ABSTRACT
Cultural Particle Swarm Optimization (CPSO) (Deng, 2016; Wu, 2010) algorithm is an intelligent algorithm that integrates cultural algorithm into particle swarm optimization algorithm. During the evolution of CPSO, the particles can be updated by tracking two goals, namely global extreme value and individual extreme value (Yan, 2012). The continuous iteration of present global optimal solution up to now form a trajectory, which will be stored and considered as global knowledge space for global iterative search. Thus, acquired knowledge is transmitted to the next generation, guiding the individuals towards perceived global optimal solution and providing a systematic method for self-evolution. At the same time, greater global searching capacity of algorithm is achieved through double evolution and mutual effect of PSO space and knowledge space (Qin, 2016). However, due to the local optimum deficiency of knowledge space in cultural algorithm, knowledge space cluster cannot achieve desirable effect during self-evolution, reducing influence on lower-layer main cluster space.
