of three controllers (PID, FOPID, and F-FOPID) and highlights the importance of tuning the controller to achieve the desired voltage response of the AVR system. In addition, the new optimal design of the filtered-FOPID (FFOPID) regulator is discussed. Usually, the researcher uses metaheuristic algorithms to determine the optimal parameter values of an AVR system controller. Table 1 categorizes a variety of optimization algorithms based on their underlying inspiration and conceptual basis. Thus, it can serve as a comprehensive guide for this study to understand the diverse sources of inspiration for modern optimization algorithms and their applications in tuning different PID controllers. Table 1Summary of the optimal controller parameters obtained in the literature review.Ref.AlgorithmControllerAlgorithm numberController parametersFilter -order \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{k}_{p}$$\end{document} \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{k}_{i}$$\end{document} \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{k}_{d}$$\end{document} µλ N 8 IKAPID11.04261.00930.5999NANANA 1 SOSPID20.56930.40970.1750NANANA 5 ACOPID30.67390.59510.2622NANANA 15 WOAPID40.99610.30610.7847NANANA 6 GAPID60.56000.50000.2000NANANA 7 BBOPID71.24640.58930.4596NANANA 23 GWOPID80.6011.1600.0781NANANA 9 CFAPID90.66920.49590.2657NANANA100.65370.47420.2467NANANA 10 GOAPID111.38251.46080.5462NANANA 11 GSAPID121.43791.22080.7363NANANA 12 SSAPID131.33811.12040.6361NANANA 13 DEPID141.94990.44300.3427NANANA 14 MOLPID150.98770.77800.5014NANANA 19 CSFOPID162.54900.17590.39041.38000.97NA 20 ABCFOPID171.96050.49220.23551.43311.55NA 22 NSGA-IIFOPID180.40800.37400.17731.33360.68NA 3 SCAFOPID191.45090.65670.30761.14421.2145NA 21 SCAFOPID202.97370.90890.53831.34621.15NA Table 1Summary of the optimal controller parameters obtained in the literature review.Ref.AlgorithmControllerAlgorithm numberController parametersFilter -order \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{k}_{p}$$\end{document} \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{k}_{i}$$\end{document} \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{k}_{d}$$\end{document} µλ N 8 IKAPID11.04261.00930.5999NANANA 1 SOSPID20.56930.40970.1750NANANA 5 ACOPID30.67390.59510.2622NANANA 15 WOAPID40.99610.30610.7847NANANA 6 GAPID60.56000.50000.2000NANANA 7 BBOPID71.24640.58930.4596NANANA 23 GWOPID80.6011.1600.0781NANANA 9 CFAPID90.66920.49590.2657NANANA100.65370.47420.2467NANANA 10 GOAPID111.38251.46080.5462NANANA 11 GSAPID121.43791.22080.7363NANANA 12 SSAPID131.33811.12040.6361NANANA 13 DEPID141.94990.44300.3427NANANA 14 MOLPID150.98770.77800.5014NANANA 19 CSFOPID162.54900.17590.39041.38000.97NA 20 ABCFOPID171.96050.49220.23551.43311.55NA 22 NSGA-IIFOPID180.40800.37400.17731.33360.68NA 3 SCAFOPID191.45090.65670.30761.14421.2145NA 21 SCAFOPID202.97370.90890.53831.34621.15NA Summary of the optimal controller parameters obtained in the literature review. The genetic algorithm (GA) are popular choices that simulate the process of natural selection by iteratively generating and evaluating candidate solutions to a problem10. The Particle Swarm Optimization (PSO) another widely used algorithm inspired by the social behaviour of birds flocking or fish schooling11. It involves a set of particles that move through a search space, updating their position and velocity according to their experiences and the experiences of their neighbours. Other implemented algorithms include the improved kidney-inspired algorithm (IKA), which models the function of a kidney and uses a gradient descent method to optimise the objective functions12. In addition, Symbiotic Organisms Search (SOS) is inspired by the mutualistic relationship between organisms, where two or more species benefit from each other’s existence1. Some nature-inspired optimisation algorithms have been developed based on the behaviour of animals, such as Whale Optimization Algorithm (WOA), which mimics the cooperative hunting behaviour of humpbacks13. The brood parasitic behaviour of cuckoo species inspires another algorithm, Cuckoo Search (CS)6. The ant colony optimisation algorithm (ACO) is based on ant colonies’ foraging behaviour, in which ants deposit pheromones to mark the path to a food source5. The artificial bee colony algorithm (ABC) algorithm is based on honeybees’ foraging behaviour7. These algorithms are based on behaviours of humpback whales, cuckoos, ants, honey-bees, and musicians. The input also mentions use of pheromones, waggle dances, and combination of chaotic maps with ant colony optimisation to improve the convergence rate and accuracy of the algorithms. Among other options is the Biogeography-based optimization (BBO) inspired by the geographic distribution of species and their migration behaviours14. The grasshopper optimization algorithm (GOA)15, which stimulates the swarming behaviour of grasshoppers in neutral. A simulated search algorithm known as Teaching Learning-Based Optimisation (TLBO) inspired the process of teaching student in a classroom16. Combining multiple swarms increases the convergence speed of the minimization problem, which can reach optimal values in less time, such as in Many Optimisation Liaisons (MOL), an algorithm that combines multiple swarm intelligence-based algorithms, which is a simplified version of PSO17. In the study presented in18, the gravitational search algorithm (GSA), which simulates the behaviour of gravitational force in a physical system, is introduced. The behaviour of salps, free-floating marine animals that swim by contracting and expanding their bodies, serves as the inspiration for Salp Swarm Algorithm (SSA)19. The differential Evolution (DE) algorithm uses differences between candidate solutions to generate new solutions20. Another search method called the continuous firefly algorithm (CFA) draws inspiration from flashing behaviour of fireflies, where their brightness attracts each other21. Meanwhile, multi-objective extremal optimisation (MOEO) is a method that finds the extreme values of multiple objectives simultaneously8. Recent research in the area of tuning controllers for AVR systems shows a significant tendency towards using advanced optimisation methods22. Ekinci et al.23, developed a hybrid search algorithm with local search capability, specifically for an AVR system and to obtained the optimum parameters of a power System Stabiliser. In24, authors studied a hybrid swarm intelligence-based algorithm for tuning the FOPID controller. In their work, the presented algorithms balanced the exploration and exploitation stages and demonstrated superior performance compared to other benchmark algorithms. Recently, Izci et al.25 proposed a new modification for the design controller of AVR systems using PID and a second-order derivative. They used an adapted Runge Kutta optimizer to fine-tune the controller parameters, thus improving the stability and robustness of the controller for the AVR system design. Another recently study by Izci et al.26 presented an intelligent master/slave tuning system that integrated the Levy flight mechanism into the Runge-Kutta optimizing algorithm, using neural computing approaches to improve controller effectiveness in the design of AVR systems. for design AVR system. This tuning controller parameters approach reflects an increasing tendency toward controller system design using optimization methods integrated with advanced artificial intelligence and machine learning approaches. Ekinci et al.27 proposed a modified opposition-based weighted mean of vector (mOBL-INFO) optimization algorithm to effectively find PID controller parameters for PID controllers with different numbers of tuning variables, thus improving the design of AVR systems. Can et al.28 developed a control design approach to enhance the transient response performance of AVR systems, focusing on the crucial need for rapid AVR system stabilization. The method emphasizes the enhancement of the PID controller with three tuning parameters to acquire faster and more stable AVR system performance. Izci et al. (2024)29 presented a combination of Harris Hawks optimization (HHO) algorithms with the SA optimization algorithm to better tune the PID controller for the AVR system. The HHO-SA optimization utilizes nature-inspired algorithms to fine-tune the parameters of different types of PID controllers. A quadratic wavelet-enhanced gradient-based optimization (QWGBO) algorithm optimized an improved cascaded RPIDD2-FOPI controller by Ekinci et al. (2024)30. This integration enhanced the performance of the controller. This algorithm describes the capability of advanced control systems to improve control performance and efficiency, with additional refinement by the QWGBO algorithm. It appears that the recent literature proposes optimization algorithms that do not adequately address all the shortcomings in the controller design of the AVR systems, especially in terms of computational efficiency, implementation complexity, and performance for disturbance rejection under different conditions. This can raise concerns about their controller performance and adaptability in dynamic conditions. Over the past five years, methods have been reported, each with a distinct optimisation algorithm specifically designed for AVR applications. Together, a limited number of studies have focused on the development and fine-tuning of filtered FOPID controllers. Lately, researchers have also used various evolutionary algorithms to create PID regulators that improve settling time, overshoot, and rise time. The goal of these studies is to make AVR systems respond better to sudden changes20. In addition to some studies using a more complex control law equation by adding multi-term integration and derivatives, these studies aim to further improve the performance of the AVR, as discussed in13,31. This paper aims to introduce a novel F-FOPID controller with a fractional filter for an AVR system to achieve superior system response compared to the approaches in the literature, including PSO11, TLBO16, BBO14, GWO32, and DE20. The main contributions of this paper can be summarised as follows: Design and effectiveness demonstration of the FOPID controller with a fractional filter, called F-FOPID, for an AVR system.Utilisation of hybrid optimisation algorithm for the first time during the design process of a FOPID controller with a fractional filter for an AVR system.A hybrid optimization algorithm that combines simulated annealing (SA) and white shark optimization (WSO) is utilized for the tuning of the PID controller of the AVR system. SA is a single-solution heuristic algorithm that finds the global optimal solution by progressing to a new solution based on iteration. On the other hand, the population-based heuristic known as the WSO is inspired by the survival-of-the-fittest principle to produce new populations. The initial solution set for WSO is generated by SA, which improves the convergence of the algorithm. The proposed SA–WSO algorithm optimizes tuning the PID controller AVR and is assessed based on step response, convergence speed, disturbance management, and resilience. The simulation results clearly indicate that the proposed SA–WSO algorithm outperforms in all performance metrics.This paper includes a comparative analysis to determine the optimum PID controller type for AVR systems. Surprisingly, the filtered FOPID (Fractional Order Proportional-Integral-Derivative) controller, which is not commonly discussed in existing literature, is found to be the most effective option. The filtered–FOPID controller or– called FFOPID controller offers enhanced performance and robustness in voltage regulation, surpassing the commonly used controller types.Three PID controllers based on different control law and five optimisation algorithms (i.e., PSO, TLBO, BBO, GWO, and DE), as well as a new F-FOPID controller, are studied and compared for the design of the AVR system.The first use of a hybrid SA–WSO, to the best of our knowledge, in the design process of the FOPID controller with an integer filter for an AVR system and a demonstration of the superiority of a controller with a fractional filter, i.e., SA–WSO-FOPID, for an AVR system over one classical PID.The first presentation of an SA–WSO-FOPID controller, to the best of our knowledge, for an AVR system and a demonstration of its advantage over the PID controller without a filter (SA–WSO-PID). Design and effectiveness demonstration of the FOPID controller with a fractional filter, called F-FOPID, for an AVR system. Design and effectiveness demonstration of the FOPID controller with a fractional filter, called F-FOPID, for an AVR system. Utilisation of hybrid optimisation algorithm for the first time during the design process of a FOPID controller with a fractional filter for an AVR system. Utilisation of hybrid optimisation algorithm for the first time during the design process of a FOPID controller with a fractional filter for an AVR system. A hybrid optimization algorithm that combines simulated annealing (SA) and white shark optimization (WSO) is utilized for the tuning of the PID controller of the AVR system. SA is a single-solution heuristic algorithm that finds the global optimal solution by progressing to a new solution based on iteration. On the other hand, the population-based heuristic known as the WSO is inspired by the survival-of-the-fittest principle to produce new populations. The initial solution set for WSO is generated by SA, which improves the convergence of the algorithm. The proposed SA–WSO algorithm optimizes tuning the PID controller AVR and is assessed based on step response, convergence speed, disturbance management, and resilience. The simulation results clearly indicate that the proposed SA–WSO algorithm outperforms in all performance metrics. A hybrid optimization algorithm that combines simulated annealing (SA) and white shark optimization (WSO) is utilized for the tuning of the PID controller of the AVR system. SA is a single-solution heuristic algorithm that finds the global optimal solution by progressing to a new solution based on iteration. On the other hand, the population-based heuristic known as the WSO is inspired by the survival-of-the-fittest principle to produce new populations. The initial solution set for WSO is generated by SA, which improves the convergence of the algorithm. The proposed SA–WSO algorithm optimizes tuning the PID controller AVR and is assessed based on step response, convergence speed, disturbance management, and resilience. The simulation results clearly indicate that the proposed SA–WSO algorithm outperforms in all performance metrics. This paper includes a comparative analysis to determine the optimum PID controller type for AVR systems. Surprisingly, the filtered FOPID (Fractional Order Proportional-Integral-Derivative) controller, which is not commonly discussed in existing literature, is found to be the most effective option. The filtered–FOPID controller or– called FFOPID controller offers enhanced performance and robustness in voltage regulation, surpassing the commonly used controller types. This paper includes a comparative analysis to determine the optimum PID controller type for AVR systems. Surprisingly, the filtered FOPID (Fractional Order Proportional-Integral-Derivative) controller, which is not commonly discussed in existing literature, is found to be the most effective option. The filtered–FOPID controller or– called FFOPID controller offers enhanced performance and robustness in voltage regulation, surpassing the commonly used controller types. Three PID controllers based on different control law and five optimisation algorithms (i.e., PSO, TLBO, BBO, GWO, and DE), as well as a new F-FOPID controller, are studied and compared for the design of the AVR system. Three PID controllers based on different control law and five optimisation algorithms (i.e., PSO, TLBO, BBO, GWO, and DE), as well as a new F-FOPID controller, are studied and compared for the design of the AVR system. The first use of a hybrid SA–WSO, to the best of our knowledge, in the design process of the FOPID controller with an integer filter for an AVR system and a demonstration of the superiority of a controller with a fractional filter, i.e., SA–WSO-FOPID, for an AVR system over one classical PID. The first use of a hybrid SA–WSO, to the best of our knowledge, in the design process of the FOPID controller with an integer filter for an AVR system and a demonstration of the superiority of a controller with a fractional filter, i.e., SA–WSO-FOPID, for an AVR system over one classical PID. The first presentation of an SA–WSO-FOPID controller, to the best of our knowledge, for an AVR system and a demonstration of its advantage over the PID controller without a filter (SA–WSO-PID). The first presentation of an SA–WSO-FOPID controller, to the best of our knowledge, for an AVR system and a demonstration of its advantage over the PID controller without a filter (SA–WSO-PID). The paper is structured as follows: Sect. 2 provides an overview of the transfer function configuration of the AVR system, as well as the typical components in the AVR block diagram. Section 3 presents an extensive literature review on the optimisation of various PID controllers. An exhaustive background survey on the state-of-the-art related to the optimum design of PID controllers for AVR systems aims to give a full understanding of optimising techniques for PID controllers. In Sect. 4, the mathematical repetitions of the WSO algorithm are briefly explained. Section 5 presents a hybrid combination of the SA and WSO algorithms, i.e., SA–WSO. The three PID controllers are optimised in Sect. 6. Section 7 provides the final conclusions. Configuration of the proposed AVR with multi-term PID controllers This section explains the configuration design for an AVR system’s proposed multi-term FOPID controller. First, we summarise the FOPID, followed by a demonstration of the AVR system mathematical model, considering three types of PID controllers. FOPID controller The present work investigates an additional effective use of fractional differential equations (FDEs), the development of an optimum controller for the voltage in a typical synchronous generator by combining FDEs with an optimisation approach based on artificial intelligence. This study aims to create the best possible controller for this task. In order to achieve this purpose, the FOPID controller, proposed by Podlubny in 1999, is employed because of its improved tuning flexibility, robustness, and frequency response relative to the conventional PID controller33. The proposed approach highlights the potential utility of FDEs in conjunction with advanced optimisation techniques for developing effective control systems in engineering applications. Overall, FDEs being increasingly used to solve problems in different areas of science shows the usefulness of fractional calculus as a tool in modern scientific research. Equation (1) represents the time-domain representation of the PID controller, while Eq. (3) depicts its FOPID version.1\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{v}_{t}\left(t\right)=k\left({v}_{s}\left(t\right)+\frac{1}{{T}_{i}}{\int\:}_{0}^{t}{v}_{s}\left(\tau\:\right)d\tau\:+{T}_{d}\frac{d{v}_{s}\left(t\right)}{dt}\right)$$\end{document}2\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{v}_{t}\left(t\right)=k\left({v}_{s}\left(t\right)+\frac{1}{{T}_{i}}{{\rm\:Y}}^{-\lambda\:}{v}_{s}\left(t\right)+{T}_{d}{{\rm\:Y}}^{\mu\:}{v}_{s}\left(t\right)\right)$$\end{document} where \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{v}_{t}\left(t\right)$$\end{document} is the PID controller output (V) based on the error between the set point and the measured value \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{v}_{s}\left(t\right)$$\end{document} (V), \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:k$$\end{document} is the gain, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{T}_{i}$$\end{document} is the integral time constant (s), and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{T}_{d}$$\end{document} is the derivative time constant (s). In addition, the FOPID version includes non-integer orders of integration \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\lambda\:}$$\end{document} and differentiation \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\mu\:}$$\end{document} and a fractional operator \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:\text{{\rm\:Y}}$$\end{document}. Unlike typical PID controller which uses three components (proportional, integral, derivative) to provide control based on the current error, past error accumulation, and the rate of change of the error. The FOPID controller extends the PID controller by incorporating fractional calculus, which deals with integrals and derivatives of non-integer orders. In the FOPID controller, the integral term is replaced with a fractional integral of the error, determined by a fractional order \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\lambda\:}$$\end{document}. Accordingly, the FOPID controller delivers more flexibility in control and allows for fine-tuning and stability improvements by adjusting the fractional order3. A typical block diagram of an automatic voltage-regulating system An exciter, amplifier, sensor, and generator are components of the AVR block diagram that combine to create a transfer function. Figure 1 presents the typical block diagram of the AVR system. Each block component has a transfer function that connects the input and output. Fig. 1A typical block diagram of an AVR system. A typical block diagram of an AVR system. Equation (3) to (6) give the transfer functions for the exciter, amplifier, sensor, and generator16. These equations involve a time constant and gain, which characterise the behaviour of the component. By connecting these transfer functions in a feedback loop and implementing appropriate control strategies, the AVR system can effectively correct any deviations in the generator’s terminal voltage. This regulation ensures that the generator maintains a stable voltage level, contributing to the overall stability of the power system. Amplifier transfer function. 3\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{G}_{a}\left(s\right)=\frac{{k}_{a}}{1+{sT}_{a}}$$\end{document} Amplifier transfer function. Amplifier transfer function. Exciter transfer function. 4\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{G}_{e}\left(s\right)=\frac{{k}_{e}}{1+{sT}_{e}}$$\end{document} Exciter transfer function. Exciter transfer function. Generator transfer function. 5\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{G}_{g}\left(s\right)=\frac{{k}_{g}}{1+{sT}_{g}}$$\end{document} Generator transfer function. Generator transfer function. Sensor transfer function. 6\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{G}_{s}\left(s\right)=\frac{{k}_{s}}{1+{sT}_{s}}$$\end{document} Sensor transfer function. Sensor transfer function. The parameters for AVR system include gains \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{k}_{a}$$\end{document}, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{k}_{e}$$\end{document}, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{k}_{g}$$\end{document}, and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{k}_{s}$$\end{document} and time constants \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{T}_{a}$$\end{document}, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{T}_{e}$$\end{document}, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{T}_{g}$$\end{document}, and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{T}_{s}\:$$\end{document}for the amplifier, exciter, generator, and sensor, respectively. These parameters are modelled as linear devices, and the range of possible values for these parameters is presented in Table 2. which, along with Eqs. (3)–(6), is based on previous works in the literature1,5,10,12–15,17–21. Table 2Lower and upper bounds of AVR parameters.Limit of parametersStudied parametersGainTime constant (s)Amplifier \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:10\le\:{k}_{a}\le\:40$$\end{document} \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:0.02\le\:{T}_{a}\le\:0.1$$\end{document} \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{T}_{a}=10,\:{k}_{a}=0.1$$\end{document} Exciter \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:0.9\le\:{k}_{e}\le\:10.0$$\end{document} \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:0.4\le\:{T}_{e}\le\:1.0$$\end{document} \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{T}_{e}=1,\:{k}_{e}=0.4$$\end{document} Generator \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:0.7\le\:{k}_{g}\le\:1.0$$\end{document} \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:1.0\le\:{T}_{g}\le\:2.0$$\end{document} \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{T}_{g}=1\:{k}_{g}=1$$\end{document} Sensor \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:0.9\le\:{k}_{s}\le\:1.1$$\end{document} \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:0.001\le\:{T}_{s}\le\:0.06$$\end{document} \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{T}_{s}=10,\:{k}_{s}=0.01$$\end{document} Table 2Lower and upper bounds of AVR parameters.Limit of parametersStudied parametersGainTime constant (s)Amplifier \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:10\le\:{k}_{a}\le\:40$$\end{document} \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:0.02\le\:{T}_{a}\le\:0.1$$\end{document} \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{T}_{a}=10,\:{k}_{a}=0.1$$\end{document} Exciter \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:0.9\le\:{k}_{e}\le\:10.0$$\end{document} \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:0.4\le\:{T}_{e}\le\:1.0$$\end{document} \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{T}_{e}=1,\:{k}_{e}=0.4$$\end{document} Generator \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:0.7\le\:{k}_{g}\le\:1.0$$\end{document} \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:1.0\le\:{T}_{g}\le\:2.0$$\end{document} \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{T}_{g}=1\:{k}_{g}=1$$\end{document} Sensor \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:0.9\le\:{k}_{s}\le\:1.1$$\end{document} \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:0.001\le\:{T}_{s}\le\:0.06$$\end{document} \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{T}_{s}=10,\:{k}_{s}=0.01$$\end{document} Lower and upper bounds of AVR parameters. The response of an uncontrolled AVR system, as shown in Fig. 2, demonstrates certain performance matters. The empirical data reveals notable steady-state error, extended settling times, and excessive overshoot. Accordingly, it is imperative to enhance both the steady-state accuracy and the transient response characteristics of the AVR system. The crucial aim of an AVR system is to provide a stable output despite disturbances or fluctuations in load. The step response analysis of an uncontrolled AVR system displays considerable deficiencies, including an inability to converge to the desired output, taking a long time to reach a final steady state after a disturbance, and exceeding the desired output level. Fig. 2Uncontrolled step response of the AVR system. Uncontrolled step response of the AVR system. Therefore, to overcome these issues, various control strategies can be implemented for the AVR system, such as a typical PID controller or its more advanced controller, like a fractional-order PID. This paper explores the efficacy of three types of PID controllers. These include the typical PID, the standard typical FOPID, and the FOPID with a fractional filter called the F-FOPID to improve the dynamic response of an AVR system, aiming to stabilize the generator voltage under all operating conditions. The three tuning parameters \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{k}_{p}$$\end{document}, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{k}_{i}$$\end{document}, and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{k}_{d}$$\end{document} make up the transfer function in the s-domain of the conventional PID controller. Here is the equation for the control law: 7\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:C\left(s\right)={k}_{p}+\frac{{k}_{i}}{s}+{k}_{d}s$$\end{document} The three tuning parameters \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{k}_{p}$$\end{document}, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{k}_{i}$$\end{document}, and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{k}_{d}$$\end{document} make up the transfer function in the s-domain of the conventional PID controller. Here is the equation for the control law: The three tuning parameters \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{k}_{p}$$\end{document}, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{k}_{i}$$\end{document}, and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{k}_{d}$$\end{document} make up the transfer function in the s-domain of the conventional PID controller. Here is the equation for the control law: Unlike the typical classical PID, the FOPID controller can accommodate non-integer orders for the components of derivative and integral. This means control law equation requires to defined five parameters –\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{k}_{p}$$\end{document}, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{k}_{i}$$\end{document}, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{k}_{d}$$\end{document}, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:\lambda\:$$\end{document}, and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\mu\:}$$\end{document}. The corresponding transfer function for the FOPID controller can be described as follows: 8\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:C\left(s\right)={k}_{p}+\frac{{k}_{i}}{{s}^{\lambda\:}}+{k}_{d}{s}^{\mu\:}$$\end{document} Unlike the typical classical PID, the FOPID controller can accommodate non-integer orders for the components of derivative and integral. This means control law equation requires to defined five parameters –\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{k}_{p}$$\end{document}, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{k}_{i}$$\end{document}, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{k}_{d}$$\end{document}, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:\lambda\:$$\end{document}, and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\mu\:}$$\end{document}. The corresponding transfer function for the FOPID controller can be described as follows: Unlike the typical classical PID, the FOPID controller can accommodate non-integer orders for the components of derivative and integral. This means control law equation requires to defined five parameters –\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{k}_{p}$$\end{document}, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{k}_{i}$$\end{document}, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{k}_{d}$$\end{document}, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:\lambda\:$$\end{document}, and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\mu\:}$$\end{document}. The corresponding transfer function for the FOPID controller can be described as follows: The F-FOPID with a fractional filter differs from the standard FOPID controller in that the orders of the derivative, integral, and filter can all be non-integers. This means that the F-FOPID controller requires seven parameters to be defined: \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{k}_{p}$$\end{document}, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{k}_{i}$$\end{document}, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{k}_{d}$$\end{document}, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\lambda\:}_{1}$$\end{document}, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:\mu\:$$\end{document},\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:\:N$$\end{document}, and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\lambda\:}_{2}$$\end{document}. The transfer function of the FOPID controller is presented as 9\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:C\left(s\right)={k}_{p}+\frac{{k}_{i}}{{s}^{{\lambda\:}_{1}}}+{k}_{d}{s}^{\mu\:}\left(\frac{N}{1+\frac{N}{{s}^{{\lambda\:}_{2}}}}\right){N}_{\beta\:}$$\end{document} The F-FOPID with a fractional filter differs from the standard FOPID controller in that the orders of the derivative, integral, and filter can all be non-integers. This means that the F-FOPID controller requires seven parameters to be defined: \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{k}_{p}$$\end{document}, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{k}_{i}$$\end{document}, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{k}_{d}$$\end{document}, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\lambda\:}_{1}$$\end{document}, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:\mu\:$$\end{document},\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:\:N$$\end{document}, and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\lambda\:}_{2}$$\end{document}. The transfer function of the FOPID controller is presented as The F-FOPID with a fractional filter differs from the standard FOPID controller in that the orders of the derivative, integral, and filter can all be non-integers. This means that the F-FOPID controller requires seven parameters to be defined: \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{k}_{p}$$\end{document}, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{k}_{i}$$\end{document}, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{k}_{d}$$\end{document}, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\lambda\:}_{1}$$\end{document}, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:\mu\:$$\end{document},\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:\:N$$\end{document}, and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\lambda\:}_{2}$$\end{document}. The transfer function of the FOPID controller is presented as The filter’s order is denoted by \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{N}_{\beta\:}$$\end{document}. The Oustaloup method creates tenth-order fractional transfer functions for the frequency range of \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\omega\:}\:\text{ϵ}\:({10}^{-5},100\text{k}$$\end{document} ) rad/s. In metaheuristic algorithm for tuning the PID controller parameters, it is critical to carefully choose a suitable objective function. This objective function serves as a critical guide for the optimization algorithm to explore optimal solutions within the problem space. Previous studies have proposed various objective functions, which form the basis for the current study on finding the best parameters for the PID controller for the AVR system. To enhance understanding of the problem in this paper, Table 3 presents some mathematical expressions of the objective functions used in previous literature. The variables in Table 3. are associated with respective definitions, including: \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:e\left(t\right)$$\end{document}: difference between the desired and terminal voltages is an error signal.\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:t$$\end{document}: duration of the simulation.\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{E}_{ss}$$\end{document} : steady-state error.\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\text{P}}_{\text{m}}$$\end{document}: phase margin measures the maximum phase lag that a system can handle without becoming unstable.\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\omega\:}_{gc}$$\end{document}: frequency at which a control system gain hits unity is known as the gain crossover frequency.\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\rho\:}_{\mathcal{W}}$$\end{document} : weighting factor that balances between overshoot, steady-state error, and timing aspects of the responses.\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\beta\:}$$\end{document} : is a weighting factor. \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:e\left(t\right)$$\end{document}: difference between the desired and terminal voltages is an error signal. \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:e\left(t\right)$$\end{document}: difference between the desired and terminal voltages is an error signal. \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:t$$\end{document}: duration of the simulation. \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:t$$\end{document}: duration of the simulation. \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{E}_{ss}$$\end{document} : steady-state error. \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{E}_{ss}$$\end{document} : steady-state error. \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\text{P}}_{\text{m}}$$\end{document}: phase margin measures the maximum phase lag that a system can handle without becoming unstable. \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\text{P}}_{\text{m}}$$\end{document}: phase margin measures the maximum phase lag that a system can handle without becoming unstable. \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\omega\:}_{gc}$$\end{document}: frequency at which a control system gain hits unity is known as the gain crossover frequency. \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\omega\:}_{gc}$$\end{document}: frequency at which a control system gain hits unity is known as the gain crossover frequency. \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\rho\:}_{\mathcal{W}}$$\end{document} : weighting factor that balances between overshoot, steady-state error, and timing aspects of the responses. \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\rho\:}_{\mathcal{W}}$$\end{document} : weighting factor that balances between overshoot, steady-state error, and timing aspects of the responses. \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\beta\:}$$\end{document} : is a weighting factor. \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\beta\:}$$\end{document} : is a weighting factor. These variables are of paramount importance in evaluating and analysing the AVR system, providing insights into its performance and stability under different conditions. Table 3Some objective functions commonly utilised in previous studies.ReferenceObjective faction (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{J}_{x}$$\end{document}) 20 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{J}_{x}=ISA={\int\:}_{0}^{\infty\:}\left|e\left(t\right)\right|dt$$\end{document} 8 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{J}_{x}=ITAE={\int\:}_{0}^{\infty\:}t\left|e\left(t\right)\right|dt$$\end{document} 7 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{J}_{x}=ITSE={\int\:}_{0}^{\infty\:}t{e\left(t\right)}^{2}dt$$\end{document} 5 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{J}_{x}=ZLG=\left(1-{e}^{-\beta\:}\right)\left({M}_{p}+{E}_{ss}\right)+{e}^{-\beta\:}({t}_{s}-{t}_{r})$$\end{document} 8 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{J_x} = \mu \: \cdot ITSE + ZLG$$\end{document} 9 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{J}_{x}=ITAE\{\left(1-{e}^{{\rho\:}_{\mathcal{W}}\:}\right)\left({M}_{p}+{E}_{ss}\right)+{e}^{{\rho\:}_{\mathcal{W}}\:}\left({t}_{s}-{t}_{r}\right)\}$$\end{document} 22 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{J}_{x\:1}={\omega\:}_{gc}.\:{J}_{x\:2}={P}_{m}$$\end{document} Table 3Some objective functions commonly utilised in previous studies.ReferenceObjective faction (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{J}_{x}$$\end{document}) 20 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{J}_{x}=ISA={\int\:}_{0}^{\infty\:}\left|e\left(t\right)\right|dt$$\end{document} 8 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{J}_{x}=ITAE={\int\:}_{0}^{\infty\:}t\left|e\left(t\right)\right|dt$$\end{document} 7 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{J}_{x}=ITSE={\int\:}_{0}^{\infty\:}t{e\left(t\right)}^{2}dt$$\end{document} 5 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{J}_{x}=ZLG=\left(1-{e}^{-\beta\:}\right)\left({M}_{p}+{E}_{ss}\right)+{e}^{-\beta\:}({t}_{s}-{t}_{r})$$\end{document} 8 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{J_x} = \mu \: \cdot ITSE + ZLG$$\end{document} 9 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{J}_{x}=ITAE\{\left(1-{e}^{{\rho\:}_{\mathcal{W}}\:}\right)\left({M}_{p}+{E}_{ss}\right)+{e}^{{\rho\:}_{\mathcal{W}}\:}\left({t}_{s}-{t}_{r}\right)\}$$\end{document} 22 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{J}_{x\:1}={\omega\:}_{gc}.\:{J}_{x\:2}={P}_{m}$$\end{document} Some objective functions commonly utilised in previous studies. The next sub-Sect. 6.1 of this study describes the objective function used to fi-ne-tune the considered AVR system controllers. Comprehensive analysis of prior research In order to underscore the novel contributions of this paper, a comprehensive analysis of prior research is presented. Specifically, Table 1 displays the ideal parameter values for different PID controllers, which were tuned using various optimisation algorithms. Subsequently, simulations were performed using these parameters to demonstrate step response of the AVR system. Figure 3 shows the step response data for each optimization algorithm, as outlined in Table 1. This analysis underscores the distinct discoveries and contributions of each presented study in context of existing studies. Fig. 3The step responses of the AVR system using controller parameters derived from the reviews described in Table 1. The step responses of the AVR system using controller parameters derived from the reviews described in Table 1. All the methods achieve no steady-state error, but significant differences lie in the overshoot, peak time, rise time, and settling time. In control theory, a feedback system’s dynamic response is generally described using four key parameters – rise time \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\text{t}}_{\text{r}}$$\end{document}, settling time \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{t}_{s}$$\end{document}, peak time \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{t}_{p}$$\end{document}, and maximum overshoot \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{M}_{p}$$\end{document} – to perform a comprehensive review of the methods shown in Table 1. in terms of the transient response parameters. In the same context for each algorithm, Fig. 4 depicts the transient response parameters derived from the corresponding step response. Thus, a thorough evaluation of the methods should include examining these parameters, in addition to the null steady-state error, as they play a crucial role in assessing the performance of a control system. Fig. 4Dynamic response of the AVR system obtained using controller parameters from the studies presented in Table 1. Dynamic response of the AVR system obtained using controller parameters from the studies presented in Table 1. It could be concluded from the evaluation of transient response parameters in Fig. 4 that optimization algorithms such as SOS and WOA offer notable results with reasonable trade-offs between performance metrics among typical PID regulators. The WOA algorithm has high proportional gain, which could mean faster responses at expense of overshoot. The SOS optimization algorithm, on the other hand, has an acceptable balance of stability and response characteristics. For FOPID controllers, algorithms like ACO–FPID and PSO–FOPID provide enhanced performance over classical PID controllers. The PSO-based FOPID controller, on the other hand, has balanced performance, responds quickly, and avoids overshoot. The extra feedforward feature in algorithms like JOA–FOPID can greatly improve performance in more complex fractional-order designs, like those found in FFOPID controllers. It does this by better handling disturbances and achieving more precise control. Classical PID with three variable controllers are simple to use and suitable for many situations. However, FOPID works better for AVR systems with complicated dynamics or a lot of disturbances. The specific requirements of the system, including the desired trade-off between response time, overshoot, and stability, should guide the choice of controller. In addition, it appears in previous studies that the FFOPID received less attention as a controller for the AVR systems. Accordingly. One of the main contributions presented in this paper is a new design of a filtered-fractional-order PID controller for AVR systems, which is based on the hybrid optimisation algorithm to fine-tune the proposed controller parameters. White shark optimisation algorithm WSO is a new metaheuristic algorithm based on models inspired by the hunting and foraging behaviours of white sharks, which include the ability to locate prey in the deep ocean34. In the mathematical models, the location of the optimal solution is unknown, and sharks must search extensively. The WSO algorithm models three behaviours of great white sharks in their search for prey: first, movement towards prey based on wave hesitation; second, random search for prey; and third, locating nearby prey through fish school behaviour. These behaviours are mathematically formulated and used to update the positions of sharks to find optimal solutions. The position is defined within matrix, denoted as \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\left[w\right]}_{\text{n}\times\:\text{d}}$$\end{document} which contains the positions of all white sharks in the search space. The \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:d$$\end{document} represents the number of dimensions; its scaler value, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{{{\varpi\:}}_{\text{d}}}^{\text{i}}$$\end{document}, indicates the location of the ith white shark along the dth dimension.10\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\begin{array}{l}\:{\left[ w \right]_{{\rm{n}} \times \:{\rm{d}}}} = \left[ {\begin{array}{*{20}{c}}{\varpi {\:_1}^1}&{\:\varpi {\:_2}^1}&{\: \ldots \:}&{\:\varpi {\:_d}^1}\\{\:\varpi {\:_1}^2}&{\:\varpi {\:_2}^2}&{\: \cdots \:}&{\:\varpi {\:_d}^2}\\{\: \vdots }& \vdots &{\: \vdots }& \vdots \\{\:\varpi {\:_1}^n}&{\:\varpi {\:_2}^n}&{\: \cdots \:}&{\:\varpi {\:_d}^n}\end{array}} \right]\\{\rm{so}}\:{\rm{that}}\:\varpi {\:_j}^i = Lo{w_j} + {\rm{rand}} \times \:(u{p_j} - lo{w_j})\end{array}$$\end{document} In the jth dimension index, the vector for the ith population index is denoted as \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{{\text{w}}_{\text{j}}}^{\text{i}}$$\end{document}, while \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{up}_{j}$$\end{document} and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\text{l}\text{o}\text{w}}_{\text{j}}$$\end{document}refer to vectors for upper and lower boundary of search space, respectively. Movement speed towards prey White sharks excel at hunting and tracking prey, utilising heightened senses like hearing, sight, and smell. Once they perceive the location of their prey, they move towards it using an undulating motion, as defined in Eq. (11), which describes the behaviour of a population of white sharks, where \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:\text{i}$$\end{document} represents the index of each shark in a population of size \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:\text{n}$$\end{document}, such that \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:\text{i\:}\text{=\:1,2,3…,}\text{\:n.}$$\end{document} The \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{{\nu\:}^{i}}_{k+1}$$\end{document} denotes the velocity of the ith shark at step \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:\text{(k+1)}$$\end{document}, while \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{{\nu\:}^{i}}_{k}$$\end{document} is the speed of the current position of the ith shark at step \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:\text{k}$$\end{document}.11\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\begin{array}{l}\upsilon {\:^i}_{k + 1} = \xi \:[\upsilon {\:^i}_k + \rho {\:_1}({w_{gbest}}_k - {w^i}_k) \times \:{c_1} + \rho {\:_2}({w^{\upsilon {\:^i}_k}}_{best} - {w^i}_k) \times \:{c_2}]\\{\rm{where}}\:{\rm{the}}\:{\rm{undulating}}\:{\rm{motion}}\:\upsilon \: = \left[ {n \times \:{\rm{rand}}\left( {1,n} \right)} \right] + 1\end{array}$$\end{document} Global best position vector achieved by any shark in kth iteration is denoted as \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{{w}^{i}}_{{gbest}_{k}}$$\end{document}, and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{{w}^{i}}_{k}\:$$\end{document}represents the current position of the ith shark at step k. The best-known position of the ith shark within the swarm is denoted as \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{{w}^{{\nu\:}^{i}}}_{best}$$\end{document}, where \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\nu\:}^{i}$$\end{document} is the index vector of the sharks reaching the best position. \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\text{c}}_{1}$$\end{document} and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\text{c}}_{2}$$\end{document} are random numbers generated uniformly in the range [0, 1]. The forces of the sharks controlling the effects of \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{w}_{{best}_{k}}$$\end{document}and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{{w}^{{\nu\:}^{i}}}_{best}$$\end{document} on \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{{w}^{i}}_{k}\:$$\end{document}are represented by \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{{\rho\:}}_{1}$$\end{document} and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{{\rho\:}}_{2}$$\end{document} respectively. These forces are computed using Eq. (12). Also, the constriction factor \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\xi\:}$$\end{document} is used to analyse the convergence behaviour of white sharks. The vector \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:\text{r}\text{a}\text{n}\text{d}(1,n)$$\end{document} represents a set of random numbers generated with a uniform distribution between 0 and 1. 12\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\begin{array}{l}\:\rho {\:_1} = \rho {\:_{max}} + (\rho {\:_{max}} - \rho {\:_{min}}) \times \:{e^{ - {{(\frac{{4k}}{{{k_{max}}}})}^2}}}\\\:\rho {\:_2} = \rho {\:_{min}} + (\rho {\:_{max}} - \rho {\:_{min}}) \times \:{e^{ - {{(\frac{{4k}}{{{k_{max}}}})}^2}}}\\\:\xi \: = \frac{2}{{\left| {2 - \alpha \: - \sqrt {\alpha {\:^2} - 4\alpha \:} } \right|}}\end{array}$$\end{document} where \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:k$$\end{document} and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{k}_{max}$$\end{document} denote the present iteration and total number of iterations, respectively, while \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\rho\:}_{min}$$\end{document} and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\rho\:}_{max}$$\end{document} indicate the initial and subordinate velocities for white sharks to attain optimal motion. Through the theoretical analysis, the values of \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\rho\:}_{min}$$\end{document}, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\rho\:}_{max}$$\end{document}, and acceleration coefficient \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:\alpha\:$$\end{document} are selected to be 0.5,1.5, and 4.125, respectively. Movement towards optimal prey Great white sharks actively search for prey, constantly adjusting their positions based on prey movements and scents. Equation (13) expresses their position-updating behaviour when moving towards prey34.13\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\begin{array}{*{20}{l}}{\:{w^i}_{k + 1} = \left\{ {\begin{array}{*{20}{c}}{{w^i}_k. \sim \oplus \:{w_o} + u.\:{{\rm{Y}}_a} + low.\:{{\rm{Y}}_b};R < {m_{\nu \:}}}\\{\:{w^i}_k + \frac{{\nu {\:^i}_k}}{\Omega };R \ge \:{m_{\nu \:}}}\end{array}} \right.}\\{{\rm{Thus}},\:\:}\\{{{\rm{Y}}_a} = {\rm{sign}}({w^i}_k - up) > 0}\\{\:{{\rm{Y}}_b} = {\rm{sign}}({w^i}_k - low) < 0}\\{{w_o} = \oplus \:(\:{{\rm{Y}}_a},\:{{\rm{Y}}_b})}\end{array}$$\end{document} In the \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:(k+1)$$\end{document}th iteration step, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{{w}^{i}}_{k+1}$$\end{document}denotes the updated position vector of the ith white shark, and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:\mathcal{R}=\text{rand}\in\:\left[0\:1\right]$$\end{document}. The \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:\sim$$\end{document} operator is used for negation, while \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{{\rm\:Y}}_{a}$$\end{document} and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{{\rm\:Y}}_{b}$$\end{document} are one-dimension binary vectors, and symbol \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:\oplus\:$$\end{document} represents a bit-wise logical XOR operation. The vectors\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{{\rm\:Y}}_{a}$$\end{document} and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{{\rm\:Y}}_{b}$$\end{document} enable flexible behaviour within the search space. The search space’s lower and upper limits are represented by \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:low$$\end{document} and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:up$$\end{document}, respectively, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{w}_{o}$$\end{document} is a logical vector, and Ω represents the frequency constant, as per Eq. (14). Equation (13) consists of two parts. The first part, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:\text{rand\:<\:}{\text{m}}_{\text{{\nu}}}$$\end{document}, represents the random positional adjustments of white sharks in response to the prey’s movement or scent. The second part of Eq. (13) utilizes the second equation of motion with constant acceleration, resembling \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:\text{x\:=\:}{x}_{i}+{\nu\:}_{i}\varDelta\:t+1/2a{(\varDelta\:t)}^{2}$$\end{document}. In the given context, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:\text{x}$$\end{document} represents the updated position of the white shark, while \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{x}_{i}\:$$\end{document}and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\nu\:}_{i}$$\end{document} are the initial position and the initial speed of the white shark at time \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{t}_{i}$$\end{document}, respectively, and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:\varDelta\:t$$\end{document} the time elapsed between the initial and current positions. The position equation uses \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:i$$\end{document} as the starting point of the calculation to predict the shark’s position at a later time. For example, at \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:i=1$$\end{document}, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{x}_{1}\:$$\end{document} represents the shark’s position at the first-time step, and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\nu\:}_{1}$$\end{document} represents its velocity. However, the time element is excluded as iterations serve as a time factor in optimisation problems. This equation simulates white sharks’ movement towards prey, especially when they detect prey-induced waves \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:\mathcal{R}\text{≥\:}{\text{m}}_{\text{{\nu}}}\:$$\end{document}. It highlights their position updates in the search space upon sensing prey.14\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:\Omega \: = \Omega {\:_{Min}} + \frac{{\Omega {\:_{Max}} - \Omega {\:_{Min}}}}{{\Omega {\:_{Max}} + \Omega {\:_{Min}}}}\:,\:\:at\:R \ge \:{{\rm{m}}_\nu }$$\end{document} where \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\Omega\:}_{Min}$$\end{document} and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\Omega\:}_{Max}$$\end{document} are the minimum and maximum frequencies of the undulating motion and assumed to have taken the values of 0.07 and 0.75, respectively34. The movement force as the shark approaches prey is described by \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{m}_{\nu\:}$$\end{document}, as expressed in Eq. (15).15\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{m}_{\nu\:}=\frac{1}{({a}_{o}+{\text{e}}^{\raisebox{1ex}{$\raisebox{1ex}{${\text{\:k}}_{\text{Max}}$}\!\left/\:\!\raisebox{-1ex}{$\text{2}\text{-}\text{k}$}\right.)\:$}\!\left/\:\!\raisebox{-1ex}{${\text{a}}_{\text{1}}$}\right.})}$$\end{document} where \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{a}_{o}$$\end{document}and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\text{a}}_{\text{1}}$$\end{document} are two positive constants values used to control the behaviour of exploration and exploitation. These values are assumed to be 6.25 and 100, respectively. Equation (16) describes the great white sharks’ proximity behaviour towards their prey.16\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{{\widehat{\text{w}}}^{\text{i}}}_{\text{k}+1}={\text{w}}_{{\text{g}\text{b}\text{e}\text{s}\text{t}}_{\text{k}}}+{{\zeta\:}}_{1}{\overrightarrow{\text{D}}}_{\text{w}}\text{s}\text{g}\text{n}\left({{\zeta\:}}_{2}-0.5\right),\:\:{{\zeta\:}}_{3}<{\text{s}}_{\text{s}}$$\end{document} where \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{{\widehat{\text{w}}}^{\text{i}}}_{\text{k+1}}$$\end{document} is the updated position of ith white shark relative to the prey’s position. The direction of the search is specified by the \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:\text{sgn(}{\text{ζ}}_{\text{2}}\text{-0.5)}$$\end{document} value, which can be either 1 or − 1. The variables \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\zeta\:}_{1},{\zeta\:}_{2},\:$$\end{document}and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\zeta\:}_{3}$$\end{document} are random values between 0 and 1, while \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\overrightarrow{D}}_{w}$$\end{document} is the distance between the white shark and its prey, as defined in Eq. (18). The current position of a white shark relative to its global best position \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{w}_{{gbest}_{k}}$$\end{document} is determined by \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{{w}^{i}}_{k}$$\end{document} and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:\mathcal{R}$$\end{document}.17\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\overrightarrow{D}}_{w}=\left|\mathcal{R}\times\:({w}_{{gbest}_{k}}-{{w}^{i}}_{k})\right|$$\end{document} The parameter \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{s}_{s}$$\end{document} is proposed to express the strength of the senses as defined in Eq. (18).18\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{s}_{s}=\left|1-{e}^{(-{a}_{2}\times\:k/{k}_{Max})}\right|$$\end{document} A positive constant \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{a}_{2}$$\end{document} is used to regulate the exploration and exploitation behaviours in the tested problems of the study. Its value for all test problems is 0. 0005. The sharks’ positions are updates based on the first two optimal solutions while preserving them. Equation (19) defines the behaviour of a school of white sharks34.19\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{{\text{w}}^{\text{i}}}_{\text{k}+1}=\frac{{{\text{w}}^{\text{i}}}_{\text{k}}+{{\widehat{\text{w}}}^{\text{i}}}_{\text{k}+1}}{2\times\:\mathcal{R}}$$\end{document} Algorithm 1 outlines the main steps of WSO and summarises the pseudocode of the model. The algorithm outlines the initialization of essential variables, including the number of white sharks, their positions, and the global best solution. Algorithm 1. Pseudocode of the iterative optimisation process of WSO.1. Set up the parameters for WSO initialization2. Randomly generate the initial positions for WSO Eq. (10).3. Initialise the velocity for the initial population4. Evaluate the positions of the initial population5. while \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\text{(k\:<\:k}}_{\text{Max}})$$\end{document} do6. Update \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\upsilon \: = \left[ {n \times \:{\rm{rand}}\left( {1,n} \right)} \right] + 1$$\end{document} and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\rho\:}_{1},{\rho\:}_{2},\xi\:\:,{{\rm\:Y}}_{a},{{\rm\:Y}}_{b},{w}_{o},\Omega\:,\:{m}_{\nu\:},\text{a}\text{n}\text{d},\:{s}_{s}\:$$\end{document}using Eqs. (12), (13), (14), (15), (16), (17), (18), and (19), respectively.7. for \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:\text{i}=1\:\text{t}\text{o}\:n$$\end{document} do8. \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\upsilon {\:^i}_{k + 1} = \xi \:\left[ {\upsilon {\:^i}_k + \rho {\:_1}\left( {{w_{gbest}}_k - {w^i}_k} \right) \times \:{c_1} + \rho {\:_2}\left( {{w^{\upsilon {\:^i}_k}}_{best} - {w^i}_k} \right) \times \:{c_2}} \right]$$\end{document}9. if \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\text{r}\text{a}\text{n}\text{d}