Butterfly mating optimization

Abstract

This paper presents a novel swarm intelligence algorithm named as Butterfly Mating Optimization (BMO) which is based on the mating phenomena occurring in butterflies. The BMO algorithm is developed with novel concept of dynamic local mate selection process which plays a major role in capturing multiple peaks for multimodal search spaces. This BMO algorithm was tested on 3-peaks function and various convergence plots were drawn from it. Also, BMO was tested on other benchmark functions to check and discuss thoroughly its capability in terms of capturing the local peaks. Various comparisons were made between BMO and GSO, a recent swarm algorithm for multimodal optimization problems.BMOwas also tested on a function with varying dimensionality at higher level. Finally based on various assumptions through simulations, possible future work is discussed.