双层粒子群算法及应用于压力容器设计

Double-layer particle swarm optimization algorithm and its application to pressure vessel design

针对粒子群算法控制参数调整策略,提出一种双层粒子群优化算法(DBPSO);DBPSO有内外2层粒子,内层粒子是优化问题的解,对优化问题进行寻优;外层粒子由内层粒子的参数组成,通过协进化策略,实现参数自适应调整。利用5个常用标准测试函数对DBPSO的寻优性能进行测试,结果表明,其寻优性能优于基本PSO算法与一类改进的PSO算法(NMPSO)。最后,将DBPSO用于压力容器模型参数优化,取得了满意结果。

A novel parameter adaptive strategy for PSO is proposed and named as DBPSO. DBPSO has two groups of particle swarm. The inner particle swarm is the solution of the optimized problem and employed to search the optimal solution. The outer particle swarm consists of the parameters of the inner particle swarm and is applied to optimize the parameters by using co-evolving strategy. Five standard test functions are used to test the performance of the DBPSO and the results show that the performance of the DBPSO is better than the standard PSO and the NMPSO. Further, the optimization problem of the pressure vessel model is tackled by DBPSO and the satisfactory solution is obtained.