The hybrid strategies of harmony search in optimization problem solving

The remarkable capability of overcoming the shortcomings of NIC individual algorithms without losing their advantages makes the hybrid NIC techniques superior to the stand-alone ones and become a dominant research topic in the field of NIC methods. This hybridization mechanism has been demonstrated to achieve great improvement in the performance of different optimization issues. In this chapter, four hybrid optimization methods based on the HS combined with niching technique, opposition-based learning (OBL), clonal selection algorithm, and cultural algorithm are presented. The simulation results show that they can considerably outperform the original HS method in the applications of multimodal optimization, optimization of high-dimensional benchmark functions, minimization of weight of a tension/compression spring, optimal Sugeno fuzzy classification systems, and optimal design of welded beam, gear train, pressure vessel, and permanent magnet direct-driven wind generator.