Chaos-enhanced Cuckoo search optimization algorithms for global optimization

Abstract

Cuckoo search optimization algorithm is a biologically inspired optimization algorithm, which is widely used to solve many optimization problems. However, it has been empirically demonstrated to easily get trapped into local optimal solutions and cause low precision. Therefore, in this work, we propose five modified Chaos-enhanced Cuckoo search (CCS) optimization algorithms, in which chaotic sequences are utilized to enhance initialized host nest location, change step size of Le´vy flight and reset the location of host nest beyond the boundary. These five CCS algorithms are denoted by CCS1 (with Logistic map), CCS2 (with tent map), CCS3 (with Gauss map), CCS4 (with Sinusoidal iterator) and CCS5 (with Circle map) respectively. We test our algorithms in two function groups, denoted by Group A and Group B, respectively. In Group A, which consists of four Unimodal and five simple Multimodal functions, we compare the performance of five CCS algorithms and the standard CS. The numerical results show that the novel algorithm enhances the performance of the basic Cuckoo search optimization algorithm, and CCS3 achieves the best performance. In Group B, which is derived from CEC2013 test problems, we test three optimization algorithms (CCS3, CLPSO and TCPSO). The numerical results show that the CCS3 algorithm has better performance than others.