A novel cuckoo search algorithm with multiple update rules

摘要

In this paper, we propose a novel cuckoo search algorithm with multiple update rules, referred to as a hybrid CS algorithm (HCS). In the presented approach, to overcome the mutual interference among dimensions and enhance the local search capability, two different one-dimensional update rules are integrated into CS framework for acquiring the candidate solutions. Moreover, using the characteristic of occasionally long jumps in Levy distribution, the proper selection between the one-dimensional update rules and Levy flight random walk is achieved by setting a limit value, so as to further enhance the exploration ability. The performance of the presented algorithm is then extensively investigated on 49 benchmark test functions including 11 common instances, 10 instances introduced in CEC 2005, and 28 instances presented in CEC 2013. The experimental results indicate that HCS algorithm is better than other CS variants in terms of solution accuracy and robustness, and it also outperforms the seven state-of-the-art intelligent algorithms.