Downloads: 1 | Views: 99 | Weekly Hits: ⮙1 | Monthly Hits: ⮙1
Comparative Studies | Computer Engineering | India | Volume 12 Issue 6, June 2023 | Rating: 5.3 / 10
Comparative Study of Evolutionary Algorithms
Vansh Khera
Abstract: Evolutionary algorithms (EAs) are widely used optimization techniques inspired by the principles of biological evolution. They mimic the process of natural selection and genetic variation to iteratively search for optimal solutions to complex problems. This comparative study aims to analyze and compare the performance of four popular evolutionary algorithms: Harris Hawk Optimization (HHO), Genetic Algorithms (GA), Differential Evolution (DE), and Particle Swarm Optimization (PSO). The study begins by providing a comprehensive overview of each algorithm, highlighting their key characteristics and underlying principles. HHO is a recently proposed algorithm inspired by the hunting behavior of Harris hawks. GA is a classic algorithm that utilizes genetic operators such as crossover and mutation to explore the solution space. DE is a population-based algorithm that utilizes vector arithmetic to generate new candidate solutions. PSO is a swarm intelligence algorithm where particles move through the search space to find optimal solutions based on their own experience and the influence of neighboring particles. To conduct a fair comparison, a set of benchmark functions is selected to evaluate the algorithms' performance in terms of convergence speed and solution quality. These benchmark functions encompass various optimization challenges, including multimodal, unimodal, and high-dimensional problems. The algorithms are implemented and executed using standardized parameters and termination criteria. The experimental results provide insights into the strengths and weaknesses of each algorithm. The comparative analysis considers factors such as convergence speed, global versus local optima exploration, robustness, and scalability. The results reveal that HHO demonstrates superior convergence speed and exploration capability for multimodal problems. GA showcases excellent performance in searching for global optima in unimodal problems. DE exhibits a balanced performance across different problem types, while PSO demonstrates effectiveness in dealing with high-dimensional optimization problems. The study concludes with a discussion on the implications of the findings and potential directions for future research. The comparative analysis presented in this study serves as a valuable resource for researchers and practitioners in selecting appropriate evolutionary algorithms based on the specific characteristics of optimization problems they encounter.
Keywords: evolutionary algorithm
Edition: Volume 12 Issue 6, June 2023,
Pages: 836 - 840