粒子群参数自适应调整的优化设计

摘要:

在分析粒子群优化原理基础上，引入模拟退火机制以一定的概率对部分粒子的速度及位置执行更新操作，建立了粒子群惯性量权重因子及学习因子的模糊逻辑控制器以实现粒子群参数的自适应调整，从而提高优化算法的收敛速度及获得全局解的能力。通过运用常规优化方法、遗传算法及参数自适应调整的粒子群优化方法对起重机结构主梁截面优化设计对比可知：采用粒子群参数调整的优化方法具有自适应能力强、计算效率高及优化设计精度高等优点。

关键词:粒子群优化;自适应参数调整;模糊逻辑控制器;模拟退火算法

Abstract:

This paper introduced the simulated annealing mechanism into the particle swarm optimization. In the method, only a part of particles were involved in the speed and location renewal operation with certain probability. Two fuzzy logic controllers about inertia and learning factor were built in order to improve convergence speed and obtain global optimal solution. The optimization results of the main beam cross-section for crane structure from conventional optimization method, genetic algorithm and particle swarm optimization, were compared with one another. The comparison analysis indicates that the proposed particle swarm optimization method based on fuzzy logic parameter adjusting has advantages such as better self-adaptive capacity, higher computation efficiency and design accuracy.  

Key words:粒子群优化;自适应参数调整;模糊逻辑控制器;模拟退火算法