中国机械工程 ›› 2015, Vol. 26 ›› Issue (8): 1048-1057.

• 智能制造 • 上一篇    下一篇

面向云制造系统复杂任务请求的服务组合优化框架

刘波;张自力   

  1. 西南大学,重庆,400715
  • 出版日期:2015-04-25 发布日期:2015-09-10
  • 基金资助:
    国家科技支撑计划资助项目(2012BAD35B08);中央高校基本科研业务费专项资金资助项目(XDJK2014C042,SWU113028) 

Framework of Complex Task Oriented Service Composition and Optimization in Cloud Manufacturing Systems

Liu Bo;Zhang Zili   

  1. Southwest University,Chongqing,400715
  • Online:2015-04-25 Published:2015-09-10
  • Supported by:
    The National Key Technology R&D Program(No. 2012BAD35B08);Fundamental Research Funds for the Central Universities
    ( No. XDJK2014C042,SWU113028 )

摘要:

为破解因多任务、强QoS约束及任务过量等因素导致云制造系统组合效果不佳的问题,研究并提出了面向复杂任务请求的全局优化策略框架。该框架以多任务全局优化、组合服务捆绑与共享为基本原则,提出了“单组合执行每任务”、“多组合执行每任务”及“多组合执行多任务”三种组合模式并建立了问题模型,最后利用基于混合算子的矩阵编码遗传算法予以实现。实验结果表明,该框架能高效、高质量响应云制造系统的复杂任务请求。

关键词: 云制造, 复杂任务, 服务组合优化, 全局策略

Abstract:

To circumvent the problems of SCO in the typical scenarios of complex multi-task requests, severe QoS constraints on tasks, and services shortage relative to tasks in cloud manufacturing systems, a framework of global optimal strategy for complex task oriented services composition(GOS-CTOSC) was presented. In this framework, the principles of multi-task oriented holistic optimization and the ideas of composite services binding and sharing were proposed to eliminate the drawbacks of traditional SCO approaches. Based on the principle, the composition patterns of  “each composition for each task”, “multi-composition for each task” and  “multi-composition for multi-task”  were designed in the framework, and the related problem models were also formulated. The implementation and evaluation of the framework were conducted in a prototype system, by means of the hybrid-operator based matrix coded genetic algorithm. The experimental results indicate the presented framework is sound performance-wise.

Key words: cloud manufacturing, complex task, service composition and optimization(SCO), global strategy

中图分类号: