Engineering Science and Technology, an International Journal (May 2023)
A modified seahorse optimization algorithm based on chaotic maps for solving global optimization and engineering problems
Abstract
Metaheuristic optimization algorithms are global optimization approaches that manage the search process to efficiently explore search spaces associated with different optimization problems. Seahorse optimization (SHO) is a novel swarm-based metaheuristic optimization method inspired by certain behaviors of sea horses. The SHO algorithm mimics the movement, hunting, and breeding behavior of sea horses in nature. Chaotic maps are effectively used to improve the performance of metaheuristic algorithms by avoiding the local optimum and increasing the speed of convergence. In this study, 10 different chaotic maps have been employed for the first time to produce chaotic values rather than random values in SHO, increasing the performance of the method. The purpose of using chaotic maps that generate chaotic values to their random values in SHO is to increase the convergence speed of the original SHO algorithm and avoid the local optimum. 33 different benchmark functions, consisting of unimodal, multimodal, fixed-dimension multimodal, and CEC2019, have been utilized to assess the performance of Chaotic SHO (CSHO), which is first introduced in this study. In addition, the proposed CSHO has been compared with four metaheuristic algorithms in the literature, namely Sine Cosine Algorithm, Salp Swarm Algorithm, Whale Optimization Algorithm, and Particle Swarm Optimization. Statistical analyses of the obtained results have been also performed. The proposed CSHO is then implemented to 4 different real-world engineering design problems, including the welded beam, pressure vessel, tension/compression spring, and speed reducer. The results obtained with CSHO are compared with popular metaheuristic methods in the literature. Experimental results show that it gives successful and promising results compared to the original SHO algorithm.