Improving Brain Storm Optimization Algorithm via Simplex Search

Through modeling human's brainstorming process, the brain storm optimization (BSO) algorithm has become a promising population based evolution algorithm. However, BSO is often good at global exploration but not good enough at local exploitation, just like most global optimization algorithms. In this paper, the Nelder-Mead's Simplex (NMS) method is adopted in a simple version of BSO. Our goal is to combine BSO's exploration ability and NMS's exploitation ability together, and develop an enhanced BSO via a better balance between global exploration and local exploitation. Large number of experimental results are reported, and the proposed algorithm is shown to perform better than both BSO and NMS.