本文已被：浏览 1543次 下载 3079次
Received:July 30, 2011 Revised:September 07, 2011
Received:July 30, 2011 Revised:September 07, 2011
中文摘要: 现代工业发展要求迅速、可靠地实现故障诊断。针对粒子群约简算法易陷入局部最优等问题，提出了一种多种群量子粒子群优化算法(MIQPSO)。该算法对量子粒子群算法进行分群，并通过接种疫苗，指导粒子朝更优化方向进化，提高了量子粒子群的收敛速度和寻优能力。利用UCI相关数据集，通过对Hu算法、粒子群算法、量子粒子群算法、多种群量子粒子群算法的粗糙集属性约简验证，结果表明，基于多种群量子粒子群优化的约简算法具有良好的约简效果。
中文关键词: 粒子群算法 多种群量子粒子群优化 粗糙集 属性约简
Abstract:Requirements of modern industryial development rapidly and reliably achieve the fault diagnosis. Against particle swarm algorithm for the reduction and other issues so easy to fall into local optimum problem,this paper aims to present the MIQPSO Algorithm. The quantum particle swarm algorithm for clustering by the MIQPSO Algorithm, and through vaccination, to guide the direction of the particle evolution towards more optimized, improve the convergence rates and optimization searching ability of the quantum particle swarm. The use of UCI data sets, and by Hu algorithm, particle swarm optimization, quantum particle swarm optimization, multi-species quantum particle swarm algorithm for rough set attribute reduction verification, the results show that the algorithm based on the quantum particle swarm optimization has good reduction effect on the reduction.
keywords: particle swarm optimization quantum-behaved particle swarm optimization rough set attribute reduction
文章编号： 中图分类号： 文献标志码：
基金项目:浙江省高校学科带头人资助项目(2007-209);学校培育基金(LZYA201003)
| Author Name | Affiliation |
| LI San-Bo | School of Mechanical Electronic and Information Engineering, Lishui Vocational & Technical College, Lishui 323000, China |
| Author Name | Affiliation |
| LI San-Bo | School of Mechanical Electronic and Information Engineering, Lishui Vocational & Technical College, Lishui 323000, China |
引用文本：
李三波.基于多种群量子粒子群优化的属性约简.计算机系统应用,2012,21(4):99-104
LI San-Bo.Attribute Reduction Based on Quantum-Behaved Particle Swarm Optimization with Multi- Swarm Algorithm.COMPUTER SYSTEMS APPLICATIONS,2012,21(4):99-104
李三波.基于多种群量子粒子群优化的属性约简.计算机系统应用,2012,21(4):99-104
LI San-Bo.Attribute Reduction Based on Quantum-Behaved Particle Swarm Optimization with Multi- Swarm Algorithm.COMPUTER SYSTEMS APPLICATIONS,2012,21(4):99-104
