华南理工大学学报(自然科学版) ›› 2015, Vol. 43 ›› Issue (1): 105-110.doi: 10.3969/j.issn.1000-565X.2015.01.017

• 计算机科学与技术 • 上一篇    下一篇

云制造系统中基于粒子群优化的多任务调度

武善玉1,2 张平1† 李方1   

  1.  1. 华南理工大学 计算机科学与工程学院, 广东 广州 510006 ; 2. 广东石油化工学院 计算机与电子信息学院, 广东 茂名 525000
  • 收稿日期:2014-05-09 修回日期:2014-07-04 出版日期:2015-01-25 发布日期:2014-12-01
  • 通信作者: 张平(1964-),男,教授,博士生导师,主要从事智能软件和机器人系统研究 E-mail:pzhang@scut.edu.cn
  • 作者简介:武善玉(1974-),女,在职博士生,广东石油化工学院讲师,主要从事面向服务架构、信息物理系统架构、资源优化、任务调度研究 .E-mail : wu_shyjin@163.com
  • 基金资助:

    广东省 - 教育部产学研结合项目( 2012B091100444 );华南理工大学中央高校基本科研业务费专项资金面上项目(2013ZM0091 );广州市科技计划项目( 2013Y2-00100 )

Multi-Task Scheduling Based on Particle Swarm Optimization in Cloud Manufacturing Systems

Wu Shan-yu1,2 Zhang PingLi Fang1   

  1. 1. School of Computer Science and Engineering , South China University of Technology , Guangzhou 510006 , Guangdong , China ;2. Guangdong University of Petrochemical Technology , Maoming 525000 , Guangdong , China 
  • Received:2014-05-09 Revised:2014-07-04 Online:2015-01-25 Published:2014-12-01
  • Contact: 张平(1964-),男,教授,博士生导师,主要从事智能软件和机器人系统研究 E-mail:pzhang@scut.edu.cn
  • About author:武善玉(1974-),女,在职博士生,广东石油化工学院讲师,主要从事面向服务架构、信息物理系统架构、资源优化、任务调度研究 .E-mail : wu_shyjin@163.com
  • Supported by:
    Supported by the Production , Education and Research Cooperative Project of Guangdong Province and the Mi-nistry of Education ( 2012B091100444 )

摘要: 为解决云制造系统的同类型多任务调度问题,建立了该问题的数学模型,提出了一种离散粒子群遗传混合算法,以所有任务的总完成时间及成本最优为目标进行求解 . 该算法采用整数编码方法建立粒子的位置矢量与服务分配的映射关系,在采用标准粒子群算法更新粒子位置时,引入了遗传算法的交叉和变异操作思想,使用 4 种方法按条件“逐级叠加”的方式对粒子位置进行更新,以保证种群的多样性 . 算例仿真结果表明,该算法是有效的且具有较高的执行效率 .

关键词: 云制造, 多任务调度, 面向服务架构, 服务组合, 多目标优化, 粒子群优化, 离散粒子群遗传混合算法

Abstract: In order to implement the scheduling of multiple tasks with the same type in cloud manufacturing systems, a mathematical model is established and is solved by using a discrete particle swarm-genetic hybrid algorithm with two objectives, namely the least total completing time and the least cost of all tasks being considered simultaneously. The hybrid algorithm employs integer coding method to establish the mapping between particle location matrix and service allocation scheme, and introduces the crossover and mutation idea of genetic algorithm to update particle swarm position with four formulas being conditionally used in a progressive and overlaying way, and thus the diversity of groups is ensured effectively. Simulated results indicate that the proposed algorithm is of high effectiveness and execution efficiency.

Key words: cloud manufacturing, multi-task scheduling, service-oriented architecture, service combination, multi-objective optimization, particle swarm optimization, discrete particle swarm-genetic hybrid algorithm

中图分类号: