基于均匀设计的粒子群优化算法参数设定

Parameter Establishment of Particle Swarm Optimization Algorithm Based on Uniform Design

粒子群优化算法的参数设置通常是依靠经验和试验来确定,造成试验工作量大且难以得到最优的参数组合,影响了算法的使用。通过将粒子群优化算法基本模型的参数设定问题描述成均匀设计中多因素多水平的试验设计,从而能够用较少的试验很快设定算法参数的取值。仿真试验表明该方法的可行性和有效性。

In general the parameters of particle swarm optimization algorithm （PSO） are determined by experience and experiment. This leads to heavy work load and difficult to obtain the optimal combination of the parameters and affects the use of PSO. Uniform design method was used to convert the problem of parameter establishment into the experimental design of multi--factor and multilevel and reduce the work load of experiment greatly of simulation. Simulation results for the benchmark problems show that the proposed method is feasible and valid.