Parameter-free optimization algorithm effective and precise solution of the optimal power flow problem Authors: Saket Gupta, Neeraj Kumar, Mohd Zuhaib, Marwan Ahmad Abdullah Sufyan, Upma Singh, Sandeep Sharma, Vijay Bhuria, Ankita Gupta Published: 2026 ABSTRACT Difficult Optimization problems can be efficiently solved through nature-inspired optimization algorithms, which remain problem-independent and are computational models at a conceptual level. In this work, a Jaya algorithm will be offered for tackling the Optimal Power Flow (OPF) problem. The basic idea of the Jaya algorithm for optimization problems is that the candidate solutions should move closer to the global optimum solution without converging at suboptimal solutions. Like other nature-inspired optimization techniques, the Jaya method is parameter-free and does not require any algorithm-specific control factors, such as learning or mutant parameters. The parameter freedom of an optimization algorithm not only improves its simplicity, but it also overcomes the challenge of adjusting optimization algorithm parameters, which affects performance as indicated in the literature, and can be expensive for some computations. The OPF problem aims to optimize the different objective functions using the control variables of the power systems. These include improving the voltage stability, decreasing the cost and emissions, and reducing power loss. For comparison purposes, the Jaya optimization technique for the OPF problem has been tested on the IEEE 30-Bus, 57-Bus and IEEE 118 Bus systems. The results were compared to previously reported outcomes of optimization approaches. The result of the computational study reveals the effectiveness of the Jaya optimization technique when compared with the other reported techniques. For instance, when the Jaya algorithm was applied to the IEEE 30-bus system, the fuel cost was reduced by approximately 11.31% compared to the initial operating condition. Introduction Artificial Intelligence (AI) can be considered an interdisciplinary field of study, which has different perspectives for the development of Intelligent Systems. These systems have the ability to perform different functions, which are typically done by human beings, such as reasoning, solving problems, learning, perceiving, and comprehending languages. The Nature-inspired (NI) algorithms are part of artificial intelligence. NI algorithms, which are metaheuristic algorithms based on processes observed in nature, are frequently used in AI systems to improve performance, solve complex optimization tasks, and simulate intelligent behavior. Over the previous two decades, the globe has seen a population increase as well as enormous technical advancement. All of this has caused a rise in global energy consumption. The ubiquitous application of OPF solutions spans the operation, control, planning, and even future expansion of electrical power systems1. Within a network, OPF algorithms employ control variables within feasible regions to optimize predefined objectives. Minimizing fuel cost is a frequent primary objective, but others like voltage stability enhancement, voltage profile improvement, loss reduction, and emission minimization can also be included through adjustments to continuous and discrete control variables. Common control variables for OPF include generator real power outputs (except the slack bus), bus voltages, transformer tap settings, phase shifters, shunt capacitors, and FACTS controllers. Mathematically, OPF exhibits a complex structure characterized by nonlinearity, high dimensionality, non-differentiability, multi-modality, non-convexity, and a mix of discrete and continuous control variables. At the beginning of 60’s, Carpentier addressed the OPF problem as an extension of economic load dispatch for the first time in history2. Traditional approaches significantly focused on solving the OPF problem through classical methods3. These attempts often took advantage of the gradients or derivatives of the objective function to search for an optimal solution, but were often difficult to execute because these approaches had a tendency to get stuck in local optima instead of true global minimum. Furthermore, their range of success was greatly reduced due to problems such as non-convexity, non-differentiability, multi-modality, and dependence on starting points. These problems made classical methods highly inefficient when dealing with the challenges presented by OPF. A large variety of complex optimization problems have been successfully resolved with NI algorithms. NI algorithms are, however, unconstraint search techniques and require clear mechanisms to continue searching in restricted areas. Therefore, constraint optimization problem is of great importance within the optimization field because, due to their physical, geometrical and other limitation, most problems with optimization have different types of constraints. The method of handling constraints plays a very important role when using Nature-inspired algorithms to solve constraint optimization problems. A large variety of complex optimization problems have been successfully resolved with Nature-inspired algorithms. The NI algorithms are not based on the structures mathematics does, meaning they perform perfectly in conditions of non-differentiability, non-smoothness, multi-objectivity, and non-convexity. This kind of flexibility, along with swift improvements in branches of Nature-inspired algorithms, made it attractive to apply these techniques to OPF4 and other complex optimization problems. The power engineers have been fascinated in the past few years with the hundreds of successful applications using NI algorithms to solve different optimization problems, OPF being one of them. Hybrid Rao-2 sine cosine algorithm5, Improved Liver Cancer Algorithm6,, improved pelican optimization algorithm7, hybrid artificial rabbits’ optimization8, Spider Wasp Optimization Algorithm9 Black-Hole Optimization10, Linear Adaptive Genetic Algorithm11, MSA12, WEA13, SOSA14, Tree-seed15, DSA16 have demonstrated notable results. A concise review of selected relevant studies is summarized in Table 1. Table 1A comparative review of selected relevant works.S. noYearPaper titleOPF type/systemOptimization/solution techniqueMain contribution12025 1 Solving Single- and Multi-Objective Optimal Power Flow Problems Using the Spider Wasp Optimization AlgorithmSingle & Multi-objective AC OPFSpider Wasp Optimization Algorithm (SWOA)Provides efficient solutions for both single and multi-objective OPF problems with improved convergence accuracy and stability.22025 5 Optimal Power Flow Solutions for Normal and Critical Loading Scenarios Using Hybrid Rao-2 Sine Cosine AlgorithmAC OPF under normal and critical loadingHybrid Rao-2 Algorithm + Sine Cosine AlgorithmDemonstrates improved convergence, stability, and solution quality for OPF under both normal and stressed loading conditions, outperforming conventional methods.32024 6 Stochastic Optimal Power Flow Framework with Incorporation of Wind Turbines and Solar PVs Using an Improved Liver Cancer AlgorithmStochastic OPF with Wind & Solar PVImproved Liver Cancer Algorithm (ILCA)Effectively manages renewable uncertainty and improves fuel cost, power loss, and voltage stability performance.420257An Improved Pelican Optimization Algorithm for Solving Stochastic Optimal Power Flow Problem Considering Renewable Energy UncertaintyStochastic OPF with RESImproved Pelican Optimization Algorithm (POA)Achieves reliable stochastic OPF solutions under renewable uncertainty with enhanced convergence behavior.52024 8 Optimizing Reactive Power Dispatch in Electrical Networks Using a Hybrid Artificial Rabbits and Gradient-Based OptimizationReactive Power Dispatch OPFHybrid Artificial Rabbits Algorithm + Gradient-Based OptimizationEnhances voltage profile and minimizes power losses through effective reactive power optimization.62025 17 Hybrid Brown-Bear and Hippopotamus Optimization with Quasi-Opposition-Based Learning for Optimal Power Flow with Renewable Energy IntegrationAC OPF with RESHybrid Brown-Bear Optimization + Hippopotamus Optimization + QOBLImproves convergence speed and avoids local minima in renewable-integrated OPF while reducing fuel cost, losses, and voltage deviation.72025 18 A generic optimal power flow solution associated with technical improvements and emission reduction by multi-objective ARO algorithmMulti-objective AC OPFArithmetic Optimization Algorithm (ARO)Simultaneously minimises fuel cost, emission, and technical indices using multi-objective ARO.82025 19 Optimal power flow solution using novel optimization technique: a case studyAC OPFNovel metaheuristic optimization techniqueDemonstrates the effectiveness of a newly proposed optimization technique for improving OPF solution quality and convergence through a practical case study.92024 20 A Modified Artificial Hummingbird Algorithm for Solving Optimal Power Flow Problem in Power SystemsAC OPFModified Artificial Hummingbird Algorithm (MAHA)Introduces MAHA to enhance convergence speed and solution quality for OPF, outperforming classical metaheuristics on standard IEEE test systems.102024 21 Chaotic-Quasi-Oppositional-Phasor Based Multi-Populations Gorilla Troop Optimizer for Optimal Power Flow SolutionAC OPFChaotic-Quasi-Oppositional Gorilla Troop Optimizer (CQO-GTO)Applies CQO-GTO to improve exploration and exploitation for OPF, achieving better convergence and robustness for multi-objective power system optimization.112023 22 Optimal power flow solution using a learning-based sine–cosine algorithmAC OPFLearning-Based Sine Cosine Algorithm (LB-SCA)Enhances conventional SCA by incorporating learning mechanisms to improve convergence speed, solution accuracy, and robustness for OPF problems.122023 23 Gaussian bare-bones levy circulatory system-based optimization for power flow in the presence of renewable unitsRenewable-integrated AC OPFGaussian Bare-Bones Levy Circulatory System OptimizationEnhances exploration–exploitation balance and robustness of OPF solutions under renewable uncertainty using Gaussian distribution and Levy flight mechanisms.132014 24 Multi-objective optimal power flow considering the cost, emission, voltage deviation and power losses using multi-objective modified imperialist competitive algorithmMulti-objective AC OPFMulti-Objective Modified Imperialist Competitive Algorithm (MOMICA)Simultaneously minimizes fuel cost, emissions, voltage deviation, and active power losses, establishing a benchmark framework for multi-objective OPF optimization.142025 25 Optimal power flow in power systems with renewable energy resources uncertainty, including geothermal power plantsProbabilistic OPF with RES & geothermalProbabilistic modeling + metaheuristic optimizationIncorporates geothermal power plants along with wind and solar uncertainties to improve the feasibility and reliability of OPF solutions.152022 26 Solution of probabilistic optimal power flow incorporating renewable energy uncertainty using a novel circle search algorithmProbabilistic OPF with RESCircle Search AlgorithmProposes a novel circle search optimizer to efficiently handle renewable uncertainty in probabilistic OPF while minimizing cost and losses. Table 1A comparative review of selected relevant works.S. noYearPaper titleOPF type/systemOptimization/solution techniqueMain contribution12025 1 Solving Single- and Multi-Objective Optimal Power Flow Problems Using the Spider Wasp Optimization AlgorithmSingle & Multi-objective AC OPFSpider Wasp Optimization Algorithm (SWOA)Provides efficient solutions for both single and multi-objective OPF problems with improved convergence accuracy and stability.22025 5 Optimal Power Flow Solutions for Normal and Critical Loading Scenarios Using Hybrid Rao-2 Sine Cosine AlgorithmAC OPF under normal and critical loadingHybrid Rao-2 Algorithm + Sine Cosine AlgorithmDemonstrates improved convergence, stability, and solution quality for OPF under both normal and stressed loading conditions, outperforming conventional methods.32024 6 Stochastic Optimal Power Flow Framework with Incorporation of Wind Turbines and Solar PVs Using an Improved Liver Cancer AlgorithmStochastic OPF with Wind & Solar PVImproved Liver Cancer Algorithm (ILCA)Effectively manages renewable uncertainty and improves fuel cost, power loss, and voltage stability performance.420257An Improved Pelican Optimization Algorithm for Solving Stochastic Optimal Power Flow Problem Considering Renewable Energy UncertaintyStochastic OPF with RESImproved Pelican Optimization Algorithm (POA)Achieves reliable stochastic OPF solutions under renewable uncertainty with enhanced convergence behavior.52024 8 Optimizing Reactive Power Dispatch in Electrical Networks Using a Hybrid Artificial Rabbits and Gradient-Based OptimizationReactive Power Dispatch OPFHybrid Artificial Rabbits Algorithm + Gradient-Based OptimizationEnhances voltage profile and minimizes power losses through effective reactive power optimization.62025 17 Hybrid Brown-Bear and Hippopotamus Optimization with Quasi-Opposition-Based Learning for Optimal Power Flow with Renewable Energy IntegrationAC OPF with RESHybrid Brown-Bear Optimization + Hippopotamus Optimization + QOBLImproves convergence speed and avoids local minima in renewable-integrated OPF while reducing fuel cost, losses, and voltage deviation.72025 18 A generic optimal power flow solution associated with technical improvements and emission reduction by multi-objective ARO algorithmMulti-objective AC OPFArithmetic Optimization Algorithm (ARO)Simultaneously minimises fuel cost, emission, and technical indices using multi-objective ARO.82025 19 Optimal power flow solution using novel optimization technique: a case studyAC OPFNovel metaheuristic optimization techniqueDemonstrates the effectiveness of a newly proposed optimization technique for improving OPF solution quality and convergence through a practical case study.92024 20 A Modified Artificial Hummingbird Algorithm for Solving Optimal Power Flow Problem in Power SystemsAC OPFModified Artificial Hummingbird Algorithm (MAHA)Introduces MAHA to enhance convergence speed and solution quality for OPF, outperforming classical metaheuristics on standard IEEE test systems.102024 21 Chaotic-Quasi-Oppositional-Phasor Based Multi-Populations Gorilla Troop Optimizer for Optimal Power Flow SolutionAC OPFChaotic-Quasi-Oppositional Gorilla Troop Optimizer (CQO-GTO)Applies CQO-GTO to improve exploration and exploitation for OPF, achieving better convergence and robustness for multi-objective power system optimization.112023 22 Optimal power flow solution using a learning-based sine–cosine algorithmAC OPFLearning-Based Sine Cosine Algorithm (LB-SCA)Enhances conventional SCA by incorporating learning mechanisms to improve convergence speed, solution accuracy, and robustness for OPF problems.122023 23 Gaussian bare-bones levy circulatory system-based optimization for power flow in the presence of renewable unitsRenewable-integrated AC OPFGaussian Bare-Bones Levy Circulatory System OptimizationEnhances exploration–exploitation balance and robustness of OPF solutions under renewable uncertainty using Gaussian distribution and Levy flight mechanisms.132014 24 Multi-objective optimal power flow considering the cost, emission, voltage deviation and power losses using multi-objective modified imperialist competitive algorithmMulti-objective AC OPFMulti-Objective Modified Imperialist Competitive Algorithm (MOMICA)Simultaneously minimizes fuel cost, emissions, voltage deviation, and active power losses, establishing a benchmark framework for multi-objective OPF optimization.142025 25 Optimal power flow in power systems with renewable energy resources uncertainty, including geothermal power plantsProbabilistic OPF with RES & geothermalProbabilistic modeling + metaheuristic optimizationIncorporates geothermal power plants along with wind and solar uncertainties to improve the feasibility and reliability of OPF solutions.152022 26 Solution of probabilistic optimal power flow incorporating renewable energy uncertainty using a novel circle search algorithmProbabilistic OPF with RESCircle Search AlgorithmProposes a novel circle search optimizer to efficiently handle renewable uncertainty in probabilistic OPF while minimizing cost and losses. A comparative review of selected relevant works. 2025 1 2025 5 2024 6 2024 8 2025 17 2025 18 2025 19 2024 20 2024 21 2023 22 2023 23 2014 24 2025 25 2022 26 NI algorithms are random search techniques that are simple to formulate, more convenient compared to analytical methods, and efficient in identifying solutions very close to optimal values. On the flip side, there are chances of fast convergence of the techniques in case of improper selection of parameters of the algorithm; therefore, in order to overcome poor/non-optimal solutions, there is a need for optimization of the NI algorithms’ parameters. Parameter selection of NI techniques is an arduous task and time-consuming as well. In overcoming the said drawback, the authors of the current research present parameter-less optimization techniques of the algorithm, called ‘Jaya Algorithms’. This feature of having “no dependency of exploration and exploitation search capability of the algorithm on the specific factors of the algorithm” makes the search capability of the current approach better compared to the previous studies mentioned in Section I. However, one key question emerges: What is the point of creating new algorithms when there are a multitude of existing ones? The solution lies in the inherent trade-offs present in each of the NI algorithms. They attempt to maintain a balance between the two actions: exploration, searching new areas, and exploitation, optimizing the promising solutions. Some algorithms perform well in the former but not in the latter, and the reverse, too. Moreover, each algorithm has its own set of pros and cons, depending on the nature of the problem it tackles. This is backed by the “No Free Lunch” theorem, which states that there is no optimal algorithm for broad optimization tasks27.In order to solve OPF problems, the Jaya method is presented in this study. The contributions of this paper are: The utilization of the parameter-free optimization algorithm, known as the Jaya optimization algorithm, to solve single-objective and multi-objective optimal power flow problems in power systems.The implementation and testing of the optimization algorithm on typical IEEE test cases, including large-scale networks like the IEEE 118-bus network.The evaluation of the optimization algorithm in terms of solution reliability by running the optimization algorithm several times.The comparison with some widely used optimization algorithms shows the strength of the Jaya optimization algorithm in terms of convergence characteristics. The utilization of the parameter-free optimization algorithm, known as the Jaya optimization algorithm, to solve single-objective and multi-objective optimal power flow problems in power systems. The utilization of the parameter-free optimization algorithm, known as the Jaya optimization algorithm, to solve single-objective and multi-objective optimal power flow problems in power systems. The implementation and testing of the optimization algorithm on typical IEEE test cases, including large-scale networks like the IEEE 118-bus network. The implementation and testing of the optimization algorithm on typical IEEE test cases, including large-scale networks like the IEEE 118-bus network. The evaluation of the optimization algorithm in terms of solution reliability by running the optimization algorithm several times. The evaluation of the optimization algorithm in terms of solution reliability by running the optimization algorithm several times. The comparison with some widely used optimization algorithms shows the strength of the Jaya optimization algorithm in terms of convergence characteristics. The comparison with some widely used optimization algorithms shows the strength of the Jaya optimization algorithm in terms of convergence characteristics. The authors of this article present an application of the Jaya algorithm to solve the OPF problem while accounting for technical and economic objective functions. The suggested algorithm is a parameter less constraint optimization algorithm, so algorithm-specific parameter tuning is not required. This quality makes the algorithm very simple to use and applies to different kinds of optimization problems. The algorithm-specific parameters play a very important role during the search process, and it is a very challenging task for the given problem. While the proposed implementation of the Jaya algorithm has demonstrated competitive performance across standard IEEE test systems, several limitations must be acknowledged. In particular, since the Jaya is mainly based on the greedy update strategy of Jaya, the risk of premature convergence may arise. To alleviate this issue, three mitigation strategies are employed: (i) random population initialization within sufficiently wide variable bounds, (ii) execution of multiple independent simulation runs, and (iii) incorporation of stochastic terms in the update equations to implicitly enhance exploration capability. The remainder of the paper is organized like this. Section 2 describes OPF problem modelling. Details on the proposed Jaya algorithm for solving the OPF problem are provided in Sect. 3. The numerical results achieved by the proposed method are given in Sect. 4. In the last section, the conclusion is presented. Problem formulation The OPF problem can be mathematically formulated as follows28:1\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:\mathrm{O}\mathrm{p}\mathrm{t}\mathrm{i}\mathrm{m}\mathrm{i}\mathrm{z}\mathrm{e}\:M\:\left(U,\:X\right)$$\end{document} Subject to the constraints; 2\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:\left\{\begin{array}{c}g\:(U,\:X)\:=\:0\:\:\\\:h\:(U,\:X)\:\le\:\:0\end{array}\right.\:$$\end{document} Subject to the constraints; where\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:M\:\left(U,\:X\right)$$\end{document} indicates the objective function that has to be minimised. The set of dependent variables is represented by ‘U’ and ‘X’, and it consists of: Here, U can be expressed as3\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\mathrm{U}}^{\mathrm{T}}=[{{P}_{{g}_{1}},\mathrm{V}}_{1}\dots\:{\mathrm{V}}_{\mathrm{N}\mathrm{L}\mathrm{B}},{\mathrm{Q}}_{\mathrm{g}1},\dots\:{\mathrm{Q}}_{\mathrm{g}\mathrm{N}\mathrm{G}},{S}_{1},\dots\:{S}_{\mathrm{n}\mathrm{t}\mathrm{l}}]\:\:$$\end{document} while X can be expressed as4\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\mathrm{X}}^{\mathrm{T}}=[{\mathrm{V}}_{{\mathrm{g}}_{1}}\dots\:{\mathrm{V}}_{{\mathrm{g}}_{\mathrm{N}\mathrm{G}\mathrm{N}}},{\mathrm{Q}}_{{\mathrm{C}}_{1}}\dots\:{\mathrm{Q}}_{{\mathrm{C}}_{\mathrm{N}\mathrm{C}}},{\mathrm{T}}_{1}\dots\:{\mathrm{T}}_{\mathrm{N}\mathrm{T}\mathrm{R}}]\:\:\:$$\end{document} Constraints Equality Constraints (g):Power flow equations are used to define the equality restrictions. Power flow equations are separated into two types: active and reactive. The active and reactive power flow equations are as follows.5\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:0\:\:=\:\:\:\:\:\:\:\mathrm{P}}_{\mathrm{g}\mathrm{i}}-{\mathrm{P}}_{\mathrm{d}\mathrm{i}}-{\mathrm{P}}_{\mathrm{L}\mathrm{o}\mathrm{s}\mathrm{s}}$$\end{document}6\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\:\:0\:\:=\:\:\:\:\:\:\mathrm{Q}}_{\mathrm{g}\mathrm{i}}-{\mathrm{Q}}_{\mathrm{d}\mathrm{i}}-\:\:\:{\mathrm{Q}}_{\mathrm{L}\mathrm{o}\mathrm{s}\mathrm{s}}\:\:$$\end{document}Inequality Constraints (h): Equality Constraints (g):Power flow equations are used to define the equality restrictions. Power flow equations are separated into two types: active and reactive. The active and reactive power flow equations are as follows.5\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:0\:\:=\:\:\:\:\:\:\:\mathrm{P}}_{\mathrm{g}\mathrm{i}}-{\mathrm{P}}_{\mathrm{d}\mathrm{i}}-{\mathrm{P}}_{\mathrm{L}\mathrm{o}\mathrm{s}\mathrm{s}}$$\end{document}6\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\:\:0\:\:=\:\:\:\:\:\:\mathrm{Q}}_{\mathrm{g}\mathrm{i}}-{\mathrm{Q}}_{\mathrm{d}\mathrm{i}}-\:\:\:{\mathrm{Q}}_{\mathrm{L}\mathrm{o}\mathrm{s}\mathrm{s}}\:\:$$\end{document} Equality Constraints (g): Power flow equations are used to define the equality restrictions. Power flow equations are separated into two types: active and reactive. The active and reactive power flow equations are as follows.5\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:0\:\:=\:\:\:\:\:\:\:\mathrm{P}}_{\mathrm{g}\mathrm{i}}-{\mathrm{P}}_{\mathrm{d}\mathrm{i}}-{\mathrm{P}}_{\mathrm{L}\mathrm{o}\mathrm{s}\mathrm{s}}$$\end{document}6\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\:\:0\:\:=\:\:\:\:\:\:\mathrm{Q}}_{\mathrm{g}\mathrm{i}}-{\mathrm{Q}}_{\mathrm{d}\mathrm{i}}-\:\:\:{\mathrm{Q}}_{\mathrm{L}\mathrm{o}\mathrm{s}\mathrm{s}}\:\:$$\end{document} Inequality Constraints (h): Inequality Constraints (h): Power systems are made up of numerous components and equipment. There are limitations to how these devices can operate. These devices’ minimum and maximum limits are known as inequality constraints. Generators constraints:7\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\:\:\:\:\:\:\:\:\:\:\mathrm{V}}_{{\mathrm{g}}_{\mathrm{k}}}^{\mathrm{m}\mathrm{i}\mathrm{n}}\le\:{\mathrm{V}}_{{\mathrm{g}}_{\mathrm{k}}}\le\:{\mathrm{V}}_{{\mathrm{g}}_{\mathrm{k}}}^{\mathrm{m}\mathrm{a}\mathrm{x}}\:\:\:\:\:\:\:\:\mathrm{k}=1\dots\:...\mathrm{N}\mathrm{G}\mathrm{N}\:\:\:\:\:\:\:\:\:$$\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}$$\:{\:\:\:\:\:\:\:\:\:\:\mathrm{Q}}_{{\mathrm{g}}_{\mathrm{k}}}^{\mathrm{m}\mathrm{i}\mathrm{n}}\le\:\:{\mathrm{Q}}_{{\mathrm{g}}_{\mathrm{k}}}\le\:{\mathrm{Q}}_{{\mathrm{g}}_{\mathrm{k}}}^{\mathrm{m}\mathrm{a}\mathrm{x}}\:\:\:\:\:\mathrm{k}=1\dots\:\dots\:\mathrm{N}\mathrm{G}\mathrm{N}\:\:\:\:\:\:$$\end{document}Shunt VAR compensator constraints:9\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\:\:\:\:\:\:\:\mathrm{Q}}_{{\mathrm{C}}_{\mathrm{k}}}^{\mathrm{m}\mathrm{i}\mathrm{n}}\le\:{\mathrm{Q}}_{{\mathrm{C}}_{\mathrm{k}}}\le\:{\mathrm{Q}}_{{\mathrm{C}}_{\mathrm{k}}}^{\mathrm{m}\mathrm{a}\mathrm{x}}\:\:\:\mathrm{k}=1\dots\:\dots\:\mathrm{N}\mathrm{G}\mathrm{N}\:\:\:\:$$\end{document}Transformer Constraints:10\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\:\:\:\:\:\:\:\:\:\:\mathrm{T}}_{\mathrm{k}}^{\mathrm{m}\mathrm{i}\mathrm{n}}\:\le\:{\mathrm{T}}_{\mathrm{k}}\le\:{\mathrm{T}}_{\mathrm{k}}^{\mathrm{m}\mathrm{a}\mathrm{x}}\:\:\:\mathrm{k}=1\dots\:\dots\:\mathrm{N}\mathrm{T}\mathrm{R}\:\:\:$$\end{document}Security Constraints: 11\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\:\:\:\:\:\:\:\:\:\:\:\:\mathrm{V}}_{{\mathrm{L}}_{\mathrm{k}}}^{\mathrm{m}\mathrm{i}\mathrm{n}}\le\:{\mathrm{V}}_{{\mathrm{L}}_{\mathrm{k}}}\le\:{\mathrm{V}}_{{\mathrm{L}}_{\mathrm{k}}}^{\mathrm{m}\mathrm{a}\mathrm{x}}\:\:\:\:\:\:\:\:\mathrm{k}=1\dots\:\dots\:\mathrm{N}\mathrm{L}\mathrm{B}\:\:\:\:\:\:\:\:$$\end{document} 12\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\:\:\:\:\:\mathrm{S}}_{{\mathrm{l}}_{\mathrm{k}}}\le\:{\mathrm{S}}_{{\mathrm{l}}_{\mathrm{k}}}^{\mathrm{m}\mathrm{a}\mathrm{x}}\:\:\:\:\:\:\:\:\:\mathrm{k}=1\dots\:\dots\:\mathrm{n}\mathrm{t}\mathrm{l}\:\:$$\end{document} Where, Pgi &Qgi: Generator active and reactive power output of ithbus, Pdi &Qdi: Active and reactive power demands at ithbus,\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\mathrm{Q}}_{{\mathrm{C}}_{\mathrm{k}}}$$\end{document}=Limits on output of shunt VAR compensators at kthbus, VLk: Load bus voltage limits at kth bus, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\:\mathrm{S}}_{{\mathrm{l}}_{\mathrm{k}}}$$\end{document}=Line flow limit in MVA, NTR: No of regulating transformer, NGN, and NLB: No of generator and Load bus respectively. Generators constraints:7\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\:\:\:\:\:\:\:\:\:\:\mathrm{V}}_{{\mathrm{g}}_{\mathrm{k}}}^{\mathrm{m}\mathrm{i}\mathrm{n}}\le\:{\mathrm{V}}_{{\mathrm{g}}_{\mathrm{k}}}\le\:{\mathrm{V}}_{{\mathrm{g}}_{\mathrm{k}}}^{\mathrm{m}\mathrm{a}\mathrm{x}}\:\:\:\:\:\:\:\:\mathrm{k}=1\dots\:...\mathrm{N}\mathrm{G}\mathrm{N}\:\:\:\:\:\:\:\:\:$$\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}$$\:{\:\:\:\:\:\:\:\:\:\:\mathrm{Q}}_{{\mathrm{g}}_{\mathrm{k}}}^{\mathrm{m}\mathrm{i}\mathrm{n}}\le\:\:{\mathrm{Q}}_{{\mathrm{g}}_{\mathrm{k}}}\le\:{\mathrm{Q}}_{{\mathrm{g}}_{\mathrm{k}}}^{\mathrm{m}\mathrm{a}\mathrm{x}}\:\:\:\:\:\mathrm{k}=1\dots\:\dots\:\mathrm{N}\mathrm{G}\mathrm{N}\:\:\:\:\:\:$$\end{document} Generators constraints:7\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\:\:\:\:\:\:\:\:\:\:\mathrm{V}}_{{\mathrm{g}}_{\mathrm{k}}}^{\mathrm{m}\mathrm{i}\mathrm{n}}\le\:{\mathrm{V}}_{{\mathrm{g}}_{\mathrm{k}}}\le\:{\mathrm{V}}_{{\mathrm{g}}_{\mathrm{k}}}^{\mathrm{m}\mathrm{a}\mathrm{x}}\:\:\:\:\:\:\:\:\mathrm{k}=1\dots\:...\mathrm{N}\mathrm{G}\mathrm{N}\:\:\:\:\:\:\:\:\:$$\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}$$\:{\:\:\:\:\:\:\:\:\:\:\mathrm{Q}}_{{\mathrm{g}}_{\mathrm{k}}}^{\mathrm{m}\mathrm{i}\mathrm{n}}\le\:\:{\mathrm{Q}}_{{\mathrm{g}}_{\mathrm{k}}}\le\:{\mathrm{Q}}_{{\mathrm{g}}_{\mathrm{k}}}^{\mathrm{m}\mathrm{a}\mathrm{x}}\:\:\:\:\:\mathrm{k}=1\dots\:\dots\:\mathrm{N}\mathrm{G}\mathrm{N}\:\:\:\:\:\:$$\end{document} Shunt VAR compensator constraints:9\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\:\:\:\:\:\:\:\mathrm{Q}}_{{\mathrm{C}}_{\mathrm{k}}}^{\mathrm{m}\mathrm{i}\mathrm{n}}\le\:{\mathrm{Q}}_{{\mathrm{C}}_{\mathrm{k}}}\le\:{\mathrm{Q}}_{{\mathrm{C}}_{\mathrm{k}}}^{\mathrm{m}\mathrm{a}\mathrm{x}}\:\:\:\mathrm{k}=1\dots\:\dots\:\mathrm{N}\mathrm{G}\mathrm{N}\:\:\:\:$$\end{document} Shunt VAR compensator constraints:9\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\:\:\:\:\:\:\:\mathrm{Q}}_{{\mathrm{C}}_{\mathrm{k}}}^{\mathrm{m}\mathrm{i}\mathrm{n}}\le\:{\mathrm{Q}}_{{\mathrm{C}}_{\mathrm{k}}}\le\:{\mathrm{Q}}_{{\mathrm{C}}_{\mathrm{k}}}^{\mathrm{m}\mathrm{a}\mathrm{x}}\:\:\:\mathrm{k}=1\dots\:\dots\:\mathrm{N}\mathrm{G}\mathrm{N}\:\:\:\:$$\end{document} Transformer Constraints:10\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\:\:\:\:\:\:\:\:\:\:\mathrm{T}}_{\mathrm{k}}^{\mathrm{m}\mathrm{i}\mathrm{n}}\:\le\:{\mathrm{T}}_{\mathrm{k}}\le\:{\mathrm{T}}_{\mathrm{k}}^{\mathrm{m}\mathrm{a}\mathrm{x}}\:\:\:\mathrm{k}=1\dots\:\dots\:\mathrm{N}\mathrm{T}\mathrm{R}\:\:\:$$\end{document} Transformer Constraints:10\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\:\:\:\:\:\:\:\:\:\:\mathrm{T}}_{\mathrm{k}}^{\mathrm{m}\mathrm{i}\mathrm{n}}\:\le\:{\mathrm{T}}_{\mathrm{k}}\le\:{\mathrm{T}}_{\mathrm{k}}^{\mathrm{m}\mathrm{a}\mathrm{x}}\:\:\:\mathrm{k}=1\dots\:\dots\:\mathrm{N}\mathrm{T}\mathrm{R}\:\:\:$$\end{document} Security Constraints: 11\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\:\:\:\:\:\:\:\:\:\:\:\:\mathrm{V}}_{{\mathrm{L}}_{\mathrm{k}}}^{\mathrm{m}\mathrm{i}\mathrm{n}}\le\:{\mathrm{V}}_{{\mathrm{L}}_{\mathrm{k}}}\le\:{\mathrm{V}}_{{\mathrm{L}}_{\mathrm{k}}}^{\mathrm{m}\mathrm{a}\mathrm{x}}\:\:\:\:\:\:\:\:\mathrm{k}=1\dots\:\dots\:\mathrm{N}\mathrm{L}\mathrm{B}\:\:\:\:\:\:\:\:$$\end{document} 12\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\:\:\:\:\:\mathrm{S}}_{{\mathrm{l}}_{\mathrm{k}}}\le\:{\mathrm{S}}_{{\mathrm{l}}_{\mathrm{k}}}^{\mathrm{m}\mathrm{a}\mathrm{x}}\:\:\:\:\:\:\:\:\:\mathrm{k}=1\dots\:\dots\:\mathrm{n}\mathrm{t}\mathrm{l}\:\:$$\end{document} Where, Pgi &Qgi: Generator active and reactive power output of ithbus, Pdi &Qdi: Active and reactive power demands at ithbus,\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\mathrm{Q}}_{{\mathrm{C}}_{\mathrm{k}}}$$\end{document}=Limits on output of shunt VAR compensators at kthbus, VLk: Load bus voltage limits at kth bus, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\:\mathrm{S}}_{{\mathrm{l}}_{\mathrm{k}}}$$\end{document}=Line flow limit in MVA, NTR: No of regulating transformer, NGN, and NLB: No of generator and Load bus respectively. Security Constraints: 11\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\:\:\:\:\:\:\:\:\:\:\:\:\mathrm{V}}_{{\mathrm{L}}_{\mathrm{k}}}^{\mathrm{m}\mathrm{i}\mathrm{n}}\le\:{\mathrm{V}}_{{\mathrm{L}}_{\mathrm{k}}}\le\:{\mathrm{V}}_{{\mathrm{L}}_{\mathrm{k}}}^{\mathrm{m}\mathrm{a}\mathrm{x}}\:\:\:\:\:\:\:\:\mathrm{k}=1\dots\:\dots\:\mathrm{N}\mathrm{L}\mathrm{B}\:\:\:\:\:\:\:\:$$\end{document} 12\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\:\:\:\:\:\mathrm{S}}_{{\mathrm{l}}_{\mathrm{k}}}\le\:{\mathrm{S}}_{{\mathrm{l}}_{\mathrm{k}}}^{\mathrm{m}\mathrm{a}\mathrm{x}}\:\:\:\:\:\:\:\:\:\mathrm{k}=1\dots\:\dots\:\mathrm{n}\mathrm{t}\mathrm{l}\:\:$$\end{document} Where, Pgi &Qgi: Generator active and reactive power output of ithbus, Pdi &Qdi: Active and reactive power demands at ithbus,\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\mathrm{Q}}_{{\mathrm{C}}_{\mathrm{k}}}$$\end{document}=Limits on output of shunt VAR compensators at kthbus, VLk: Load bus voltage limits at kth bus, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\:\mathrm{S}}_{{\mathrm{l}}_{\mathrm{k}}}$$\end{document}=Line flow limit in MVA, NTR: No of regulating transformer, NGN, and NLB: No of generator and Load bus respectively. Incorporation of constraints Operating constraints, encompassing generator capabilities, VAR limitations, transformer capacities, line flow limits, and bus voltage bounds, are effectively managed through penalty factors within the OPF framework. The objective function has been augmented with penalty terms corresponding to each constraint violation. The mathematical formulations of the penalty functions, along with the selected penalty coefficients and their role in guiding the optimization process toward feasible solutions, have now been clearly presented in Eq. (13).13\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\begin{aligned}{\mathrm{M}}_{\mathrm{a}\mathrm{u}\mathrm{g}} & =\mathrm{M}\left(\mathrm{U},\mathrm{X}\right)+{P}_{1}{\sum\:}_{k=1}^{NGN}h\left({P}_{{g}_{Slack}}-{P}_{{g}_{Slack}}^{lim}\right)+{CP}_{2}{\sum\:}_{k=1}^{NGN}h\left({Q}_{{g}_{k}}-{Q}_{{g}_{k}}^{lim}\right) \\ & \quad +{P}_{3}{\sum\:}_{k=1}^{NLB}h\left({V}_{{L}_{k}}-{V}_{{L}_{k}}^{lim}\right)+{P}_{4}{\sum\:}_{k=1}^{NGN}h\left({S}_{{l}_{k}}-{S}_{{l}_{k}}^{lim}\right)\end{aligned}$$\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}$$\:{\:\mathrm{P}}_{1}$$\end{document},\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\mathrm{P}}_{2}$$\end{document}, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\mathrm{P}}_{3}$$\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}$$\:{\mathrm{P}}_{4}$$\end{document} represent penalty factors for limit violations, which are chosen by the user. Jaya algorithm The Jaya algorithm proposed by Rao29 is a powerful global optimization method, which is mainly characterized by the absence of algorithm-specific parameters. This algorithm imitates the natural phenomenon of problem-solving in which every attempt aims toward an optimal solution while avoiding a lesser solution. In this respect, the present algorithm is similar to TLBO, and its non-availability of algorithm-specific parameters makes it easier to implement for various engineering optimization problems. The flow chart of the Jaya algorithm is given in Fig. 1. Fig. 1Flowchart for Jaya algorithm. Flowchart for Jaya algorithm. Key characteristics: Objective function: Maximizes (or minimizes) M(z).Initial solution: Randomly generated within variable bounds.Update mechanism: Variables are updated using Eq. (14): 14\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{z}_{i+1,\:\:j,\:k}={z}_{i,\:\:j,\:k}+{\alpha\:\:}_{i,\:\:j,\:1}({z}_{i,\:\:j,\:B}-\left|{(z}_{i,\:\:j,\:k})\right|)-{\alpha\:\:}_{i,\:\:j,\:2}({z}_{i,\:\:j,\:W}-\:\left|{(z}_{i,\:\:j,\:k})\right|)\:$$\end{document} Objective function: Maximizes (or minimizes) M(z). Objective function: Maximizes (or minimizes) M(z). Initial solution: Randomly generated within variable bounds. Initial solution: Randomly generated within variable bounds. Update mechanism: Variables are updated using Eq. (14): Update mechanism: Variables are updated using Eq. (14): Where: \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{z}_{i,\:\:j,\:B}$$\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}$$\:{z}_{i,\:\:j,\:W}$$\end{document} denote the indices of the population’s best and worst solutions, respectively.i, j, and k represent the indices of the candidate solution, variable, and iteration, respectively.\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{z}_{i,\:\:j,\:k}$$\end{document}signifies the jth variable of the ith candidate solution in the kth iteration.\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\alpha\:\:}_{i,\:\:j,\:1}$$\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}$$\:{\alpha\:\:}_{i,\:\:j,\:2}$$\end{document}are randomly generated numbers within [0, 1], acting as scaling factors for exploration.Absolute values of variables further promote exploration. \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{z}_{i,\:\:j,\:B}$$\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}$$\:{z}_{i,\:\:j,\:W}$$\end{document} denote the indices of the population’s best and worst solutions, respectively. \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{z}_{i,\:\:j,\:B}$$\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}$$\:{z}_{i,\:\:j,\:W}$$\end{document} denote the indices of the population’s best and worst solutions, respectively. i, j, and k represent the indices of the candidate solution, variable, and iteration, respectively. i, j, and k represent the indices of the candidate solution, variable, and iteration, respectively. \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{z}_{i,\:\:j,\:k}$$\end{document}signifies the jth variable of the ith candidate solution in the kth iteration. \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{z}_{i,\:\:j,\:k}$$\end{document}signifies the jth variable of the ith candidate solution in the kth iteration. \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\alpha\:\:}_{i,\:\:j,\:1}$$\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}$$\:{\alpha\:\:}_{i,\:\:j,\:2}$$\end{document}are randomly generated numbers within [0, 1], acting as scaling factors for exploration. \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\alpha\:\:}_{i,\:\:j,\:1}$$\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}$$\:{\alpha\:\:}_{i,\:\:j,\:2}$$\end{document}are randomly generated numbers within [0, 1], acting as scaling factors for exploration. Absolute values of variables further promote exploration. Absolute values of variables further promote exploration. Computational steps for the OPF problem Input: Population size (P), maximum iterations (I), control variables (D), and variable bounds (Xmin and Xmax).Initialization: Set iteration count I = 0, randomly generate ‘P’ initial solutions within bounds, and evaluate fitness using Eq. (13).Identify best and worst solutions: Determine \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{z}_{i,\:\:j,\:B}$$\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}$$\:{z}_{i,\:\:j,\:W}$$\end{document}based on fitness in the feasible search space.Generate new solutions: Apply Eq. (14) to create new candidate solutions.Acceptance: Accept new solutions if superior to previous ones, otherwise retain previous solutions.Iteration: If I < Imax, increment I and return to step 3. Otherwise, proceed to step 7.Output: Show the fitness value of the optimal solution. Input: Population size (P), maximum iterations (I), control variables (D), and variable bounds (Xmin and Xmax). Input: Population size (P), maximum iterations (I), control variables (D), and variable bounds (Xmin and Xmax). Initialization: Set iteration count I = 0, randomly generate ‘P’ initial solutions within bounds, and evaluate fitness using Eq. (13). Initialization: Set iteration count I = 0, randomly generate ‘P’ initial solutions within bounds, and evaluate fitness using Eq. (13). Identify best and worst solutions: Determine \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{z}_{i,\:\:j,\:B}$$\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}$$\:{z}_{i,\:\:j,\:W}$$\end{document}based on fitness in the feasible search space. Identify best and worst solutions: Determine \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{z}_{i,\:\:j,\:B}$$\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}$$\:{z}_{i,\:\:j,\:W}$$\end{document}based on fitness in the feasible search space. Generate new solutions: Apply Eq. (14) to create new candidate solutions. Generate new solutions: Apply Eq. (14) to create new candidate solutions. Acceptance: Accept new solutions if superior to previous ones, otherwise retain previous solutions. Acceptance: Accept new solutions if superior to previous ones, otherwise retain previous solutions. Iteration: If I < Imax, increment I and return to step 3. Otherwise, proceed to step 7. Iteration: If I < Imax, increment I and return to step 3. Otherwise, proceed to step 7. Output: Show the fitness value of the optimal solution. Output: Show the fitness value of the optimal solution. Numerical results and discussion The efficacy of the proposed OPF algorithm was rigorously evaluated on three distinct test systems: the IEEE 30-bus, 57-bus, and 118-bus systems. In order to evaluate the functioning of the algorithm, 11 unique case studies were examined. The IEEE 30-bus system used a population size of 40, and maximum iteration limit of 100 is set. Subsequently, the IEEE 57-bus and IEEE 118-bus systems had these values increased to 50 and 150. The parameters were set this way to guarantee that the search space was correctly optimised while still being computationally effective. The simulations were carried out using a personal computer (PC) containing a 1.7 GHz Intel Core i5 processor, 4GB RAM, and a 64-bit operating system, utilising the MATLAB computing environment. IEEE 30-bus test system The Jaya method is then tested with the same system for a total of 30 independent runs with varying objective functions of the OPF. The performance potential of the Jaya method at its optimum is then highlighted by displaying its best possible outcome amongst these runs. For a better comparison, PSO, DE, and TLBO algorithms were also tested with the same objective functions of the OPF with the same parameters, like population, number of generations and control variables. The control variables’ operational limits and specific system characteristics were taken from the standard benchmark data available in30. The IEEE-30 test bus system is shown in Fig. 2 (Table 2). Fig. 2Single-line diagram of IEEE-30 Bus Test System. Single-line diagram of IEEE-30 Bus Test System. Table 2Overview of OPF optimization cases with corresponding objectives.Case no.Objective(s)Optimization typeDescriptionIEEE 30 Bus system Case 1Fuel cost minimizationSingle-objective optimizationMinimizes the total generation cost under standard OPF constraints Case 2Fuel cost + voltage profile improvementBi-objective optimizationBalances economic operation with voltage quality across the network Case 3Fuel cost + voltage stability enhancementBi-objective optimizationSimultaneously optimizes cost and enhances voltage stability Case 4Fuel cost + voltage stability enhancement during contingencyBi-objective optimizationSimultaneously optimises cost, voltage stability enhancement during contingency Case 5Minimize the real power lossesSingle-objective optimizationMinimizes the total power losses Case 6Emission minimizationSingle-objective optimizationReducing environmental impact from generation emissionsIEEE 57 Bus system Case 7Fuel cost minimizationSingle-objective optimizationMinimizes the total generation cost under standard OPF constraints Case 8Fuel cost + voltage profile improvementBi-objective optimizationBalances economic operation with voltage quality across the network Case 9Fuel cost + voltage stability enhancementSingle-objective optimizationoptimizes voltage stability Case 10Minimize the real power lossesSingle-objective optimizationMinimizes the total power lossesIEEE 118 Bus system Case 11Fuel cost minimizationSingle-objective optimizationMinimizes the total generation cost under standard OPF constraints Table 2Overview of OPF optimization cases with corresponding objectives.Case no.Objective(s)Optimization typeDescriptionIEEE 30 Bus system Case 1Fuel cost minimizationSingle-objective optimizationMinimizes the total generation cost under standard OPF constraints Case 2Fuel cost + voltage profile improvementBi-objective optimizationBalances economic operation with voltage quality across the network Case 3Fuel cost + voltage stability enhancementBi-objective optimizationSimultaneously optimizes cost and enhances voltage stability Case 4Fuel cost + voltage stability enhancement during contingencyBi-objective optimizationSimultaneously optimises cost, voltage stability enhancement during contingency Case 5Minimize the real power lossesSingle-objective optimizationMinimizes the total power losses Case 6Emission minimizationSingle-objective optimizationReducing environmental impact from generation emissionsIEEE 57 Bus system Case 7Fuel cost minimizationSingle-objective optimizationMinimizes the total generation cost under standard OPF constraints Case 8Fuel cost + voltage profile improvementBi-objective optimizationBalances economic operation with voltage quality across the network Case 9Fuel cost + voltage stability enhancementSingle-objective optimizationoptimizes voltage stability Case 10Minimize the real power lossesSingle-objective optimizationMinimizes the total power lossesIEEE 118 Bus system Case 11Fuel cost minimizationSingle-objective optimizationMinimizes the total generation cost under standard OPF constraints Overview of OPF optimization cases with corresponding objectives. Case 1: Total fuel cost minimization The objective is to reduce the price of generation/fuel cost. It can be described as31:15\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\:\mathrm{F}}_{\mathrm{c}\mathrm{o}\mathrm{s}\mathrm{t}}\left(\mathrm{x},\mathrm{u}\right)={\sum\:}_{\mathrm{i}=1}^{\mathrm{N}\mathrm{G}\mathrm{N}\:\:}{\mathrm{A}}_{\mathrm{i}}+{\mathrm{B}}_{\mathrm{i}}{\mathrm{P}}_{{\mathrm{g}}_{\mathrm{i}}}+{\mathrm{C}}_{\mathrm{i}}{\mathrm{P}}_{{\mathrm{g}}_{\mathrm{i}}}^{2}\:\:(\mathrm{\$}/\mathrm{h})\:\:\:\:$$\end{document} In the direct competition for the minimum fuel cost, Jaya outperformed all others. Equipped with the fuel cost coefficients in31, Jaya thus obtained an outstanding result of 799.9676 $/h, while the others trailed behind in the competition: PSO (800.4391 $/h), DE (800.6636 $/h), and TLBO (800.0492 $/h). A full comparison of the performance is seen in Table 3 below, where the outstanding optimizing abilities of Jaya are seen in its performance compared to the existing algorithms. The convergence patterns of case 1 are seen in the picture below in Fig. 3. The results indicate that the Jaya algorithm provides competitive performance in minimizing fuel cost; however, it does not consistently outperform all benchmark methods across all evaluation metrics. The optimized fuel cost, compared with other algorithms, is presented in Fig. 4. Table 3Comparison of Case 1 numerical results.AlgorithmFCECPLOSSVDLICTNR (Base Case)902.0040.2225.8421.1600.177229.8312 Jaya 799.9676 0.3358 8.9137 1.0988 0.1286 89.8123 TLBO800.04920.33578.94101.14810.128092.1267DE800.66360.33709.14900.89600.130294.6523PSO800.43910.33629.06130.97140.130799.7621MSA12800.50990.366459.03450.903570.13833–DSA16800.38870.366728.9819-0.12624–PSO28800.40750.3669.0060.91290.1276–TLBO28800.41040.36639.0080.90680.1269–ARCBBO32800.51590.36639.02550.88670.1385–RCBBO32800.8703–––––ABC31800.660.365149.03280.92090.1381–SKH33800.51410.36629.0282–0.1382–GWO34801.41–9.3–––DE34801.23–9.22–––MGBICA35801.14090.3296––––GBICA35801.15130.3296––––ECHT-DE36800.41480.365938.99990.912310.13777–**CT= computation time in seconds. Table 3Comparison of Case 1 numerical results.AlgorithmFCECPLOSSVDLICTNR (Base Case)902.0040.2225.8421.1600.177229.8312 Jaya 799.9676 0.3358 8.9137 1.0988 0.1286 89.8123 TLBO800.04920.33578.94101.14810.128092.1267DE800.66360.33709.14900.89600.130294.6523PSO800.43910.33629.06130.97140.130799.7621MSA12800.50990.366459.03450.903570.13833–DSA16800.38870.366728.9819-0.12624–PSO28800.40750.3669.0060.91290.1276–TLBO28800.41040.36639.0080.90680.1269–ARCBBO32800.51590.36639.02550.88670.1385–RCBBO32800.8703–––––ABC31800.660.365149.03280.92090.1381–SKH33800.51410.36629.0282–0.1382–GWO34801.41–9.3–––DE34801.23–9.22–––MGBICA35801.14090.3296––––GBICA35801.15130.3296––––ECHT-DE36800.41480.365938.99990.912310.13777–**CT= computation time in seconds. Comparison of Case 1 numerical results. **CT= computation time in seconds. Fig. 3Convergence characteristics of Case 1. Convergence characteristics of Case 1. Fig. 4Comparative analysis of optimized OPF results with different algorithms. Comparative analysis of optimized OPF results with different algorithms. Case 2: Bi-objective optimization – fuel cost with voltage profile improvement The fuel cost function is commonly coupled with voltage profile improvement to construct a multi-objective formulation, represented in Eq. (16). The suggested objective function simultaneously addresses the following two aspects: minimization fuel expenses and the voltage deviations from 1 p.u. at all loads31.16\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\:\mathrm{F}}_{\mathrm{V}\mathrm{D}}\left(\mathrm{x},\mathrm{u}\right)={\sum\:}_{\mathrm{i}=1}^{\mathrm{N}\mathrm{G}\mathrm{N}\:\:}{\mathrm{f}}_{\mathrm{i}}\left(.\right)\mathrm{\$}/\mathrm{h}+\:{{\upalpha\:}}_{\mathrm{V}\mathrm{D}}{\sum\:}_{\mathrm{i} \in \mathrm{N}\mathrm{L}}^{\:\:}\left|{\mathrm{V}}_{\mathrm{i}}-1\right|$$\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}$$\:{{\upalpha\:}}_{\mathrm{V}\mathrm{D}}$$\end{document} is a weight factor. The minimum voltage deviation is 0.0896 p.u., and a total fuel cost of 841.4409 $/h was found, respectively, while the voltage deviation obtained by PSO, DE, and TLBO algorithms is 0.0994, 0.0962, and 0.0906 p.u. respectively. The voltage profile is significantly improved in case 2. As earlier in case 1, the Total Voltage Deviation (TVD) was 1.0988 p.u, which is reduced to 0.0896 p.u. The optimized OPF results compared to other algorithms are provided in Table 4. In Fig. 5, the voltage profile is shown in case 1 and case 2. Table 4Comparison of Case 2 numerical results.AlgorithmFCECPLOSSVDLICTNR (Base Case)902.0040.2225.8421.1600.177229.8312 Jaya 841.4409 0.4460 15.5359 0.0896 0.1402 92.1478 TLBO878.35930.30638.74080.09060.141195.0917DE804.51360.333110.07620.09620.140896.1176PSO853.18390.327513.41920.09940.140998.2271MSA12803.31250.36349.72060.10840.1478–MPSO12803.97870.36369.92420.12020.149–PSO28803.47360.36399.8210.09780.1371–TLBO28803.56750.36369.8130.09390.1369–ECHT-DE36803.71980.36389.84140.094540.14888–SF-DE36803.42410.36429.78070.097720.14893–SP-DE36803.41960.36329.75730.097760.14893– Table 4Comparison of Case 2 numerical results.AlgorithmFCECPLOSSVDLICTNR (Base Case)902.0040.2225.8421.1600.177229.8312 Jaya 841.4409 0.4460 15.5359 0.0896 0.1402 92.1478 TLBO878.35930.30638.74080.09060.141195.0917DE804.51360.333110.07620.09620.140896.1176PSO853.18390.327513.41920.09940.140998.2271MSA12803.31250.36349.72060.10840.1478–MPSO12803.97870.36369.92420.12020.149–PSO28803.47360.36399.8210.09780.1371–TLBO28803.56750.36369.8130.09390.1369–ECHT-DE36803.71980.36389.84140.094540.14888–SF-DE36803.42410.36429.78070.097720.14893–SP-DE36803.41960.36329.75730.097760.14893– Comparison of Case 2 numerical results. Fig. 5Convergence characteristics of Case 2. Convergence characteristics of Case 2. The weighted sum method (WSM), a classical and widely used scalarization technique in multi-objective optimization, is employed in this study to aggregate multiple objectives. While WSM is appropriate for problems with moderately aligned objectives, we acknowledge its limitations when objectives are strongly conflicting. Case 3: Bi-objective optimization – fuel cost with voltage stability enhancement The growing requirement for electricity exerts pressure on the power system because the generating capacity is unable to match it. The result of the growing gap between the two is a concerning “downward spiral situation,” where the lack of power causes low voltage conditions at the load buses, thereby further aggravating the voltage stability condition in the system. A critical requirement in a healthy power system is the regulation of the bus voltage within acceptable margins in the normal functioning condition as well as during the sudden change in the loading condition. All load buses employ the “L-index,” which ranges from 0 to 1, to monitor the system’s voltage stability conditions. Buses having greater L-index values are more prone to voltage instability. A significant improvement in the voltage stability conditions in the system can be achieved by lowering the “L-index,” a crucial metric for assessing the voltage conditions of the relevant buses31. The fuel cost function is integrated with L-index to formulate a combined objective function, as expressed in Eq. (17).17\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{{\mathrm{F}}_{\mathrm{L}\mathrm{I}}\left(\mathrm{x},\mathrm{u}\right)\:={\sum\:}_{\mathrm{i}=1}^{\mathrm{N}\mathrm{G}\mathrm{N}}{\mathrm{f}}_{\mathrm{i}}\left(.\right)\mathrm{\$}/\mathrm{h}+{{\upalpha\:}}_{\mathrm{L}}\mathrm{L}}_{\mathrm{m}\mathrm{a}\mathrm{x}}\:\:\:\:\:\:\:$$\end{document} The Jaya algorithm resulted in a minimum L-index of 0.1283 and a corresponding total fuel cost of 841.4409 $/h. The Jaya approach presents better results than TLBO, DE, PSO and other reported methods. The optimized OPF results compared to other algorithms are provided in Table 5. The weighted sum method is used in this study due to its simplicity; we recognize that it may not fully capture the Pareto structure of the OPF problem when objectives are strongly conflicting. Table 5Comparison of Case 3 numerical results.AlgorithmFCECPLOSSVDLICTNR (Base Case)902.0040.2225.8421.1600.177229.8312 Jaya 810.2319 0.3408 11.9651 0.6223 0.1283 96.2789 TLBO807.13040.339011.00790.55740.1305100.2611DE809.80850.332611.41770.33270.133299.8901PSO808.00290.339511.29780.37650.1298102.1098MSA12801.22480.36118.97610.92650.1371–IMFO28800.47620.36629.0160.90860.1255–ECHT-DE36800.43210.365859.00430.912440.13739–SF-DE36800.42030.365928.99850.938460.13745–SP-DE36800.43650.365178.98380.937060.13748– Table 5Comparison of Case 3 numerical results.AlgorithmFCECPLOSSVDLICTNR (Base Case)902.0040.2225.8421.1600.177229.8312 Jaya 810.2319 0.3408 11.9651 0.6223 0.1283 96.2789 TLBO807.13040.339011.00790.55740.1305100.2611DE809.80850.332611.41770.33270.133299.8901PSO808.00290.339511.29780.37650.1298102.1098MSA12801.22480.36118.97610.92650.1371–IMFO28800.47620.36629.0160.90860.1255–ECHT-DE36800.43210.365859.00430.912440.13739–SF-DE36800.42030.365928.99850.938460.13745–SP-DE36800.43650.365178.98380.937060.13748– Comparison of Case 3 numerical results. Case 4: Bi-objective optimization – fuel cost with improvement of voltage stability during contingency Regular power system operations experience unplanned events like transmission line failures and generator blackouts, which are termed contingencies. The configuration of a system can change due to contingencies that may impact voltage stability levels31. The objective of Case 4 addresses voltage stability enhancement alongside minimization of fuel cost under scenarios involving single-line outages at buses 2–6. This is to retain operability despite contingencies or disturbances. The proposed technique does an excellent job in handling contingencies, along with optimizing fuel cost and voltage stability. Case 4 is devoted to studies of (n-1) contingencies, taking into consideration single-line outage contingencies for a robust operational system. Under this approach, a minimum value of the L-index was achieved at 0.1363 with a fuel cost of $818.5353/h, outperforming algorithms such as TLBO, DE, and PSO, where the L-indices were 0.1469, 0.1439, and 0.1485, respectively. Table 6 presents comparisons with some other evolutionary computation-based algorithms. Table 6Comparison of Case 4 numerical results.AlgorithmFCECPLOSSVDLICTNR (Base Case)902.0040.2225.8421.1600.177229.8312 Jaya 818.5353 0.3375 14.145 0.7439 0.1363 101.9081 TLBO825.38080.28009.38700.63170.1469108.2716DE810.30120.332411.38680.57540.1439107.1101PSO827.33750.27929.26450.59250.1485109.2651MSA12804.48380.36139.94860.91770.1392– Table 6Comparison of Case 4 numerical results.AlgorithmFCECPLOSSVDLICTNR (Base Case)902.0040.2225.8421.1600.177229.8312 Jaya 818.5353 0.3375 14.145 0.7439 0.1363 101.9081 TLBO825.38080.28009.38700.63170.1469108.2716DE810.30120.332411.38680.57540.1439107.1101PSO827.33750.27929.26450.59250.1485109.2651MSA12804.48380.36139.94860.91770.1392– Comparison of Case 4 numerical results. Case 5: Minimize the real power losses In this section, real power loss minimization was selected as the main objective function. The mathematically \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{F}_{\mathrm{l}\mathrm{o}\mathrm{s}\mathrm{s}}$$\end{document} can be represented as follows:18\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\mathrm{F}}_{\mathrm{L}\mathrm{O}\mathrm{S}\mathrm{S}}\left(\mathrm{x},\mathrm{u}\right){\sum\:}_{\mathrm{k}=1}^{\mathrm{N}\mathrm{T}\:}{\mathrm{G}}_{\mathrm{k}}\left|{\mathrm{v}}_{\mathrm{i}}^{2}+{\mathrm{v}}_{\mathrm{j}}^{2}+2\left|{\mathrm{V}}_{\mathrm{i}}\right|\left|{\mathrm{V}}_{\mathrm{j}}\right|\mathrm{cos}({{\updelta\:}}_{\mathrm{i}}-{{\updelta\:}}_{\mathrm{j}})\:\right|$$\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}$$\:{\mathrm{G}}_{\mathrm{k}}\:$$\end{document}is the conductance of the kth line connected between the ith and jth buses: NT is the number of transmission lines: \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\mathrm{V}}_{\mathrm{i}\:}$$\end{document}is the voltage magnitude at bus i: \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\mathrm{V}}_{\mathrm{j}}$$\end{document}is the magnitude at bus j: \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{{\updelta\:}}_{\mathrm{i}}$$\end{document} is the voltage angles at bus i: \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{{\updelta\:}}_{\mathrm{j}}$$\end{document} is the voltage angle at bus j. The\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\:\mathrm{F}}_{\mathrm{L}\mathrm{o}\mathrm{s}\mathrm{s}\:}$$\end{document}is selected for the real power loss minimization as described in Eq. (18). Here, the nature of all generator cost characteristics is quadratic. The numerical results found after optimization were compared with PSO, DE, TLBO, and other reported results, as given in Table 7. In the present case, the Jaya algorithm gives 3.1086 MW after optimization which is slightly higher than the result obtained by MSA, MPSO, IMFO, PSO, and TLBO methods. The convergence characteristic obtained by the Jaya algorithm for real power loss minimization is represented in Fig. 6. In this Case, the Jaya optimizer yields a result that is slightly inferior to other competing algorithms. This behavior is attributed to (i) the conflicting nature of OPF objectives under the weighted-sum formulation, (ii) the sensitivity of local search dynamics to penalty/constraint interactions, and (iii) the inherent trade-offs between exploration and exploitation. Despite this, the Jaya algorithm maintains overall competitive performance across all test cases and demonstrates stable convergence with lower computational complexity. Table 7Comparison of Case 5 numerical results.AlgorithmFCECPLOSSVDLICTR (Base Case)902.0040.2225.8421.1600.177229.8312 Jaya 967.6830 0.2066 3.1086 1.0361 0.1302 94.1191 TLBO967.75500.20663.13881.00110.129498.1141DE968.14960.20663.30410.41250.139196.1109PSO968.52480.20663.46120.71070.141899.8601MSA12967.66360.207273.10050.88860.1385–MPSO12967.65230.207273.10310.90630.1381–TLBO28967.58410.20733.0880.90290.1274–SKH33967.65940.20723.0987NA0.1385–ECHT-DE36967.60010.207263.08500.909370.13836–SF-DE36967.61980.207263.08450.906480.13818–SP-DE36967.59620.207263.08440.903590.13832– Table 7Comparison of Case 5 numerical results.AlgorithmFCECPLOSSVDLICTR (Base Case)902.0040.2225.8421.1600.177229.8312 Jaya 967.6830 0.2066 3.1086 1.0361 0.1302 94.1191 TLBO967.75500.20663.13881.00110.129498.1141DE968.14960.20663.30410.41250.139196.1109PSO968.52480.20663.46120.71070.141899.8601MSA12967.66360.207273.10050.88860.1385–MPSO12967.65230.207273.10310.90630.1381–TLBO28967.58410.20733.0880.90290.1274–SKH33967.65940.20723.0987NA0.1385–ECHT-DE36967.60010.207263.08500.909370.13836–SF-DE36967.61980.207263.08450.906480.13818–SP-DE36967.59620.207263.08440.903590.13832– Comparison of Case 5 numerical results. Fig. 6Convergence characteristics of Case 5. Convergence characteristics of Case 5. Case 6: Total emission cost minimization The power plant’s emissions of the key pollutants SOx and NOx are monitored closely. These gases are produced by all generating units and their output is estimated using a combination of exponential and quadratic functions based on each unit’s active power. The total emission cost is then determined based on these calculations31.19\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\mathrm{F}}_{\mathrm{E}\mathrm{C}\mathrm{O}\mathrm{S}\mathrm{T}}\left(\mathrm{x},\mathrm{u}\right)={\sum\:}_{\mathrm{i}=1}^{\mathrm{N}\mathrm{G}\:\:}{{\upalpha\:}}_{\mathrm{i}}+\:{{\upbeta\:}}_{\mathrm{i}}{\mathrm{P}}_{{\mathrm{G}}_{\mathrm{i}}}+\:{{\upgamma\:}}_{\mathrm{i}}{\mathrm{P}}_{{\mathrm{G}}_{\mathrm{i}}}^{2}+\:\:{{\upxi\:}}_{\mathrm{i}}\mathrm{exp}\left({{\uplambda\:}}_{\mathrm{i}}{\mathrm{P}}_{{\mathrm{G}}_{\mathrm{i}}}\right)\left(ton/h \right)\:\:\:$$\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}$$\:{\:{\upalpha\:}}_{\mathrm{i}}$$\end{document},\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{{\upbeta\:}}_{\mathrm{i}}$$\end{document},\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{{\upgamma\:}}_{\mathrm{i}}$$\end{document},\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{{\upxi\:}}_{\mathrm{i}}$$\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}$$\:{{\uplambda\:}}_{\mathrm{i}}$$\end{document} are the emission coefficients. The \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\mathrm{F}}_{\mathrm{E}\mathrm{C}\mathrm{O}\mathrm{S}\mathrm{T}}\left(\mathrm{x},\mathrm{u}\right)$$\end{document}is selected for total emission cost optimisation as described in (19). Here, the nature of the generating cost characteristics is quadratic. As earlier in case 1, the emission cost was 0.3358 ton/h, which is reduced to 0.2037 ton/h (39.33%) in case 6. The simulation results of PSO, DE, TLBO, and other methods mentioned in the literature are compared with the Jaya algorithm shown in Table 8. It is seen that the results from the suggested method are excellent in comparison to other methods. The convergence characteristic is shown in Fig. 7. Appendix: Table 15 lists the Jaya algorithm control variables for Cases 1–6. Table 8Comparison of case 6 numerical results.AlgorithmFCECPLOSSVDLICTNR (Base Case)902.0040.2225.8421.1600.177229.8312 Jaya 942.3443 0.2037 3.2162 1.1261 0.1286 97.1118 TLBO944.33740.20393.78070.67570.140296.7198DE915.21850.21264.23250.54930.146798.0191PSO944.17220.20403.96230.82100.1328100.0914MSA12944.39040.204823.23610.880590.1387–DSA16944.40860.20582553.24373–0.12734–TLBO28944.32780.20483.2210.89880.1271–ABC31944.43910.2048263.2470.84630.1402–SKH33944.38020.20483.2326–0.1384–MGBICA35942.84010.2048––––GBICA35944.65160.2049––––ECHT-DE36944.37820.204823.21790.892400.13844–SF-DE36944.32420.204823.21790.896170.13844–SP-DE36944.33230.204823.21780.902150.13830–MAHA37943.97740.20473.20350.12730.8765–MGTO38943.25000.20472.92182.97830.1060–MARO39943.8080.204793.2360.76680.1296– Table 8Comparison of case 6 numerical results.AlgorithmFCECPLOSSVDLICTNR (Base Case)902.0040.2225.8421.1600.177229.8312 Jaya 942.3443 0.2037 3.2162 1.1261 0.1286 97.1118 TLBO944.33740.20393.78070.67570.140296.7198DE915.21850.21264.23250.54930.146798.0191PSO944.17220.20403.96230.82100.1328100.0914MSA12944.39040.204823.23610.880590.1387–DSA16944.40860.20582553.24373–0.12734–TLBO28944.32780.20483.2210.89880.1271–ABC31944.43910.2048263.2470.84630.1402–SKH33944.38020.20483.2326–0.1384–MGBICA35942.84010.2048––––GBICA35944.65160.2049––––ECHT-DE36944.37820.204823.21790.892400.13844–SF-DE36944.32420.204823.21790.896170.13844–SP-DE36944.33230.204823.21780.902150.13830–MAHA37943.97740.20473.20350.12730.8765–MGTO38943.25000.20472.92182.97830.1060–MARO39943.8080.204793.2360.76680.1296– Comparison of case 6 numerical results. Fig. 7Convergence characteristics of Case 6. Convergence characteristics of Case 6. IEEE 57-bus system The effectiveness of the Jaya algorithm was evaluated on the well-known IEEE 57 bus system, a benchmark for power system optimization. Detailed system parameters and operating limits of the control variables were adopted from standard benchmark data reported in40. The IEEE-57 test bus system is shown in Fig. 8. Fig. 8Single-line diagram of the IEEE-57 Bus Test System. Single-line diagram of the IEEE-57 Bus Test System. Case 7: Total fuel cost minimization Minimizing total fuel cost was the primary objective, as in case 1, and the objective function was defined in Eq. (15). All generating cost characteristics were quadratic. The Jaya algorithm achieved the optimal fuel cost of $41657.3521/h, outperforming other reported results. Notably, TLBO, DE, and PSO algorithms obtained costs of $41,872.0668/h, $41,771.1088/h, and $41,909.5363/h, respectively. Table 9 clearly demonstrates the dominance of the Jaya algorithm’s results compared to other methods. Figure 9 illustrates the fuel cost characteristic for case 7. A comparative representation of the optimized fuel cost with different algorithms is illustrated in Fig. 10. Table 9Comparison of Case 7 numerical results.AlgorithmFuel CostPlossVDL-indexCTNR (Base Case)51,395.5706428.365891.255170.309940.7423 Jaya 41657.3521 14.73391.71220.2327 130.2561 TLBO41872.066816.48371.67130.2411132.9081DE41771.108817.36641.54650.2310132.9871PSO41908.536316.29621.02630.2527134.8891MSA1241673.723115.05261.55080.28392–TSA1541,685.0716.9732.92––DSA1641686.82–0.243531.0833–TLBO2841694.777815.74260.7120.2409–SKH3341676.915215.0795–0.2807–ECHT –DE3641670.56214.94790.503190.28886–SF-DE3641667.8514.88641.642090.27971–SP-DE3641667.8214.9091.543670.28123– Table 9Comparison of Case 7 numerical results.AlgorithmFuel CostPlossVDL-indexCTNR (Base Case)51,395.5706428.365891.255170.309940.7423 Jaya 41657.3521 14.73391.71220.2327 130.2561 TLBO41872.066816.48371.67130.2411132.9081DE41771.108817.36641.54650.2310132.9871PSO41908.536316.29621.02630.2527134.8891MSA1241673.723115.05261.55080.28392–TSA1541,685.0716.9732.92––DSA1641686.82–0.243531.0833–TLBO2841694.777815.74260.7120.2409–SKH3341676.915215.0795–0.2807–ECHT –DE3641670.56214.94790.503190.28886–SF-DE3641667.8514.88641.642090.27971–SP-DE3641667.8214.9091.543670.28123– Comparison of Case 7 numerical results. Fig. 9Convergence Characteristics of Case 7. Convergence Characteristics of Case 7. Fig. 10A comparative representation of the optimized fuel cost with different algorithms. A comparative representation of the optimized fuel cost with different algorithms. Case 8: Bi-objective optimization – fuel cost with voltage profile improvement For a stable power system, the bus voltages should satisfy the standard of 1 p.u. In this condition, the objective function sought to concurrently reduce the total cost and enhance the voltage profile, as explained by Eq. (16). In this study, the Jaya method is effective in obtaining a minimal voltage variation of 0.5889p.u. with a fuel cost of $41755.6721/h. The optimized values of the results are shown in Table 10. Case 8 led to a considerable improvement in the voltage profiles. From a total voltage deviation of 1.6953 p.u. in case 7, a highly decreased TVD of 0.5725 p.u. has been achieved in case 8. The results of voltage deviation for the proposed algorithm are compared with the outcomes of PSO, DE, TLBO, and other existing algorithms in Table 10. The convergence curve for this case is illustrated in Fig. 11. Table 10Comparison of Case 8 numerical results.AlgorithmFuel CostPlossVDL-indexCT NR (base case) 51,395.5706428.365891.255170.309940.7423 Jaya 41755.6721 16.6558 0.5889 0.2348 131.8891 TLBO41710.383415.92150.80070.2443133.4912DE4169.110215.42140.76440.2414133.8912PSO41688.441715.47190.98820.2438134.0211FPA1241726.375816.0270.697230.29488–TSA1554,045.1728.5130.72––DSA1641699.4–0.7620.2471–IMFO2841692.717815.75520.71820.24259–ECHT-DE3641694.8215.58060.816590.29198–SF-DE3641697.5215.56160.775720.29262–SP-DE3641697.515.58970.772530.29228– Table 10Comparison of Case 8 numerical results.AlgorithmFuel CostPlossVDL-indexCT NR (base case) 51,395.5706428.365891.255170.309940.7423 Jaya 41755.6721 16.6558 0.5889 0.2348 131.8891 TLBO41710.383415.92150.80070.2443133.4912DE4169.110215.42140.76440.2414133.8912PSO41688.441715.47190.98820.2438134.0211FPA1241726.375816.0270.697230.29488–TSA1554,045.1728.5130.72––DSA1641699.4–0.7620.2471–IMFO2841692.717815.75520.71820.24259–ECHT-DE3641694.8215.58060.816590.29198–SF-DE3641697.5215.56160.775720.29262–SP-DE3641697.515.58970.772530.29228– Comparison of Case 8 numerical results. Fig. 11Convergence characteristics of Case 8. Convergence characteristics of Case 8. Case 9: Voltage stability enhancement With both aims of fuel cost and voltage stability improvement, the objective function was expressed by Eq. (16). The Jaya method was successful and achieved a minimum value of the L-index of 0.2200, signifying a substantial improvement in voltage stability. A comparison of numerical results of both fuel cost and minimum value of L-index using proposed Jaya, TLBO, DE, PSO, and other recently developed meta-heuristic algorithms is provided in Table 11. The results truly revealed that the Jaya method performs better than other methods existing in the literature. Table 11Comparison of Case 9 results.AlgorithmFuel CostPlossVDL-indexCTNR (Base Case)51,395.5706428.365891.255170.309940.7423 Jaya 41665.3565 14.801 1.8417 0.2200 132.9142 TLBO41692.972015.5611.78350.2191132.9912DE41814.963017.36431.57480.2459133.9084PSO41670.4726915.01751.76370.2200134.9071MSA1241675.994815.00261.72360.27481–MPSO1241694.140715.45541.50840.27918–MDE1241689.587815.70921.54470.27677–MFO1241680.193715.10261.72450.27467–FPA1241684.185915.21931.74780.27429–DSA1641761.22–1.05730.2383–IMFO2841673.620415.2991.176250.23525–MFO2841688.652215.4751.39880.2395–GA2841670.087215.1651.24730.2413–PSO2841670.175515.2051.43230.242–TLBO2841685.35315.4781.34530.24787–SKH3343937.105813.3154–0.2721–ECHT-DE3641671.0915.02751.561880.28152–SF-DE3641667.5314.89631.611740.28022–SP-DE3641668.4515.0121.608030.28092– Table 11Comparison of Case 9 results.AlgorithmFuel CostPlossVDL-indexCTNR (Base Case)51,395.5706428.365891.255170.309940.7423 Jaya 41665.3565 14.801 1.8417 0.2200 132.9142 TLBO41692.972015.5611.78350.2191132.9912DE41814.963017.36431.57480.2459133.9084PSO41670.4726915.01751.76370.2200134.9071MSA1241675.994815.00261.72360.27481–MPSO1241694.140715.45541.50840.27918–MDE1241689.587815.70921.54470.27677–MFO1241680.193715.10261.72450.27467–FPA1241684.185915.21931.74780.27429–DSA1641761.22–1.05730.2383–IMFO2841673.620415.2991.176250.23525–MFO2841688.652215.4751.39880.2395–GA2841670.087215.1651.24730.2413–PSO2841670.175515.2051.43230.242–TLBO2841685.35315.4781.34530.24787–SKH3343937.105813.3154–0.2721–ECHT-DE3641671.0915.02751.561880.28152–SF-DE3641667.5314.89631.611740.28022–SP-DE3641668.4515.0121.608030.28092– Comparison of Case 9 results. Case 10: Minimize the real power losses Minimizing real power loss was the objective, as in case 5. The Jaya algorithm achieved optimal values of 9.7924 MW for real power loss and $4441.4980/h for fuel cost, outperforming TLBO, DE, and PSO algorithms with losses of 9.770, 10.30, and 10.00 MW, respectively. Figure 8 shows the convergence characteristic for real power loss minimization. Table 12 compares Jaya’s OPF results with other evolutionary computation algorithms. Appendix: Table 16 lists the Jaya algorithm control variables for Cases 7–10 (Fig. 12). Table 12Comparison of Case 10 numerical results.AlgorithmFuel costPlossVDL-indexCTNR (Base Case)51,395.5706428.365891.255170.309940.7423 Jaya 4441.4980 9.7924 1.7192 0.2348 134.0962 TLBO44,4389.77001.74640.2353136.1989DE44,74710.30001.24950.2435135.8910PSO44,41810.00001.52780.2434135.0914TSA1544,519.0712.4732.69––SKH3345044.240710.6877–0.2857– Table 12Comparison of Case 10 numerical results.AlgorithmFuel costPlossVDL-indexCTNR (Base Case)51,395.5706428.365891.255170.309940.7423 Jaya 4441.4980 9.7924 1.7192 0.2348 134.0962 TLBO44,4389.77001.74640.2353136.1989DE44,74710.30001.24950.2435135.8910PSO44,41810.00001.52780.2434135.0914TSA1544,519.0712.4732.69––SKH3345044.240710.6877–0.2857– Comparison of Case 10 numerical results. Fig. 12Convergence characteristics of Case 10. Convergence characteristics of Case 10. IEEE 118-bus test system To evaluate the scalability and effectiveness of Jaya for large-scale power systems, Jaya was applied to the IEEE 118-bus system. This benchmark system consists of 54 generating units, two shunt reactors, twelve capacitors, 186 transmission lines, and nine tap-changing transformers. Detailed system parameters and operating limits of the control variables were adopted from standard benchmark data reported in41. Jaya was executed for 30 independent trials to ensure statistical reliability, and the best-performing solutions among these runs are reported in this section. The IEEE-118 test bus system is shown in Fig. 13. Fig. 13Single-line diagram of IEEE-118 Bus Test System. Single-line diagram of IEEE-118 Bus Test System. Case 11: Fuel cost minimization In Case 11, minimization of generation fuel cost was considered as the sole objective. The Jaya algorithm achieved the lowest operating cost of 129,216.3118 $/h, whereas TLBO, DE and PSO resulted in minimum costs of 129,240.3789 $/h,129,236.9901 $/h and 129,246.1098 $/h, respectively. A comparative analysis of the OPF results obtained using the Jaya algorithm and other recently reported approaches is provided in Table 13. The comparison clearly indicates that Jaya provides the most economical solution among all evaluated techniques, confirming its better optimization capability. The convergence characteristics of fuel cost for case 11 are shown in Fig. 14. Figure 15 shows a comparison of optimized fuel cost with various algorithms. Table 13Comparison of Case 11 numerical results.AlgorithmFuel costCTNR (Base Case)131,220.020890.1287 Jaya 129,216.3118 162.8871 TLBO129,240.3789166.9812DE129,236.9901168.1192PSO129,246.1098163.1231IMFO28131820.0–GPU-PSO41129,627.03–PSOGSA42129,733.58– Table 13Comparison of Case 11 numerical results.AlgorithmFuel costCTNR (Base Case)131,220.020890.1287 Jaya 129,216.3118 162.8871 TLBO129,240.3789166.9812DE129,236.9901168.1192PSO129,246.1098163.1231IMFO28131820.0–GPU-PSO41129,627.03–PSOGSA42129,733.58– Comparison of Case 11 numerical results. Fig. 14Convergence Characteristics of Case 11. Convergence Characteristics of Case 11. Fig. 15A comparative representation of the optimized fuel cost with different algorithms. A comparative representation of the optimized fuel cost with different algorithms. Statistical analysis Since stochastic methods are based on random behaviour in nature, running these methods several times independently for the same problem is necessary to achieve better results because the results can vary from one run to another. To perform a fair and unbiased comparison of the proposed algorithm with other competing algorithms, a one-way ANOVA test was used as a statistical tool to check the performance of the proposed optimizer algorithm. In this study, the value of the significance level is set to 0.05. The result, where the p-value is less than 0.05, shows that the difference observed in the algorithms is statistically significant and not due to any variation. Therefore, this test is important in proving the effectiveness and robustness of the suggested approach. Table 14 shows the result obtained from the one-way ANOVA test, which was conducted on all optimizers on the IEEE 118 bus system. The result shows that the obtained p-value is less than 0.05, indicating clearly that there is a difference between the algorithms. Moreover, from the better result, it is deduced that the Jaya algorithm is better than the others with a high level of statistical confidence. Figure 16 shows the one-way ANOVA analysis of the OPF algorithms on the IEEE 118 bus system. Table 14Statistical results of the one-way ANOVA test for the IEEE 118 bus system.SourceSSdfMSFProb > FGroups15710.935236.9615407.591.12256e−150Error39.41160.34Total15750.3119 Table 14Statistical results of the one-way ANOVA test for the IEEE 118 bus system.SourceSSdfMSFProb > FGroups15710.935236.9615407.591.12256e−150Error39.41160.34Total15750.3119 Statistical results of the one-way ANOVA test for the IEEE 118 bus system. Fig. 16One-way ANOVA comparison of OPF algorithms for IEEE 118 Bus System. One-way ANOVA comparison of OPF algorithms for IEEE 118 Bus System. Discussion The experimental results obtained for the OPF optimization problem show that the optimization approach can achieve the minimum value for the objective functions while maintaining stable convergence characteristics. The convergence characteristics show that the algorithm performs an initial wide range of exploration followed by gradual exploitation in the search space. This is important in solving OPF optimization problems due to the presence of several local optima in the solution space, due to generator limits, power balance conditions, and transmission limits. As can be seen from the convergence characteristics, near-optimal values are achieved in a small number of iterations. Another significant feature that can be noted from the above simulation result is the robustness of the algorithm for repeated trials. The statistical analysis carried out for multiple independent trials of the algorithm demonstrates that there is a relatively low variance in the objective function values. This demonstrates that the algorithm is not too sensitive to initializations and is capable of producing consistent solutions of high quality. This is a significant feature for OPF problems since they are typically of high dimensionality and are subject to equality and inequality constraints that are nonlinear in nature. The consistent performance of the algorithm demonstrates that it is maintaining a good balance between exploration and exploitation of the search space. The scalability of the suggested method was also examined by utilizing the IEEE-118 bus test system, which portrays a power network of considerable size with substantially more variables and operational constraints compared to other power systems of lower sizes. The results show that the algorithm retains its stable convergence characteristics even for this large-scale OPF problem. The capacity of the algorithm to deal with this expanded dimension of the search space points to the fact that it can handle power system optimization problems of considerable complexity. This is especially true for modern power systems that are characterized by the presence of numerous power networks and the integration of renewable energy sources. Although promising results are obtained as demonstrated within this study, some limitations are noted. Like most population-based metaheuristic optimization techniques, this algorithm necessitates repetitive objective function evaluations. This could result in higher computational time for large-scale systems. Moreover, although this algorithm has demonstrated promising results on the test systems, the results could be affected by some characteristics of the OPF problem, such as network size, number of control variables, or constraint level. Therefore, future research could be directed at improving the efficiency of this algorithm using hybrid optimization techniques or parallel search mechanisms. The overall findings of the presented work indicate that the suggested optimization method is a reliable solution for the OPF problems. The robust convergence characteristic of the suggested method, along with the statistically proven performance improvements, indicates the potential applicability of the suggested method for power system operation. Conclusions In order to achieve an economic operation of a modern power system, system operators use OPF solutions in operation, control and future expansion of system planning. The research paper presents the Jaya optimization tool as a revolutionary solution to overcome OPF problems for maintaining grid stability through improving the voltage profile of load buses and efficient loss reduction of the power system. The Jaya method has been used on IEEE 30-bus, IEEE-57-bus, and IEEE 118-bus test systems to assess its effectiveness and superiority. The numerical results achieved by the Jaya algorithm were compared with recent competitive global optimization methods. The results were particularly noteworthy as all inequality constraints were fully satisfied during the optimization process. These bounds include outputs of active and reactive power generation, thermal limits of lines, voltage magnitudes of load buses, and line loading. The research demonstrates that Jaya achieves higher accuracy and efficiency in most cases in comparison to established algorithms, TLBO, DE, and PSO, with evidence from modern literature investigations. We further acknowledge that the weighted sum method (WSM), while simple and widely used, may not always capture these trade-offs effectively—especially when objectives conflict. Future research will explore Pareto-based multi-objective optimization methods to more accurately represent trade-offs inherent to OPF problems.