Self-adaptive harmony search algorithm for optimization

Recently, a new meta-heuristic optimization algorithm – harmony search (HS) with continuous design variables was developed. This algorithm is conceptualized using the musical improvisation process of searching for a perfect state of harmony. Although several variants and an increasing number of applications have appeared, one of its main difficulties is how to select suitable parameter values. In this paper, we used the consciousness (i.e., harmony memory) to automatically adjust parameter values. In addition, the pseudo-random number generator is also replaced by the low-discrepancy sequences for initialization of the harmony memory. Finally, the experimental results revealed the superiority of the proposed method to the original HS and recently developed variants.