基于正态云模型的自适应变异量子粒子群优化算法

摘要

为提高量子粒子群算法的寻优能力，文中提出一种新的正态云模型自适应变异量子粒子群算法。该方法采用正态云模型优化策略，引入自身最差粒子和全局最差粒子，结合自身最优粒子和全局最优粒子自适应调整势阱中心位置与收缩-扩张系数，每次迭代后生成的新粒子，以一定概率采用正态云模型对粒子进行变异操作。最后标准函数极值优化的实验结果表明，该算法的单步迭代时间较长但优化能力较同类算法有大幅度提高。%To improve the ability of quantum particle swarm optimization algorithm, this paper proposes a new normal cloud model adaptive mutation quantum particle swarm optimization algorithm which uses normal cloud model optimization strategy, introduced its own worst particle and particle worst global, combined with own best particle and global best particle adaptive trap central location and contraction-expansion coefficient, After each iteration to generate new particles ,to a certain probability, using normal cloud model mutation particles. Experimental results show that the standard function extreme optimization, single iteration of the algorithm is a long time, but relatively similar algorithms, optimization capabilities have greatly improved.