Locating Multiple Optima via Brain Storm Optimization Algorithms

Abstract

Locating multiple optima/peaks in a single run and maintaining these found optima until the end of a run is the goal of multimodal optimization. Three variants of brain storm optimization (BSO) algorithms, which include original BSO algorithm, BSO in objective space algorithm with Gaussian random variable, and BSO in objective space algorithm with Cauchy random variable, were utilized to solve multimodal optimization problems in this paper. The experimental tests were conducted on eight benchmark problems and its applications in seven nonlinear equation system problems. The performance and effectiveness of various BSO algorithms on solving multimodal optimization problems were validated based on the experimental results. The conclusions could be made that the global search ability and solutions maintenance ability of an algorithm needs to be balanced simultaneously on solving multimodal optimization problems.