Novel self-adaptive harmony search algorithm for continuous optimization problems

This paper presents a novel harmony search algorithm for solving the global continuous optimization problems. In the proposed algorithm, a new version of improvisation scheme is proposed to generate candidate harmonies, where two control parameters, bandwidth distance (bw) and pitch adjusting rate (PAR), are either self-adaptive or automatically updated with respect to both the evolution of the search process and the useful information taken from the optimization problems with different features. Numerical results show that the proposed algorithm is very competitive in comparison with the compared HS algorithms.