Adaptive Particle Swarm Optimization with Gaussian Perturbation and Mutation

The particle swarm optimization (PSO) is a wide used algorithm, which yet suffers from trapping in local optimum and the premature convergence. Many studies have proposed improvements to address drawbacks above. Most of them implemented single strategy for one problem or fixed neighborhood structure during whole search process. To further improve PSO performance, we introduced simple but effective method, named adaptive with Gaussian perturbation mutation (AGMPSO), consisting three strategies. are incorporated promote exploration exploitation capability, while ensure dynamic implement former two strategies, guarantee balance searching ability accuracy. Comparison experiments AGMPSO existing variants solving 29 benchmark functions CEC 2017 test suites suggest that, despite simplicity architecture, obtains high convergence accuracy significant robustness proven by conducted Wilcoxon’s rank sum test.