计算机工程 ›› 2008, Vol. 34 ›› Issue (14): 197-199.doi: 10.3969/j.issn.1000-3428.2008.14.070

• 人工智能及识别技术 • 上一篇    下一篇

面向产品优化配置的粒子群优化算法

俞文静,王建维,魏小鹏   

  1. (大连大学先进设计技术中心,大连 116622)
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2008-07-20 发布日期:2008-07-20

Improved PSO Algorithm for Product Optimization Configuration

YU Wen-jing, WANG Jian-wei, WEI Xiao-peng   

  1. (Center for Advanced Design Technology, Dalian University, Dalian 116622)
  • Received:1900-01-01 Revised:1900-01-01 Online:2008-07-20 Published:2008-07-20

摘要: 针对产品配置大规模、多约束、多目标及组合优化等特性,建立一种有效的配置模型,将复杂的产品优化配置问题转化为图的路径寻优问题。针对基本粒子群算法(PSO)的缺陷,将遗传原理、蚁群机制和模拟退火理论引入PSO算法,提出一种改进的PSO算法。根据产品优化配置问题的离散特点,对PSO算法进行离散化处理,重新定义粒子的位置和速度表示,确立这些量的运算规律和粒子运动方程。典型产品配置实例验证了提出的模型和算法的可行性。

关键词: 粒子群算法, 产品优化配置, 离散, 粒子

Abstract: Due to the combinatorial characteristics of product configuration, an effective product configuration model is established to transform the complicated problem into the path optimization problem. An improved hybrid intelligent algorithm, which unites the genetic principle, the ant colony mechanism and the simulation anneal theory into the original Particle Swarm Optimization(PSO), is adopted. Based on the discrete characteristic of the product configuration problem, the hybrid intelligent algorithm is transformed from the sequential to the discrete. An example is given to evaluate the effectiveness of the model and the algorithm.

Key words: Particle Swarm Optimization(PSO), product optimization configuration, discrete, particle

中图分类号: