Computer Engineering and Applications ›› 2020, Vol. 56 ›› Issue (13): 54-62.DOI: 10.3778/j.issn.1002-8331.1906-0038

Previous Articles     Next Articles

Improved Dynamic Dual Adaptive PSO Algorithm Based on Theory of Co-evolution

SONG Mei, GE Yuhui, LIU Jusheng   

  1. Business School, Shanghai University of Science and Technology, Shanghai 200093, China
  • Online:2020-07-01 Published:2020-07-02



  1. 上海理工大学 管理学院,上海 200093


To explore the problem of the PSO algorithm easy to fall into the local optimum and the birth defect, under certain error tolerance, this paper references related characteristics of the theory of co-evolution such as the characteristic of the dynamic, the nonlinearity of the system, the co-evolution of the individual and the environment and the self-adaptability of the individual to design a kind of improved algorithm. Firstly, it uses the Feigenbaum iteration to calculate the initial chaotic value of the population. Secondly, it uses the nonlinear and the adaptive strategy to calculate the self-learning factor, social learning factor and the inertia weight. Finally, it does a simulation on single-mode and multi-modal Benchmark functions, in the meanwhile, it does a comparison with the other five algorithms. The simulation results show that the DDAPSO algorithm has better performance than other algorithms in the aspect of solving precision, optimizing efficiency and stability. It has the advantage of strong ability to find the global optimal solution and a wide application prospect in the future.

Key words: theory of co-evolution, dynamic dual adaptive, Feigenbaum iteration, adaptive adjustment, PSO algorithm



关键词: 协同进化理论, 动态双重自适应, Feigenbaum迭代, 自适应调整, PSO算法