中国机械工程 ›› 2010, Vol. 21 ›› Issue (20): 2482-2487.

• 机械基础工程 • 上一篇    下一篇

支持扩展QoS约束的动态虚拟企业伙伴选择模型

张以文1,2;倪志伟1;王力1;李志东1
  

  1. 1.合肥工业大学过程优化与智能决策教育部重点实验室,合肥,230009
    2.安徽大学,合肥,230039
  • 出版日期:2010-10-25 发布日期:2010-10-29
  • 基金资助:
    国家高技术研究发展计划(863计划)资助项目(2007AA04Z116);国家自然科学基金资助项目(70871033) 
    National High-tech R&D Program of China (863 Program) (No. 2007AA04Z116);
    National Natural Science Foundation of China(No. 70871033)

Model of Partner Selection for Dynamic Virtual Enterprise with Extended QoS Constraints

Zhang Yiwen1,2;Ni Zhiwei1;Wang Li1;Li Zhidong1
  

  1. 1.Key Laboratory of Process Optimization and Intelligent Decision-making,Ministry of Education,Hefei University of Technology,Heifei,230009
    2.Anhui University,Hefei,230039
  • Online:2010-10-25 Published:2010-10-29
  • Supported by:
     
    National High-tech R&D Program of China (863 Program) (No. 2007AA04Z116);
    National Natural Science Foundation of China(No. 70871033)

摘要:

现有Web服务架构为动态虚拟企业系统集成提供了良好的解决方案,但不能很好地支持扩展QoS(QoSextr)约束的动态虚拟企业伙伴选择。提出一种基于Web服务架构的动态虚拟企业伙伴选择扩展模型(DVEPSEM),对传统QoS进行了扩展,利用QoS监控系统对企业QoSextr进行实时监控和处理,同时发挥本体语义优势,建立企业服务分类集和领域本体间的映射,以提高企业服务发现的准确性和效率。基于DVEPSEM模型提出一种改进的自适应遗传伙伴选择算法(IAGA),实现QoSextr约束和大规模的动态虚拟企业伙伴选择。实验表明,IAGA算法能够较大程度地减少进化代数,提高伙伴选择的准确性。

关键词:

Abstract:

The present Web service architecture provides a good solution for the
integration of dynamic VE system,but it is still not good enough as to support the selection of dynamic VE partner with QoSextr constraints.So here an extended model(DVEPSEM) of dynamic VE partner selection was presented, By using QoS monitoring system, it extended the traditional QoS, monitoring and processing the enterprise QoSextr in a real time sense.Meanwhile,based on ontology semantics, it established the mapping between services category sets and domain ontology,thus the accuracy and efficiency in enterprise service discovery were achieved. An improved AGA partner selection algorithm based on DVEPSEM was proposed.The problem of partner selection with extended QoSextr constraints and in the large extended dynamic VE was resolved.Experimental
 results show that IAGA algorithm can further reduce the evolution generation and improve the accuracy in partner selection.

Key words:

中图分类号: