SIA OpenIR  > 自动化系统研究室
基于事件的生产过程建模及调度算法的研究与实现
其他题名Research and Implementation of Production Process Modeling and Scheduling Algorithm Based on Event
林硕1,2
导师史海波
分类号TP273
关键词制造执行系统 过程建模 实时调度 消息发布 工作流 生产过程事件
索取号TP273/L62/2012
页数103页
学位专业机械电子工程
学位名称博士
2012-11-27
学位授予单位中国科学院沈阳自动化研究所
学位授予地点沈阳
作者部门自动化系统研究室
摘要本文从对生产过程建模入手,对制造执行系统中的生产过程模型进行了深入研究,改进系统性能,解决目前存在的问题。主要工作有以下几点:(1)对生产过程的建模方法进行研究,并提出新的建模方法,同时建立生产过程事件模型。首先通过对3种不同行业产品的生产过程进行分析和总结,提出关键生产过程中的事件,并对事件之间的共性和特性进行总结。在此基础上,参考ISA95标准,提出一种基于生产过程事件驱动的过程建模方法,包括生产过程事件的定义、属性和分类,以及事件间的关系和触发规则,对生产过程建立模型。建立触发规则库,对触发规则分层次管理。通过对事件属性和触发规则的配置,该建模方法可以对不同行业和产品的生产过程建立事件模型。同其他建模方法比较,该方法符合ISA95标准,生产过程通过事件的属性和触发规则描述,直观、易于理解,具有可适应性高和可集成、可配置的优点。(2)对生产过程事件模型的运行机制进行研究,并提出一种与业务流程低耦合的模型运行机制。通过对事件处理的特点进行分析,利用基于消息发布/订阅机制,将处理过程转变为消息的发布接收。使模型的运行与实际业务流程的关联分离,提高模型的可适应性和可配置性。并利用工作流引擎,对消息进行并发处理,有效提高模型中事件处理的效率。(3)对实时调度算法进行研究和设计。首先对算法的调度对象,即生产过程事件模型中的事件执行特点进行分析,结合调度算法要提高效率和减少关键任务响应时间的目标,综合考虑了调度对象的多个影响调度结果的属性,包括任务的周期、截止期、关键度和系统负载,提出一种多特征综合的实时调度算法。最后对该算法进行仿真验证,证明其满足设计需求而且可用。 (4)将生产过程事件模型和实时调度算法应用于MES平台中并在实际应用中进行验证。通过在实际项目中的部署运行,对比原来平台性能,证明该模型和实时调度算法的可用性,并且能有效地提高平台处理效率。
其他摘要This article targeted to solve the current MES problem. The article focused on the following main points: (1) Analyze on the current modeling method in production process, introduce fine-tuned modeling method with real-time models. The article first described and summarized production process of 3 different products, point out the key events, and analyzed on the common and unique properties of the events. A new production-process-driven event modeling method is then introduced which includes the definition, proprieties, classification and inter-relations, and the triggering rules among different events. The article was trying to argue that, comparing to other modeling methods, this method provided a better way to describe the production process of 9 main models in 4 classes of ISA95. Furthermore, through the configuration of the event proprieties, this method could describe the production process in an ISA95 model more comprehensively than the ordinary methods. The production process event modeling method could model different industries by creating the triggering rules database, managing by different levels, and configuring the triggering rules of the relevant property of the event. The new method could significantly increase the ease of integration, adaptability, and configurability of the MES platforms. (2) Analyze on the operating mechanism of the production process event modeling and propose the execution method of the events. The article first analyzed the characteristics of different event handling methods, chose the workflow that manages the execution processes, and then transforms the execution processes into message lists. A message broadcast/subscription based workflow engine are then introduced which can manage the messages created in the event handling processes. The message broadcast/subscription processes are based on the triggering rules between the events: the pre-events are the broadcaster, and the post-event are the subscriber, every triggering rule can be mapped back to a pair of broadcaster/subscriber. (3) Analyze on the real-time dispatch method. The article first listed and analyzed the current classic real-time dispatch methods, compared the different methods on their characteristics in production process to define the dispatch requirements and strategy, then introduced a new multi-characteristic real-time dispatch method to replace the current ones. By applying verification and simulation tools, the method was proved to be able to fulfill all event handling requirements in a production process event model. (4) Apply the production process event model and real-time dispatch method into MES platforms and verify their efficiency in real world production projects. The model and the real-time dispatch method increased the platform processing efficiency in comparison with the platform which the model and the method are not applied, and then verified to be efficient in real world productions.
语种中文
产权排序1
文献类型学位论文
条目标识符http://ir.sia.cn/handle/173321/10643
专题自动化系统研究室
作者单位1.中国科学院沈阳自动化研究所
2.中国科学院大学
推荐引用方式
GB/T 7714
林硕. 基于事件的生产过程建模及调度算法的研究与实现[D]. 沈阳. 中国科学院沈阳自动化研究所,2012.
条目包含的文件
文件名称/大小 文献类型 版本类型 开放类型 使用许可
基于事件的生产过程建模及调度算法的研究与(1418KB) 开放获取使用许可请求全文
个性服务
推荐该条目
保存到收藏夹
查看访问统计
导出为Endnote文件
谷歌学术
谷歌学术中相似的文章
[林硕]的文章
百度学术
百度学术中相似的文章
[林硕]的文章
必应学术
必应学术中相似的文章
[林硕]的文章
相关权益政策
暂无数据
收藏/分享
所有评论 (0)
暂无评论
 

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。