Improvement of Population Diversity of Meta-heuristics Algorithm Using Chaotic Map

Particle swarm optimization (PSO) is a global optimization and nature-inspired algorithm known for its good quality and easily applied in various realworld optimization challenges. Nevertheless, PSO has some weaknesses such as slow convergence, converging prematurely and simply gets stuck at local optima. This study aims to solve the problem of deprived population diversity in the search process of PSO which causes premature convergence. Therefore, in this research, a method is brought to PSO to keep away from early stagnation which explains premature convergence. The aim of this research is to propose a chaotic dynamic weight particle swarm optimization (CHPSO) wherein a chaotic logistic map is utilized to enhance the populace diversity within the search technique of PSO with the aid of editing the inertia weight of PSO in an effort to avoid premature convergence. This study additionally investigates the overall performance and feasibility of the proposed CHPSO as a function selection set of rules for fixing problems of optimization. 8 benchmark functions had been used to assess the overall performance and seek accuracy of the proposed (CHPSO) algorithms and as compared with a few other meta-heuristics optimization set of rules. The outcomes of the experiments show that the CHPSO achieves correct consequences in fixing an optimization and has established to be a dependable and green metaheuristics algorithm for selection of features.