面向敏捷维修的维修资源智能优化配置技术研究
|
摘要
敏捷维修是近年来针对先进制造系统可靠性保障而提出的新概念。敏捷维修技术的覆盖面很广,既包括涉及具体制造设备的维修技术,也包括维修过程中对维修任务的协同指挥和各项维修资源的优化配置。论文重点研究了维修资源配置过程中的关键技术和相关问题,并在此基础上建立了面向敏捷维修的维修资源智能优化配置系统。
首先通过对制造系统设备功能结构的分析,分别从制造设备功能部件维修级别的视角及其部件单元维修策略的视角来描述制造设备的维修需求状态;研究人工神经网络技术在维修需求识别中的应用,并将基于Rough集的连续值多属性决策信息系统模型和推理算法运用于维修需求识别中。
在对制造设备的维修需求进行识别的基础上,通过建立维修资源需求预测专家系统提高维修资源需求预测的敏捷性,进一步实现了维修资源需求的快速预测。
对维修资源智能优化配置的相关技术进行了研究,提出运用进程调度理论和基于维修时间窗口方法的资源调度方法指导维修资源的预配置;在分析了敏捷维修中维修资源冲突产生根源的基础上,提出了在维修资源配置过程中采用结构化处理和非结构化处理相结合的资源冲突消解策略,取得了良好的效果。
最后以某制造设备为维修对象进行系统验证。先由其触发维修需求,再通过辅助决策原型系统进行相应的维修需求识别、维修资源需求预测以及维修资源配置,验证了面向敏捷维修的维修资源智能优化配置技术的可行性及其方法的有效性。
Agile maintenance is a new concept presented recently, which is oriented to the logistics of advanced manufacturing system. The technology of Agile Maintenance has covered many aspects, and mainly contains both the practical maintenance technology referred to the equipments and the scheme technology of the maintenance resources. This research has done some studies about the key technology of the maintenance resources scheme, and built an intelligent maintenance-resources-scheme system oriented to the Agile Maintenance.
Firstly, by analyzing the structure of the manufacturing equipments system from the function perspective, maintenance requirement might be defined with two aspects, the maintenance demand level of the equipment units and the maintenance mode strategy to the parts of them. Then, some intelligent algorithms, such as Artificial Neural Network and rough set theory have been introduced to identify the maintenance requirement. Based on the identification of the maintenance requirement, it has moved to study the technology predicting the maintenance-resources requirement according to the maintenance requirement, such as the theory of expert system, which has increased the promptitude and the intelligence of the prediction. After that, it continued with the research of applying process scheduling of operation system theory and time-window setting method to the maintenance resources scheme. Besides, a strategy has been discussed to eliminate the conflicts in multitask maintenance process and balance the benefit of maintenance tasks. Finally, based on a simple manufacturing system, a software prototype system has been built to validate the model and solution.
首先通过对制造系统设备功能结构的分析,分别从制造设备功能部件维修级别的视角及其部件单元维修策略的视角来描述制造设备的维修需求状态;研究人工神经网络技术在维修需求识别中的应用,并将基于Rough集的连续值多属性决策信息系统模型和推理算法运用于维修需求识别中。
在对制造设备的维修需求进行识别的基础上,通过建立维修资源需求预测专家系统提高维修资源需求预测的敏捷性,进一步实现了维修资源需求的快速预测。
对维修资源智能优化配置的相关技术进行了研究,提出运用进程调度理论和基于维修时间窗口方法的资源调度方法指导维修资源的预配置;在分析了敏捷维修中维修资源冲突产生根源的基础上,提出了在维修资源配置过程中采用结构化处理和非结构化处理相结合的资源冲突消解策略,取得了良好的效果。
最后以某制造设备为维修对象进行系统验证。先由其触发维修需求,再通过辅助决策原型系统进行相应的维修需求识别、维修资源需求预测以及维修资源配置,验证了面向敏捷维修的维修资源智能优化配置技术的可行性及其方法的有效性。
Agile maintenance is a new concept presented recently, which is oriented to the logistics of advanced manufacturing system. The technology of Agile Maintenance has covered many aspects, and mainly contains both the practical maintenance technology referred to the equipments and the scheme technology of the maintenance resources. This research has done some studies about the key technology of the maintenance resources scheme, and built an intelligent maintenance-resources-scheme system oriented to the Agile Maintenance.
Firstly, by analyzing the structure of the manufacturing equipments system from the function perspective, maintenance requirement might be defined with two aspects, the maintenance demand level of the equipment units and the maintenance mode strategy to the parts of them. Then, some intelligent algorithms, such as Artificial Neural Network and rough set theory have been introduced to identify the maintenance requirement. Based on the identification of the maintenance requirement, it has moved to study the technology predicting the maintenance-resources requirement according to the maintenance requirement, such as the theory of expert system, which has increased the promptitude and the intelligence of the prediction. After that, it continued with the research of applying process scheduling of operation system theory and time-window setting method to the maintenance resources scheme. Besides, a strategy has been discussed to eliminate the conflicts in multitask maintenance process and balance the benefit of maintenance tasks. Finally, based on a simple manufacturing system, a software prototype system has been built to validate the model and solution.
引文
[1] John Moubray.有关“简化的”RCM 的剖析.中国设备工程,2003.8.
[2] Sandy Dunn. Maintenance terminology — some key terms. 2004.1 ,http://www.plant-maintenance.com.
[3] V.S. Deshpande,J.P. Modak.Application of RCM to a medium scale industry.Reliability Engineering and System Safety,2002(77): 31-43.
[4] Diego J. Pedregal,Fausto P. Garcia,Felix Schmid.RCM2 predictive maintenance of railway systems based on unobserved components models.Reliability Engineering and System Safety,2004(83): 103-110.
[5] Yong Sun,Lin Ma,Joseph Mathew.Mechanical system bazard estimation using condition monitoring.Mechanical System and Signal Processing,2006(20): 1189-1201.
[6] 代科学,李国辉,万歆睿.状态维修理论、技术与实施.航空制造技术,2005.1:101-105.
[7] 臧铁钢,杨明忠.基于网络的智能维修信息系统的研究与开发.机械工程师,2003.01.
[8] 胡荣华.面向敏捷维修的制造设备监控与故障诊断技术研究,[硕士学位论文] .南京,南京航空航天大学,2006.3:1.
[9] 谢小鹏.设备状态识别与维修决策.北京,中国石化出版社,2000.
[10] 蒋瑜,陶利民,杨雪等.机械设备状态预测方法的发展和研究.现代机械,2001(3):84-87.
[11] 黄金国,臧铁钢,张淑猛等.智能维修决策支持系统的研究.制造业信息化,2003.12:6-8.
[12] 王生凤,郭昊,徐宗昌.设备资源优化管理的 EVA 研究.现代机械,2003.2:67-68.
[13] 涂忆柳,李晓东.维修工程管理研究与发展综述.工业工程于管理,2004.4:7-12.
[14] 巫世晶,向农.电力设备维修计划优化管理研究.电力建设,2004.2:48-51.
[15] Rommert Dekker.Application of maintenance optimization models: a review and analysis.Reliability Engineering and System Safety,1996(51): 229-240.
[16] 张涛,武小悦,谭跃进.Petri 网在系统可靠性分析中的应用.电子产品可靠性与环境试验,2003.2(1):60-65.
[17] 丁彩虹,黄文虎.应用 Petri 网改进基于故障树的诊断方法.振动工程学报,2000.6(13):277-282.
[18] 徐宗昌,朴允华,刘义乐.基于 Petri 网的维修保障系统建设方案研究.系统工程理论与实践,2004.8:141-144.
[19] 冯静,吴孟达.Profust 故障树建模与分析.国防科技大学学报,2001.1(23):85-88.
[20] 刘芳,赵建印,郭波.基于扩展 CGSPN 的维修保障系统仿真评估与优化方法.计算机仿真,2004.4(22):82-85.
[21] 王亚彬,贾希胜,康建设.仿真技术在维修资源预测中的应用研究.计算机仿真,2005.7(22):249-258.
[22] 赵昔.高风险不确定性维修决策方法.中国设备工程,2004.6:7-9.
[23] 张戌社,宁晨校.模糊数学在机械设备故障诊断中的应用.河北科技大学学报,2002(2):55-59.
[24] 宋林郁,郭顺生,李益兵.Rough 集在智能维修决策支持系统中的研究与应用.组合机床与自动化加工技术,2005.1:24-29.
[25] 涛基斌,郭应征,周太全.基于前馈式神经网络的化工设备维修决策.南京化工大学学报,2000.9(22):11-14.
[26] 辛道勇,王健,张琦等.基于 ANN 的工程机械预知维修专家系统研究.机械与电子,2004.4:63-66.
[27] 邱永庆,刘毅,吴海桥等.基于事例优化的故障树诊断.中国民航学院学报,2003.7(21):37-39.
[28] 戈鹏,殷国富,高伟.基于知识的制造资源建模与应用.中国机械工程,2003.9(14):1478-1481.
[29] Jay.Lee , Hai Qiu , Jun Ni . Information Technologies and Predictive Tools for Next-Generation Maintenance System,11th IFAC Symposium on Information Control Problems in Manufacturing,Salvador,Brazil,2004.5.
[30] Laura Swanson . An information-processing model of maintenance management.International Journal of Production Economics,2003(83): 45-64.
[31] Martin T. Hagan, Howard B. Demuth, Mark H. Beale.Neural Network Design[M].机械工业出版社,2002.9:197-259.
[32] LIU Wen-jun, XIAO Qi-mei.Fuzzy Decision Algorithm Based on Rough Sets[A].Fuzzy Systems and Mathematics, 2006, 20 (2): 127-132.
[33] 刘震宇,郭宝龙,杨林耀.一种新的用于连续值属性离散化的约简算法[J].控制与决策, 2002, 17 (5): 545-549.
[34] 张文修,仇国芳.基于粗糙集的不确定性决策[M].清华大学出版社, 2005, 7.
[35] Krysgkiewing M.Comparative studies of alternative type of knowledge reduction in consistent systems [J].International Journal of Intelligent Systems, 2001, 16, (3): 105-120.
[36] 张文修,米据生,吴伟志.不协调目标信息系统的知识约简[J].计算机学报, 2003, 26 (1): 12-18.
[37] 李然.连续值域信息系统的规则提取与知识约简[J].模糊系统与数学, 2003, 17 (4): 40-47.
[38] 王涛.机电系统趋势预测及智能诊断技术的研究,[硕士学位论文].北京,北京机械工业学院,2005.1:32-39.
[39] 夏洪山,许峰.分布式实时系统中的多 Agent 调度[J].南京航空航天大学学报,2004,36(3):302-307.
[40] Graham J R . Real-time scheduling in distributed multi agent systems [EB/OL].http://www.eecis.udel.edu,2001.
[41] 乔颖,邹冰.一种实时异构系统的集成动态调度算法[J].软件学报,2002,13(12):2253.
[42] Konstantinos Vasseions Prouskas . Real-time extensions of APRL for time-aware multi-agent system programming [EB/OL].http://casbah.ee.ic.ac.uk,2002-10.
[43] Wanger T,Garvey A,Lesser V.Criteria-directed task scheduling [J].International Journal of Approximate Reasoning,Special Issue on Scheduling,1998,91-98.
[44] Liu C,Laylan J.Scheduling algorithms for multi-programming in a hard real-time environment [J].Journal of the ACM,1993,1(20):2033.
[2] Sandy Dunn. Maintenance terminology — some key terms. 2004.1 ,http://www.plant-maintenance.com.
[3] V.S. Deshpande,J.P. Modak.Application of RCM to a medium scale industry.Reliability Engineering and System Safety,2002(77): 31-43.
[4] Diego J. Pedregal,Fausto P. Garcia,Felix Schmid.RCM2 predictive maintenance of railway systems based on unobserved components models.Reliability Engineering and System Safety,2004(83): 103-110.
[5] Yong Sun,Lin Ma,Joseph Mathew.Mechanical system bazard estimation using condition monitoring.Mechanical System and Signal Processing,2006(20): 1189-1201.
[6] 代科学,李国辉,万歆睿.状态维修理论、技术与实施.航空制造技术,2005.1:101-105.
[7] 臧铁钢,杨明忠.基于网络的智能维修信息系统的研究与开发.机械工程师,2003.01.
[8] 胡荣华.面向敏捷维修的制造设备监控与故障诊断技术研究,[硕士学位论文] .南京,南京航空航天大学,2006.3:1.
[9] 谢小鹏.设备状态识别与维修决策.北京,中国石化出版社,2000.
[10] 蒋瑜,陶利民,杨雪等.机械设备状态预测方法的发展和研究.现代机械,2001(3):84-87.
[11] 黄金国,臧铁钢,张淑猛等.智能维修决策支持系统的研究.制造业信息化,2003.12:6-8.
[12] 王生凤,郭昊,徐宗昌.设备资源优化管理的 EVA 研究.现代机械,2003.2:67-68.
[13] 涂忆柳,李晓东.维修工程管理研究与发展综述.工业工程于管理,2004.4:7-12.
[14] 巫世晶,向农.电力设备维修计划优化管理研究.电力建设,2004.2:48-51.
[15] Rommert Dekker.Application of maintenance optimization models: a review and analysis.Reliability Engineering and System Safety,1996(51): 229-240.
[16] 张涛,武小悦,谭跃进.Petri 网在系统可靠性分析中的应用.电子产品可靠性与环境试验,2003.2(1):60-65.
[17] 丁彩虹,黄文虎.应用 Petri 网改进基于故障树的诊断方法.振动工程学报,2000.6(13):277-282.
[18] 徐宗昌,朴允华,刘义乐.基于 Petri 网的维修保障系统建设方案研究.系统工程理论与实践,2004.8:141-144.
[19] 冯静,吴孟达.Profust 故障树建模与分析.国防科技大学学报,2001.1(23):85-88.
[20] 刘芳,赵建印,郭波.基于扩展 CGSPN 的维修保障系统仿真评估与优化方法.计算机仿真,2004.4(22):82-85.
[21] 王亚彬,贾希胜,康建设.仿真技术在维修资源预测中的应用研究.计算机仿真,2005.7(22):249-258.
[22] 赵昔.高风险不确定性维修决策方法.中国设备工程,2004.6:7-9.
[23] 张戌社,宁晨校.模糊数学在机械设备故障诊断中的应用.河北科技大学学报,2002(2):55-59.
[24] 宋林郁,郭顺生,李益兵.Rough 集在智能维修决策支持系统中的研究与应用.组合机床与自动化加工技术,2005.1:24-29.
[25] 涛基斌,郭应征,周太全.基于前馈式神经网络的化工设备维修决策.南京化工大学学报,2000.9(22):11-14.
[26] 辛道勇,王健,张琦等.基于 ANN 的工程机械预知维修专家系统研究.机械与电子,2004.4:63-66.
[27] 邱永庆,刘毅,吴海桥等.基于事例优化的故障树诊断.中国民航学院学报,2003.7(21):37-39.
[28] 戈鹏,殷国富,高伟.基于知识的制造资源建模与应用.中国机械工程,2003.9(14):1478-1481.
[29] Jay.Lee , Hai Qiu , Jun Ni . Information Technologies and Predictive Tools for Next-Generation Maintenance System,11th IFAC Symposium on Information Control Problems in Manufacturing,Salvador,Brazil,2004.5.
[30] Laura Swanson . An information-processing model of maintenance management.International Journal of Production Economics,2003(83): 45-64.
[31] Martin T. Hagan, Howard B. Demuth, Mark H. Beale.Neural Network Design[M].机械工业出版社,2002.9:197-259.
[32] LIU Wen-jun, XIAO Qi-mei.Fuzzy Decision Algorithm Based on Rough Sets[A].Fuzzy Systems and Mathematics, 2006, 20 (2): 127-132.
[33] 刘震宇,郭宝龙,杨林耀.一种新的用于连续值属性离散化的约简算法[J].控制与决策, 2002, 17 (5): 545-549.
[34] 张文修,仇国芳.基于粗糙集的不确定性决策[M].清华大学出版社, 2005, 7.
[35] Krysgkiewing M.Comparative studies of alternative type of knowledge reduction in consistent systems [J].International Journal of Intelligent Systems, 2001, 16, (3): 105-120.
[36] 张文修,米据生,吴伟志.不协调目标信息系统的知识约简[J].计算机学报, 2003, 26 (1): 12-18.
[37] 李然.连续值域信息系统的规则提取与知识约简[J].模糊系统与数学, 2003, 17 (4): 40-47.
[38] 王涛.机电系统趋势预测及智能诊断技术的研究,[硕士学位论文].北京,北京机械工业学院,2005.1:32-39.
[39] 夏洪山,许峰.分布式实时系统中的多 Agent 调度[J].南京航空航天大学学报,2004,36(3):302-307.
[40] Graham J R . Real-time scheduling in distributed multi agent systems [EB/OL].http://www.eecis.udel.edu,2001.
[41] 乔颖,邹冰.一种实时异构系统的集成动态调度算法[J].软件学报,2002,13(12):2253.
[42] Konstantinos Vasseions Prouskas . Real-time extensions of APRL for time-aware multi-agent system programming [EB/OL].http://casbah.ee.ic.ac.uk,2002-10.
[43] Wanger T,Garvey A,Lesser V.Criteria-directed task scheduling [J].International Journal of Approximate Reasoning,Special Issue on Scheduling,1998,91-98.
[44] Liu C,Laylan J.Scheduling algorithms for multi-programming in a hard real-time environment [J].Journal of the ACM,1993,1(20):2033.