基于云变异的云自适应粒子群算法

摘要

In order to overcome the faults of cloud particle swarm optimization algorithm, such as easy to premature convergence, and to improve the performance of optimization multi-modal function, a cloud adaptive particle swarm optimization algorithm based on cloud variation was proposed to overcome this problem. For the ratio of the global optimal value and particle fitness reflects the characteristics of the particle optimal difference, cloud model generator was used to adaptively adjust the inertia value of every particles, and normal clouds operator was used to realize variation operation of part particle and set the parameters of algorithm. Simulation results of typical test functions show that the proposed algorithm has fine capability of finding global optimum and much higher accuracy, improves the velocity of convergence, and is suitable for find the value of multi-modal function.%研究云粒子群优化算法问题,为了克服云粒子群优化算法易过早收敛的缺点和提高优化多峰函数的性能.提出了一种云变异的云自适应粒子群优化新算法,结合全局最优值和粒子适应度的比值体现出粒子优差的特点,利用正态云发生器自适应调整粒子个体惯性权重,并且对粒子位置进行了基于云模型的变异操作,合理的对粒子群各参数进行设置,典型测试函数仿真结果表明,改进优化算法能有效找出全局最优解,提高了收敛精度和收敛速度,且适宜于多峰值问题寻优,是一种可行而有效的优化方法.