SIA OpenIR  > 自动化系统研究室
复杂离散制造过程建模与性能评价方法研究
Alternative TitleResearch on modeling and performance evaluation of complex discrete manufacturing system
刘昶1,2
Department自动化系统研究室
Thesis Advisor史海波
ClassificationTP273
Keyword复杂离散制造过程 建模 性能评价 生产控制 Petri网
Call NumberTP273/L71/2008
Pages122页
Degree Discipline机械电子工程
Degree Name博士
2008-05-30
Degree Grantor中国科学院沈阳自动化研究所
Place of Conferral沈阳
Abstract离散制造系统是制造业中最为复杂的制造系统之一,其生产特点表现在:生产方式的异步、并行;生产柔性强;调度复杂性高等方面。要实现对复杂离散制造系统的控制,首先必须对其建模。随着制造系统自动化程度提高,规模的增大,组成因素的复杂化,性能评价问题变得日益复杂并越来越受到人们的重视。提供直观的模型描述方法、有效的数学理论工具和模型分析方法及分析软件,是制造系统性能评价迫切需要解决的问题。为此本文进行了以下工作: 首先,提出了一种支持性能评价的制造过程模型框架。该框架体现了车间生产闭环控制机制,全面描述了车间制造过程相关组成要素之间的关系,可用于指导制造过程的建模和基于性能评价的优化控制。模型框架核心部分是生产过程模型、过程控制模型和过程性能模型,这三部分构成了制造过程模型。该过程模型在工厂模型的支持下,由事件模型驱动,受策略模型约束,通过一些过程准则和方法进行测量和监视,对性能相关数据进行统计、分析和评估,反映在性能指标上。 第二,确定了离散车间性能评价指标体系。该指标体系建立了不同性能指标与不同层次、不同类型的制造过程以及不同性能评价主体之间的关联,为选择评价方法奠定了基础。该指标体系按照车间目标将性能指标分为三大类,即时间、质量、成本;每一类细化为三个层次,分别是基本性能指标、扩展性能指标和综合性能指标。 第三,基于随机Petri网建立了具有随机机器故障的制造过程性能模型。该模型包含两个基本的状态转换环,将正常状态和异常情况分别描述;基于故障的随机特征,设置了服从泊松分布的故障发生模块;根据不同的故障发现模式,衍生出若干状态转换环,清晰地将故障发生、发现分离;还考虑了中断作业的处理,使模型更接近实际系统动态行为特征。另外,本文设计了考虑成本信息的作业优先级调度规则,并基于随机高级Petri网(SHLPN)应用到混流装配线动态调度问题中;针对流水线上存在的一类共享资源—维修工人,建立了考虑系统状态的维修排序规则SHLPN模型。通过数学分析和仿真方法结合实现了对以上制造过程模型的性能评价,验证了模型的有效性。 第四,针对性能评价的需求和Petri网在制造过程成本分析方面存在的不足,提出了一种适合制造过程成本分析的价格时间Petri网,并给出了相应的状态空间化简方法。通过对装配生产线多目标工人指派问题的求解,验证了模型和方法的有效性。 最后,根据管理控制过程和生产过程的特点以及它们之间的关系,提出了基于进程代数与Petri网的分层生产控制方法。采用Petri网和进程代数相结合的方法将物流和信息流集成在一个模型中,Petri网描述基于状态的与生产过程有关的过程流,进程代数描述计划、调度等与控制有关的推理过程,构成了两层分布式的车间生产控制结构。 综上可见,本文提出的制造过程建模、性能评价和控制方法,可以支持复杂离散制造系统的分析和综合;基于过程模型实现了复杂离散制造系统的生产过程模拟和性能分析;能够发现影响系统性能的瓶颈环节;为制造过程的改进和制造系统性能的提高提供了依据;进而实现了制造系统自主闭环自适应的优化控制。本文的研究工作具有重要的理论意义和实际应用价值。
Other AbstractA discrete manufacturing system is one of the most complex manufacturing systems in industries, the main characteristics of which are asynchronous, parallel, flexible, and of higher scheduling complexity. To control the complex discrete manufacturing systems, we must model the complex discrete manufacturing systems firstly. With the increasing degree of automation, increasing size, complicated composition factors of manufacturing systems, performance evaluation increasingly becomes complicated and arouses people's attention. It is an urgent issue that provides the intuitive description of the model, the effective mathematical theory tools,analysis methods, and analysis softwares. So the following work is done in this paper. Firstly, the manufacturing process model framework to support performance evaluation is presented, which reflects the shop floor closed-loop production control mechanism and provides a comprehensive description of the related factors of the shop floor manufacturing processes to guide the modeling and optimal control of production processes. The core parts of the framework are the production process model, the process control model and the process performance model; the three models form the manufacturing process model. The model makes statistical analysis and assessment of the performance-related data according to performance indicators, supported by the factory model, driven by the event model, constrained by the policy model, measured and supervised through a number of principles and methods. Secondly, the performance evaluation index system for discrete manufacturing shop floor is determined. The association among the performance indicators, the manufacturing processes of different levels and types, and the different evaluation subjects is established. The indicators are divided into three categories according to the objectives of the shop floor, namely, the time, the quality and the cost. Each of the categories includes three levels, namely, the basic performance indicators, the extended performance indicators and the comprehensive performance indicators. The determination of performance indices is the premise and foundation for choosing the evaluation methods. Thirdly, the manufacturing process model with stochastic machine failures is developed for temporal performance analysis based on Generalized Stochastic Petri Nets (GSPN). The model consists of two basic parallel state conversion rings, which differentiate the normal states and the abnormal situations; the module is set up to simulate machine breakdown, according to the stochastic characteristics of failures obeying a Poisson distribution; several state conversion rings are derived in accordance with the different failure detection modes, which clearly separate the failure occurrence and failure detection; the interrupted job handling policies are considered, making the model closer to the characteristics of the actual system dynamic behaviors. Several cost-based priority scheduling rules are designed, and applied to the dynamic scheduling problem of assembly lines based on Stochastic High Level Petri Nets (SHLPN). The resource allocation rules considering system states are established using SHLPN model, with the application in the assignment of maintenance workers of assembly lines. The performance is evaluated based on the above manufacturing process models with the combination of the mathematical analysis method and simulation. The validity and efficiency of the above method is proved. Fourthly, against the disadvantages of Petri nets in the cost analysis of manufacturing processes, a priced timed Petri net is proposed, and two approaches are provided for state space simplification. The proposed Petri net model can solve the multi-objective worker assignment problems considering both time and cost effectively. The effectiveness of the proposed model and method is verified. Finally, the hierarchical production control method based on Petri nets and process algebra are developed, according to the characteristics of the contol processes and the production processes, as well as the relationship between them. The logistics and information flows are integrated in a model. Petri net models the process flow associated with production processes based on the state conversation, and process algebra describes the reasoning process, including planning, scheduling and control functions. The two-level hierarchical and distributed control structure is realized. In summary, the modeling, performance evaluation and control methods proposed in this dissertation can support the analysis and synthesis of the complex discrete manufacturing systems; do production process simulation and performance analysis based on the proposed process model; identify the system performance bottlenecks; provides basis for the manufacturing process betterment and the manufacturing system performance improvement; further implement the autonomous, closed-loop and self-adaptive optimal control of the manufacturing systems. The research has the academic value and the practical significance.
Language中文
Contribution Rank1
Document Type学位论文
Identifierhttp://ir.sia.cn/handle/173321/136
Collection自动化系统研究室
Affiliation1.中国科学院沈阳自动化研究所
2.中国科学院研究生院
Recommended Citation
GB/T 7714
刘昶. 复杂离散制造过程建模与性能评价方法研究[D]. 沈阳. 中国科学院沈阳自动化研究所,2008.
Files in This Item:
File Name/Size DocType Version Access License
10001_20031801470305(2094KB) 开放获取--Application Full Text
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[刘昶]'s Articles
Baidu academic
Similar articles in Baidu academic
[刘昶]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[刘昶]'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.