PDF《具有随机机器故障的制造过程建模与性能分析》

第14 卷第

9 期 计算机集成制造系统 Vol.

14 No.

9 2

0 0

8 年9月Computer Integrated Manufacturing Systems Sep.

2 0

0 8 文章编号: 1006- 5911( 2008) 09- 1704-

06 收稿日期:

2007211 212;

修订日期:

2008 201 228.Received

12 Nov. 2007;

accepted

28 Jan. 2008. 基金项目: 国家自然科学基金资助项目( 60674114);

国家

863 计划资助项目( 2006AA04Z164, 2007AA04Z1A4) .Foundation item: Project sup2 ported by the National Natural Science Foundation, China( No. 60674114), and the National H igh2T ech. R&

D Pr ogram, China ( No. 2006AA04Z164, 2007AA04Z1A4) . 作者简介: 刘昶( 1973) ) , 女, 辽宁辽阳人, 中国科学院沈阳自动化研究所博士研究生, 主要从事制造执行系统、 制造过程建模与仿真等的研 究.E2mail: changl@ sia. cn. 具有随机机器故障的制造过程建模与性能分析 刘昶1,

2 , 史海波1 , 袁杰1,

2 ( 1. 中国科学院 沈阳自动化研究所, 辽宁 沈阳 110016;

2. 中国科学院 研究生院, 北京 100039) 摘要: 为研究机器故障和维修活动对制造过程性能的影响, 提出一种基于广义随机 Petri 网的制造过程建模 与性能分析方法.分析了随机机器故障特征;

定义了两种故障发现模式和两种中断作业处理策略;

给出具有随机 机器故障的制造过程的不同模型方法;

通过对模型结构特征的分析, 证明了其有效性.针对不同策略和参数设置 进行了性能仿真.分别以平均产量和平均过程流时间等性能指标, 分析了单个工作站的性能;

采用平均产量, 分析 了具有两个工作站的流水线的性能.仿真结果表明, 故障率、 平均维修时间、 缓存数量配置、 维修工人数量、 故障发 现模式和中断作业处理策略是影响具有随机机器故障的制造过程性能的主要因素. 关键词: 制造过程建模;

维修;

随机故障;

广义随机 Pet ri 网 中图分类号: T P302.

7 文献标识码: A Modeling and performance analysis of manufacturing processes with stochastic machine failures L IU Chang1,

2 , SH I H ai2bo1 , YUA N Jie1,

2 ( 1. Shenyang Institute of Automation, Chinese Academy of Sciences, Shenyang 110016, China;

2. Graduate School, Chinese Academy of Sciences, Beijing 100039, China) Abstract: To study the influence on the performance of manufacturing processes exerted by the machine failures and repairing activities, an approach based on Generalized Stochast ic Petri Net ( GSPN) w as proposed for manufacturing processesp modeling and performance analysis. The characteristics of stochastic machine failures w ere analyzed. T wo failure 2detection modes and two interrupted2job 2handling policies w ere defined respectively. Different models for manufacturing processes which w ere subject to stochastic failures w ere presented. Validit y of the models w as veri2 fied by analyzing the structural characterist ics. Simulation were performed to deal w it h different policies and parame2 ter configurations. Performance of single w ork station was analyzed by the average throughput and t he mean process flow time. Performance of the f low line with tw o w ork stations was analyzed by average throughput. Simulation re2 sults demonst rated t hat t he failure rate, the mean time to repair, the number of buffers, the number of repair per2 sonnel, failure 2detection modes and interrupted2job2handling policies were the main inf luencing factors for manufac2 turing processesp performance. Key words:manufacturing process modeling;