论文部分内容阅读
微粒群算法(PSO)是一种随机群体优化算法,相对于遗传算法等其它的进化算法,它模型简单、操作参数少、智能程度高、运算速度快,已受到许多相关领域学者的关注与研究。但是,标准微粒群算法在寻优过程中往往陷入局部最优解,而不是全局最优解。在研究均匀设计与惰性变异的基础上,提出了改进的微粒群算法(UMPSO)。该算法利用均匀设计的思想来确定算法的初始粒子,以使其均匀分布于解空间,从而使算法以更高的概率、更快的速度找到全局最优解;在进化过程中,对惰性粒子以概率为1进行随机变异,则能够更好地保证微粒群的多样性。仿真结果表明,与标准的PSO相比,UMPSO的寻优精度更高、寻优速度更快。
Particle Swarm Optimization (PSO) is a stochastic population optimization algorithm. Compared with other evolutionary algorithms, such as genetic algorithm, its simple model, low operating parameters, high intelligence and fast computing speed have attracted the attention of many scholars in related fields . However, the standard particle swarm optimization often falls into the local optimal solution rather than the global optimal solution. On the basis of researching uniform design and inert variation, an improved Particle Swarm Optimization (UMPSO) is proposed. The algorithm uses the idea of uniform design to determine the initial particle of the algorithm so that it can be uniformly distributed in the solution space, so that the algorithm can find the global optimal solution with higher probability and faster speed. In the process of evolution, Random mutation with a probability of 1 can better ensure the diversity of particle swarm. Simulation results show that, compared with the standard PSO, UMPSO has higher precision and faster optimization.