基于云遗传的混合混沌粒子群算法研究

为了解决PSO算法种群随机初始化遍历性差、易陷入早熟收敛以及不具备全局收敛性的缺点,提出一种基于云遗传的混合混沌粒子群优化算法,使用均匀性更优的无限折混沌叠映射实现粒子初始化,通过自适应云算子、改进的Metropolis接受准则以及动态调整粒子集规模等策略,实现了云遗传算法和PSO算法的协同,最后通过全局收敛性证明、时间复杂度和实验分析,验证了算法的有效性。

Research on Hybrid Chaotic Particle Swarm Optimization Algorithm Based on Cloud Theory-based Genetic

In order to solve the PSO population randomly initialized ergodicity is poor, easy to fall into premature convergence and does not have the disadvantages of global convergence, this paper puts forward a method of a cloud genetic algorithm hybrid chaos particle swarm optimization algorithm based on, using uniformity better map with infinite collapses to achieve particle initialization, cloud adaptive operator, improving the metropolis acceptance criteria and dynamic adjustment of the particle set scale strategy, to achieve synergies cloud genetic algorithm and PSO algorithm, the global convergence results prove, time complexity and experimental analysis, to verify the effectiveness of the algorithm