A Niching Harmony Search Method for Multi-modal Optimization

Abstract

The Harmony Search (HS) method is an emerging meta-heuristic optimization algorithm. However, it is generally not so efficient in dealing with multi-modal optimization problems, in which instead of only a single optimum, multiple optima need to be found. In our paper, a novel HS method based on the niching technique (deterministic crowding), n-HS, is proposed and studied to overcome this shortcoming. A total of eight typical multi-modal optimization functions are used to demonstrate that the n-HS can successfully locate most of the global and local optima.