Research on Improved QPSO Algorithm Based on Cooperative Evolution with Two Populations
This paper presents a Cooperative Evolutionary Quantum-behaved Particle Swarm Optimization (CEQPSO) algorithm with two populations to tackle the shortcomings of the original QPSO algorithm on premature convergence and easily trapping into local extremum. In the proposed CEQPSO algorithm, the QPSO algorithm is used to update individual and global extremum in each population; the operations of absorbing and cooperation are used to exchange and share information between the two populations. The absorbing strategy makes the worse population attracted by the other population with a certain probability, and the cooperation strategy makes the two populations mutually exchange their respective best information. Moreover, when the two populations are trapped into the same optimum value, Cauchy mutation operator is adopted in one population. Four benchmark functions are used to test the performance of the CEQPSO algorithm at a fixed iteration, and the simulation results showed that the proposed algorithm in this paper had a better optimization performance and faster convergence rate than PSO and QPSO algorithms.
Cooperative Evolution Quantum-behaved Particle Swarm Optimization Cauchy Mutation
Longhan Cao Shentao Wang Xiaoli Liu Rui Dai Mingliang Wu
Key Laboratory of Control Engineering, Chongqing Institute of Communication,Chongqing, 400035 Key La Key Laboratory of Control Engineering, Chongqing Institute of Communication,Chongqing, 400035
国际会议
无锡
英文
93-100
2010-09-17(万方平台首次上网日期,不代表论文的发表时间)