基于混沌变异的自适应双粒子群优化

针对粒子群优化在解决高维优化问题时收敛性差、搜索效率不高的问题,在对粒子群优化算法收敛性分析的基础上,提出了混沌变异对极值进行扰动的方法,以增强算法摆脱局部最优解的能力.采用自适应惯性权重和局部邻域搜索保持较高的局部搜索性能,并结合双粒子群协同进化的方法,综合平衡优化算法的全局搜索和局部搜索能力.通过对4个典型测试函数进行的对比实验,表明了所提出的算法能大大提高粒子群优化的搜索效率和收敛精度.

Adaptive double particle swarms optimization algorithm based on chaotic mutation

Considering the bad convergence and search efficiency of particle swarm optimization(PSO)when applied to multi-dimensional optimization problem,an algorithm of adaptive chaotic variation particle swarm optimization(ACPSO) is proposed after analyzing the convergence of PSO.The proposed algorithm uses the technique of double subpopulation cooperative evolution.One subpopulation uses a new disturbed extremum method of chaotic mutation to improve the global convergence ability,and the other uses the adaptive in...