A novel approach for optimization in dynamic environments based on modified cuckoo search algorithm

作者：Fouladgar Nazanin^*; Lotfi Shahriar

来源：Soft Computing, 2016, 20(7): 2889-2903.

DOI：10.1007/s00500-015-1951-7

• Swarm intelligence
• Dynamic optimization problems
• Multi-swarm
• Cuckoo search
• Moving peak benchmark

Cuckoo search algorithm is one of the famous algorithms in the area of swarm intelligence algorithms. It has been supplied widely for solving static optimization problems. However, it should be considered that a great number of optimization problems in the real world, are in the form of dynamic optimization problems. In fact, the algorithms which have been implemented for static environments are not able to solve problems in dynamic environments. In this paper, a novel multi-swarm algorithm based on modified cuckoo search algorithm (MCSA) has been proposed to find and track the optimum (optima) of the problem space in dynamic environments. Each swarm performs optimization process based on MCSA. Also, a deactivation mechanism has been utilized to improve the efficiency of this approach. Finally the proposed algorithm has been tested on moving peak benchmark, one of the most well-known benchmarks of this domain, and compared with several prominent algorithms in this area. The results indicate the superiority of this approach.

• 出版日期2016-7