Abstract:
A new method of adjusting inertial weight adaptively with diversity feedback control is proposed. The diversity of swarm is maintained especially in the early phase of iterations. The risk of premature convergence is reduced and the precision of the optimum is improved remarkably, the influence of the swarm size on an optimum is weakened. The efficiency of the algorithm is verified by the simulation results of three benchmark functions and the comparison with two adaptive inertial weight Particle Swarm Optimization(PSO) algorithms.
Key words:
Particle Swarm Optimization(PSO),
diversity,
inertial weight
摘要: 利用粒子群多样性的反馈信息,给出带有粒子群多样性测度反馈控制的新惯性权值动态自适应调节方法,有效地维持进化初期的种群多样性,降低粒子群优化算法在进化初期发生早熟的风险,提高最优化解的精度,减小种群规模对优化精度的影响。几个典型函数的仿真结果以及与2种典型的惯性权值调节粒子群算法的比较结果表明了算法的有效性。
关键词:
粒子群优化,
多样性,
惯性权值
CLC Number:
JIAO Wei; LIU Guang-bin. Particle Swarm Optimization Algorithm Based on Diversity Feedback[J]. Computer Engineering, 2009, 35(22): 202-204.
焦 巍;刘光斌. 基于多样性反馈的粒子群优化算法[J]. 计算机工程, 2009, 35(22): 202-204.