基于配置的可重构制造系统建模与优化研究
文献类型：学位论文
| 作者 | 任思成 |
| 学位类别 | 工学博士 |
| 答辩日期 | 2006-06-02 |
| 授予单位 | 中国科学院研究生院 |
| 授予地点 | 中国科学院自动化研究所 |
| 导师 | 谭民 ; 徐德 |
| 关键词 | 可重构制造系统 配置 建模 优化 reconfigurable manufacturing system configuration modeling optimization |
| 其他题名 | Research on Modeling and Optimization of Configuration-based Reconfigurable Manufacturing Systems |
| 学位专业 | 控制理论与控制工程 |
| 中文摘要 | 传统的制造模式无法满足制造业从“大规模生产”到“大规模定制化”转变的要求。正是在这种背景之下，科研人员提出了一种新的制造模式—可重构制造系统。可重构制造系统能够通过重构对系统的生产能力和功能进行动态调整，以快速响应市场需求的变化。 论文首先概述了可重构制造系统的研究背景、定义和特点，综述了现有的研究成果。针对传统机床结构设计中加工功能不变的特性，论文提出了一种可行的解决方法。本文将机床的不同加工功能看作是由相应的加工模块所实现，加工模块不再束缚于某台机床，可以自由的从机床上添加或删除。在此基础上，本文提出了“功能模块与机床组合”的观点和基于配置的可重构制造系统。功能模块与机床的任何一种组合称为一种配置，系统从一种配置切换到另一种配置的过程称为系统重构。一般来说，不同配置下的系统性能表现会不同。可重构制造系统研究的一个重要问题，就是寻找一种合理的配置方法使系统达到最优性能。 结合“功能模块与机床组合”的观点，论文提出并研究了非周期和周期可重构流水作业的建模与优化问题。非周期可重构流水作业由析取图模型表示，在析取图模型的基础上按照一定规则构建优化模型，以完成所有作业的最大时间最小化作为优化的目标。周期可重构流水作业由赋时事件图或扩展赋时事件图模型表示，优化模型由赋时事件图或扩展赋时事件图按一定的规则派生得到，以系统周期最小化作为优化的目标。通过求解对应的优化模型，可以得到不同情况下非周期可重构流水作业和周期可重构流水作业的最优配置。 在考虑到重构代价的情况下，论文对可重构制造系统的动态重构过程进行了分析，建立了系统动态重构过程的有向图模型，并在扩展有向图模型的基础上给出了求解最优重构路径的动态规划和线性规划方法。 最后，论文总结了取得的研究成果，并给出了需继续开展的研究工作。 |
| 英文摘要 | Under the background that traditional manufacturing paradigms cannot meet the requirement of manufacturing industry shifting from mass production to mass customization, a new manufacturing paradigm named reconfigurable manufacturing system (RMS) was proposed by researchers. RMS can adjust the capacities and functionalities dynamically by means of reconfigurations in quick response to the market requirement changes. The research background, definitions and characteristics of RMS are briefly discussed and a literature review is given at the beginning. A feasible solution to the invariability of processing functionalities in the traditional machine architecture design is put forward that different processing functionalities can be achieved by corresponding processing modules and that the processing modules are no longer fixed on the machine and can be added to or removed from the machine. A configuration-based RMS is proposed based on the idea “combination of functional modules and machines” in this dissertation. One configuration is defined as one possible combination of functional modules and machines and reconfiguration is defined as the process of the system changing from one configuration to another. Generally speaking, the system performances differ under different configurations and one goal of RMS is to find a reasonable configuration method to achieve the best system performance. Based on the idea “combination of functional modules and machines”, the problems of acyclic and cyclic reconfigurable flow shops modeling and optimization are studied. The acyclic reconfigurable flow shops can be modeled as disjunctive graphs, based on which the optimization models are constructed according to some rules with makespan minimization as objective. The cyclic reconfigurable flow shops are modeled as timed event graphs or extended timed event graphs from which optimization models are derived with system cycle time minimization as objective. The best configurations of acyclic and cyclic reconfigurable flow shops can be obtained by solving the corresponding optimization models. Considering the reconfiguration cost, the dynamic reconfiguration processes are analyzed and modeled as directed graph. On the basis of the extended directed graphs, the best reconfiguration paths are obtained using the methods of dynamic programming and linear programming. Finally, the research results are summarized and future work is discussed. |
| 语种 | 中文 |
| 其他标识符 | 200318014602981 |
| 源URL | [http://ir.ia.ac.cn/handle/173211/5926]  |
| 专题 | 毕业生_博士学位论文 |
| 推荐引用方式 GB/T 7714 | 任思成. 基于配置的可重构制造系统建模与优化研究[D]. 中国科学院自动化研究所. 中国科学院研究生院. 2006. |
入库方式： OAI收割
来源：自动化研究所
浏览0
下载0
收藏0
其他版本
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。