基于多Agent 可互操作知识化制造动态自适应调度策略.pdfVIP

第28 卷 第2 期 控 制 与 决 策 2013 年 2 月 Vol. 28 No. 2 Control and Decision Feb. 2013 文章编号: 1001-0920 (2013) 02-0161-08 基于多Agent 可互操作知识化制造动态自适应调度策略 汪浩祥, 严洪森 (东南大学 a. 自动化学院，b. 复杂工程系统测量与控制教育部重点实验室，南京210096) 摘 要: 针对知识化制造系统生产环境的不确定性, 构建一个基于多Agent 可互操作的知识化动态调度系统. 该系统 对各种调度问题采用具有一系列问题特征的知识表示, 利用Agent 技术构建基于问题的功能模块, 提出一种基于改 进-学习算法(WSQ) 的自适应调度机制, 以此指导设备Agent 在动态环境下的调度策略选择. 通过对其进行复杂性 分析和仿真实验, 验证了该控制策略的有效性. 该系统具有自适应和自学习特征, 具有高度智能化和可互操作性. 关键词: 知识化制造系统；动态调度；多Agent 技术；-学习 中图分类号: TH165 文献标志码: A Interoperable dynamic adaptive scheduling strategy in knowledgeable manufacturing based on multi-agent WANG Hao-xiang, YAN Hong-sen (a. School of Automation，b. Key Laboratory of Measurement and Control of CSE ，Ministry of Education ，Southeast University ，Nanjing 210096 ，China ．Correspondent ：WANG Hao-xiang ，E-mail ：whx39@) Abstract ：：：Aiming to the uncetainty of production enviroment in knowledgeable manufacturing system, an interoperable knowledgeable dynamic scheduling system based on multi-agent is built, in which a knowledge representation with a series of problem characteristics for various scheduling problems is adopted and problem-based function modules are constructed by using agent technology. An adaptive scheduling mechanism based on modified -learning agrorithm known as weighted subordination based -learning(WSQ) is proposed for guiding the equipment agent to select scheduling strategy in a dynamic enviroment. By the analysis of its complexity and simulation experiments, the results show the effectiveness of this strategy. The system is of adaptive and self-learning features, and high intelligence and interoperability. Key words ：：：kno