Opposition-Based Learning Harmony Search Algorithm Solving Unconstrained Optimization Problems

Abstract

Recently, a new meta-heuristic optimization algorithm–harmony search (HS) was developed,which imitates the behaviors of music improvisation. Although several variants and an increasing number of applications have appeared, one of its main difficulties is how to enhance diversity and prevent it trapped into local optimal. This paper develops an opposition-based learning harmony search algorithm (OLHS) for solving unconstrained optimization problems. The proposed method uses the best harmony to play pitch adjustment, and bring the concept of opposition-base learning into improvisation, which enlarged the algorithm search space. Besides, we design a new parameter setting strategy to directly tune the parameters in the search process, and balance the process of exploitation and exploration. Numerical results demonstrate that the proposed algorithm performs much better than the existing HS variants in terms of the solution quality and the stability.