부산대학교 도서관

A novel hybrid Krill Herd (KH) and Cuckoo Search(CS) with Differential Evolution(DE), called DECSKH, is proposed for the global optimization. DECSKH is designed for enhancing the ability of the local search and increasing the individual diversity in the population. In DECSKH, all individuals are divided into two parts, denoted as part I and part II, respectively. Two different optimization algorithm combined with DE, DECS and DEKH, are incorporated into these two parts, respectively. DECSKH, therefore, is expected to obtain an efficient balance between exploration and exploitation. DECSKH can make all the individuals proceed to the true global optimum without introducing additional operators to the basic CS, KH and DE algorithms. The effectiveness of these improvements is tested by 18 standard benchmark functions. Based on the results, we can easily infer that DECSKH is more effective and efficient than the original CS, KH, DE and other approaches.