混沌映射的粒子群算法分析比较

Analysis and comparison of particle swarm optimization algorithms for chaotic mapping

为解决粒子群算法(Particle Swarm Optimization, PSO)易受初值影响、迭代后期早熟收敛、局部寻优结果不稳定等问题,提出混沌映射粒子种群初始化的方案。即在算法前期根据混沌映射初值敏感性、随机性等特征,基于6种混沌映射对粒子群算法分别进行初始化,进而增加种群多样性和解的覆盖。引入动态惯性权重系数,提高算法收敛速度。通过6个测试函数进行仿真实验,对比不同混沌映射数据结果,实验证明该算法能在不改变原有时间复杂度的基础上较好地提高算法的收敛速度和寻优精度。

In order to solve the problems of Particle Swarm Optimization(PSO) which is easily affected by initial value, premature convergence in the later stage of iteration, and unstable local optimization results, a chaotic mapping particle population initialization scheme is proposed.That is, in the early stage of the algorithm, according to the characteristics of initial value sensitivity and randomness of chaotic map, the particle swarm optimization algorithm is initialized based on six chaotic maps respectively, and then the coverage of population diversity and solution is increased.The dynamic inertia weight coefficient is introduced to improve the convergence speed of the algorithm.Through the simulation experiment of six test functions and comparing the results of different chaotic mapping data, the experiment shows that the algorithm can better improve the convergence speed and optimization accuracy of the algorithm without changing the original time complexity.