CASIA OpenIR  > 毕业生  > 硕士学位论文
Thesis Advisor郑应平
Degree Grantor中国科学院自动化研究所
Place of Conferral中国科学院自动化研究所
Degree Discipline控制理论与控制工程
Abstract离散事件动态系统(简称DEDS)作为系统学和控制学领域内的一 个非常活跃的前沿方向,正受到越来越多人的重视。本文的内容是作 者硕士研究生期间在DEDS的一个应用领域——制造系统的调度与控制 方面所做的主要工作,全文分四章。 第一章是绪论,概述了当前研究DEDS的几种主要方法,并分别 加以阐述。然后介绍了研究制造系统调度与控制的方法。 第二章研究制造系统的实时调度策略。本章阐述了Gershwin等人 提出的递阶控制:各类事件按其发生的频率分层, 由最低频的顶层往 下,最高层离线地决定决策参数,如每一种工作的需求量。中间层需 在线地计算出每一种工件在短期内的生产速度。最低层需确定待加工 工件的加工时刻和加工次序,即确定调度策略,使得工件实际生产率 尽量接近中间层次决定的生产率。Perkins和Kumar在最低层作了些工作 ,他们模型中最突出的内容是反馈、稳定性和实时性,但他们提出 的CLB、CAF策略中,没有体现最优性。于是,在他们的工作的基础上 ,本章提出了一个新的库存模型,并为最低层提供了一个优化目标, 然后求出了基于最优准则的调度策略。同时研究了该策略的稳定性。 为体现该策略的优越性,本章举出了一个实例,将该策略与CLB、CAF 策略进行了比较,证实了所求得的策略的优越性。 第三章是研究生产线系统的建模、性能分析和优化控制。生产 线系统的建模一一般有三种方法: 排队论方法,马氏过程方法以及近似 法。本章所讨论的性能指标是指系统的可用度与生产率。本章研究一 类装配式生产线系统,其中装配机器为瓶颈过程。该系统为连续工件 流的确定性生产线系统。为研究方便,将其拆解为装配部分与输出部 分,装配部分采用近似法建模,并应用有报酬的更新理论求得了该部 分的稳态生产率。并利用所求得的表达式讨论了缓冲区容量分配的优 化问题,这是一个非线性规划问题。输出部分主要研究了为使系统正 常运行,输出缓冲区的容量大小以及输出机器生产速率的大小。 第四章总结了全文,并给出今后要继续研究的若干问题。
Other AbstractDiscrete Event Dynamic Systems (abbreviated as DEDS) as an actively advancing aspect are paid more and more attention by people studying in the field of systems and controls. This paper centers around one of the applied aspect of DEDS, that is the scheduling and control of manufacturing system. The contents of this paper are done by the author during her graduate study and divided into four chapters. Chapter 1 is Introduction. First outline some important methods by which we study DEDS at present, and then explain them respectively. Finally introduce the way by which we do research in the field of the scheduling and control of manufacturing system. Chapter 2 is concerned about the real-time scheduling policies of manufacturing systems. The chapter outlines the Gershwin's hierarchical scheduling scheme in which various decisions and events are grouped into various level of a hierarchy according to their characteristic frequencies. The least frequent events are assigned to the top level which generates the decision parameters of the policy, such as the demand rate of each part type. The middle level computers the short-term production rates for each part type for each machine state. The lower level decides the time and the order to produce the parts, i.e. deciding the scheduling policy, with the aim of maintaining the real production rates equal to the production rates computerd by the middle level. The Perkins and Kumar have done some research in the lower level. The characteristics of their model is of feedback, stability and real-time. But the policies of CLB and CAF developed by them did not reflect the thought of optimization. So, based on their work, we develope a new capacity model and provide an optimizing aim for the lower level. Then we obtain a new scheduling policy according to optimizing principle. We also have proven the stability of this policy. In order to show the superiority of this policy, we give an example and compare the result of this policy with the results of CLB and CAF policies. The comparing results demonstrate the superiority of the new policy obtained by us. Chapter 3 talks about the modeling, the analysis of performance and the optimizing control of production line systems. Generally we model production systems by three methods: Queuing theory, Markov process theory and approximate method. The performance measures discussed by this chapter include the availability and the throughput of systems. We conside an assembly-like flow system where the assembly machine is the bottleneck of the system. For convenience, we decompose the system into two parts, one is assembly, the other is output. We model the assembly part by the approximate method and derive an approximate expression for the steady-state average throughput of the part according to the renewal reward theory. We also indicate how it can be used to buffer allocation which is a nonlinear program. In the o
Other Identifier347
Document Type学位论文
Recommended Citation
GB/T 7714
刘雪婷. 制造系统的调度与控制[D]. 中国科学院自动化研究所. 中国科学院自动化研究所,1995.
Files in This Item:
There are no files associated with this item.
Related Services
Recommend this item
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[刘雪婷]'s Articles
Baidu academic
Similar articles in Baidu academic
[刘雪婷]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[刘雪婷]'s Articles
Terms of Use
No data!
Social Bookmark/Share
All comments (0)
No comment.

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.