Improving Harmony Search by Zipf Distribution

Harmony search (HS) is a promising metaheuristic algorithm inspired by music improvisation process for various hard optimization problems. HS has attracted attention of researchers from different areas because it is easy to implement and can be applied to different optimization problems. in this paper, we present a novel method to improve the result of HS called Zipf harmony search (ZHS)â"by using the Zipf distribution to reference the harmonies within the harmony memory. We want to balance the intensification and diversification. Experimental results show that ZHS can provide better results than simple HS in the higher number of dimensions in numerical function optimization problem.