基于规则和案例的压缩机集成故障诊断专家系统研究
|
摘要
论文在分析现有故障诊断方法特点和局限性的基础上,对基于规则推理和案例推理的集成诊断系统的设计方案、集成机制和关键技术进行了研究。
规则获取是专家系统的“瓶颈”问题,论文采用基于粗集的数据挖掘方法进行规则获取,对属性约简的基本算法做出了改进,利用属性约减算法和启发式值约减算法,在一定程度上简化了计算,提高了约简效率。粗糙集从现有的故障诊断的数据出发,能有效地处理不完备、不一致数据,实现高效快速地提取故障诊断规则的目的。
在运用知识工程技术进行案例的表示、组织以及案例库的构建中,论文根据压缩机故障诊断领域的特点以及故障诊断专家系统的应用需求,以基于特征的故障案例表示为基础,依据故障的特征属性和内容将压缩机故障案例按照故障特征分类。在案例检索中,根据故障案例特征的层次结构,将整个案例库组织成多级分层索引结构。这种按故障特征分类,分层索引的策略思想不仅降低了案例表示、组织以及案例库的构建难度,而且极大地提高了案例检索的效率。
在上述研究工作基础上,研究建立了一个集规则诊断(RBR)和案例推理(CBR)于一体的集成智能诊断专家系统。该集成系统既可以充分利用相关的领域知识又可以利用以往的诊断经验,相互补充,从而提高诊断系统的故障分辨能力和准确性。
The article analyzes the characteristic and limitation of the applied fault diagnosis method. Based on that, a new integrated diagnosis method based on RBR (rule-based reasoning) and CBR (case-based reasoning) is presented in this paper. The design plan, integrated mechanism and key technology of the integrated diagnosis method have been developed.
A Rough sets data mining method for rule acquisition is selected to over come the shortages of some knowledge attaining methods existed before. The basic algorithm is improved by using attribute reduction and heuristic value reduction, which can simplify the calculation and enhance the efficiency of seeking the reduction in some extent. Rough sets theory can gathers the existing diagnoses data and obtains the diagnosis rule efficiently and quickly through handle incomplete and inconsistent data availably.
According to the demand of fault diagnosis expert system and the characteristic of compressor fault diagnosis, the compressor fault case is classified by characteristic and detail on the foundation of characteristic-based case expression. The multi-layered structure of the case database of the system is formed in accordance with the structure of the key characteristic which is abstracted from fault cases. The strategy of layering index and characteristic classification have reduced the complexity of case expression, the management and construction of case database, it also have enhanced the index speed and efficiency dramatically.
Based on the research above, this paper introduces an integrated fault diagnosis expert system integrated with rule-based reasoning and case-based reasoning. The integrated system can not only make use of the field knowledge but can take advantage of diagnose experience in the past, There by,it is mutually complementary to improve the accuracy and identification capability of the diagnosis System.
规则获取是专家系统的“瓶颈”问题,论文采用基于粗集的数据挖掘方法进行规则获取,对属性约简的基本算法做出了改进,利用属性约减算法和启发式值约减算法,在一定程度上简化了计算,提高了约简效率。粗糙集从现有的故障诊断的数据出发,能有效地处理不完备、不一致数据,实现高效快速地提取故障诊断规则的目的。
在运用知识工程技术进行案例的表示、组织以及案例库的构建中,论文根据压缩机故障诊断领域的特点以及故障诊断专家系统的应用需求,以基于特征的故障案例表示为基础,依据故障的特征属性和内容将压缩机故障案例按照故障特征分类。在案例检索中,根据故障案例特征的层次结构,将整个案例库组织成多级分层索引结构。这种按故障特征分类,分层索引的策略思想不仅降低了案例表示、组织以及案例库的构建难度,而且极大地提高了案例检索的效率。
在上述研究工作基础上,研究建立了一个集规则诊断(RBR)和案例推理(CBR)于一体的集成智能诊断专家系统。该集成系统既可以充分利用相关的领域知识又可以利用以往的诊断经验,相互补充,从而提高诊断系统的故障分辨能力和准确性。
The article analyzes the characteristic and limitation of the applied fault diagnosis method. Based on that, a new integrated diagnosis method based on RBR (rule-based reasoning) and CBR (case-based reasoning) is presented in this paper. The design plan, integrated mechanism and key technology of the integrated diagnosis method have been developed.
A Rough sets data mining method for rule acquisition is selected to over come the shortages of some knowledge attaining methods existed before. The basic algorithm is improved by using attribute reduction and heuristic value reduction, which can simplify the calculation and enhance the efficiency of seeking the reduction in some extent. Rough sets theory can gathers the existing diagnoses data and obtains the diagnosis rule efficiently and quickly through handle incomplete and inconsistent data availably.
According to the demand of fault diagnosis expert system and the characteristic of compressor fault diagnosis, the compressor fault case is classified by characteristic and detail on the foundation of characteristic-based case expression. The multi-layered structure of the case database of the system is formed in accordance with the structure of the key characteristic which is abstracted from fault cases. The strategy of layering index and characteristic classification have reduced the complexity of case expression, the management and construction of case database, it also have enhanced the index speed and efficiency dramatically.
Based on the research above, this paper introduces an integrated fault diagnosis expert system integrated with rule-based reasoning and case-based reasoning. The integrated system can not only make use of the field knowledge but can take advantage of diagnose experience in the past, There by,it is mutually complementary to improve the accuracy and identification capability of the diagnosis System.
引文
[1] 张安华.机电设备状态监测与故障诊断技术[M].西安:西北工业大学出版社,2000
[2] 苏育刊.面向预知的设备群监测系统研究与开发:[湖南大学硕士学位论文].长沙:湖南大学,2007,1-2
[3] 吴今培,肖健华.智能故障诊断与专家系统[M].北京:科学出版社,1997,9
[4] 王道平.故障智能诊断系统的理论与方法[M].北京:冶金出版社,2001:5-7
[5] 赵亮.数据仓库与数据挖掘在远程诊断中的应用研究:[北京科技大学硕士学位论文].北京:北京科技大学,2005
[6] 蒋瑜,陈循,杨雪.智能故障诊断研究与发展[J].信息技术,2002,21(2)
[7] Ian Watson.Applying Case-Based Reasonging:Techniques for Enterprise Systems.Morgan Kaufmann Publishers,1997
[8] 范兴铎,盛颂恩.基于模糊神经网络的压缩机运行状态预报模型的研究[J].压缩机技术,2005,5
[9] 姚华堂,盛颂恩,范兴铎.往复式压缩机故障诊断专家系统设计与实现[J].压缩机技术,2005,1
[10] 王日新,徐敏强,张嘉钟.离心式压缩机故障诊断专家系统研究[J].风机技术,2002,(03):33-35
[11] 王致杰,王耀才,李冬.基于组合式智能预测的提升机故障诊断策略研究[EB].http://www.ca800.com/apply/html/2007-12-17/n25559.html,2004-1-13
[12] L.Xu.An Integrated Rule-and Case-Based Approach to AIDS initial Assessment. International Journal of Bio-Medical Computing.1996,40:197-207
[13] 骆霞武,景旭文.基于数据挖掘的故障智能诊断系统研究[J],机械设计与制造,2005,3
[14] Jiawei Han,Micheline Kamber.Data Mining Comcepts and Technique.Morgan Kaufmann Publishers,2000
[15] 阮跃.基于案例、规则和模型推理的电站智能诊断系统 [J].电站系统工程,2006,14(3):25-28
[16] Slowinski R.Rough set approach to decision analysis.AlExPert,1995:19-25
[17] 刘清.Rough 集及 Rough 推理[M].北京:科学出版社,2001 年,第一版
[18] Usama Fayyad, Paul Stolorz, Data mining and KDD: Promise and challenges, F uture Generation Computer Systems.1997,13:99-115
[19] 史忠植.高级人工智能[M].北京:科学出版社,1998
[20] 范明等.数据挖掘概念与技术[M].北京:机械工业出版社,200l
[21] Han J.数据挖掘:概念与技术[M].北京:高等教育出版社,2001
[22] 朱明.数据挖掘[M].合肥:中国科学技术大学出版社,2002
[23] 王国胤.Rough 集理论与知识获取[M].西安:西安交通大学出版社,2001
[24] 邵峰晶,等.数据挖掘原理与算法[M].北京:中国水利水电出版社,2003
[25] D.Cheung,J.Han,V.Ng and C.Y.Wong.Maintenance of discovered association rules in large databases: An incremental updating technique. Stanley Y.W.Su. Proceeings of the 12th International Conference on Data Engineering(ICDE'96),New Orleans, IEE Computer Society,1996:1 06-114
[26] 伍星.基于数据挖掘的设备远程监测和故障诊断系统的研究:[上海交通大学博士学位论文].上海:上海交通大学,2005:41-43
[27] 李宗杰.基于粗糙集的飞机远程诊断知识获取模型研究:[中国民航大学硕士学位论文].北京:中国民航大学,2007
[28] 曹倬瑝,许南山.基于粗集理论的压缩机故障诊断方法研究[J].计算机与数字工程,2005,34(2):36-37
[29] 曾黄麟.粗集理论及其应用-关于数据推理的新方法[M].重庆大学出版社
[30] J W Guan,D A Bell . Rough Computatlonal Methods for information Systems[J].Artifica1 Intelligence,1998,35(105):77-103
[31] Skowron A,Crausze.The Diseernlbility Matrix,and functions in Information Systems,Handbook of Applications,and Advances of the Rough Set Theory.Kluwer Academic Publlshers,1992:331-362
[32] Brown P,Della Pietra V.Class-based n-gram models of natural language [J]. Computational Linguistics,2005,28 (4) 477-480
[33] 王艳梅,胡小平,李舟军.基于粗糙集理论的液体火箭发动机故障诊断[J].导弹与航天运载技术,2006,(2):51-53
[34] Barletta, R.. An intreduction to case-based reasoning.AI Expert,1991:42-49
[35] Helton.T.The hottest new AI techno1ogy-case-based reasoning.The spang robinson report on artificial intelligence,1991,7(8)
[36] Slade.S. Case-basedreasoni: A research Paradigm.AI Magazine,1991,42-55
[37] 史忠值.知识发现[M].北京:清华大学出版社,2002
[38] 周平,柴天佑.基于案例推理的磨矿分级系统智能设定控制[J].东北大学学报,2007,28(5):613-616
[39] D.Alia,K.Branting.Stratified Case-Based Reasoning:Reusing Hierar-chical Problem Solving Episodes. IJCAI-95. 1995,384-390
[40] I.Watson,S.Perera.A Hierarchical Case Representation Using Context Guided Retrieval.Knowledge-Based Systems,1998,11:285-292
[41] D.Alia,K.Branting.Stratified Case-Based Reasoning:Reusing Hierarchical Problem Solving Episodes.IJCAI-95.1995,384-390
[42] 杨健,杨晓光.一种基于 K-NN 的案例相似度权重调整算法[J].计算机工程与应用,2007,43(23):8-11
[43] 王日新.基于案例推理的智能诊断技术研究及应用:[哈尔滨工业大学博士学位论文].哈尔滨:哈尔滨工业大学,2002
[44] 安淑芝等编著.数据仓库与数据挖掘.北京:清华大学出版社,2005.6
[45] Inmon WH.数据仓库[M].北京:机械工业出版社,2000
[46] 关惠玲,韩捷.设备故障诊断专家系统原理及实践[M].北京:机械工业出版社,2000
[47] 涂又光.数据挖掘在网络故障诊断中的应用研究:[重庆大学硕士学位论文].重庆:重庆大学,2004,5-6
[48] 曹倬瑝.基于数据挖掘的故障诊断:[北京化工大学硕士学位论文].北京:北京化工大学,2005,4-5
[2] 苏育刊.面向预知的设备群监测系统研究与开发:[湖南大学硕士学位论文].长沙:湖南大学,2007,1-2
[3] 吴今培,肖健华.智能故障诊断与专家系统[M].北京:科学出版社,1997,9
[4] 王道平.故障智能诊断系统的理论与方法[M].北京:冶金出版社,2001:5-7
[5] 赵亮.数据仓库与数据挖掘在远程诊断中的应用研究:[北京科技大学硕士学位论文].北京:北京科技大学,2005
[6] 蒋瑜,陈循,杨雪.智能故障诊断研究与发展[J].信息技术,2002,21(2)
[7] Ian Watson.Applying Case-Based Reasonging:Techniques for Enterprise Systems.Morgan Kaufmann Publishers,1997
[8] 范兴铎,盛颂恩.基于模糊神经网络的压缩机运行状态预报模型的研究[J].压缩机技术,2005,5
[9] 姚华堂,盛颂恩,范兴铎.往复式压缩机故障诊断专家系统设计与实现[J].压缩机技术,2005,1
[10] 王日新,徐敏强,张嘉钟.离心式压缩机故障诊断专家系统研究[J].风机技术,2002,(03):33-35
[11] 王致杰,王耀才,李冬.基于组合式智能预测的提升机故障诊断策略研究[EB].http://www.ca800.com/apply/html/2007-12-17/n25559.html,2004-1-13
[12] L.Xu.An Integrated Rule-and Case-Based Approach to AIDS initial Assessment. International Journal of Bio-Medical Computing.1996,40:197-207
[13] 骆霞武,景旭文.基于数据挖掘的故障智能诊断系统研究[J],机械设计与制造,2005,3
[14] Jiawei Han,Micheline Kamber.Data Mining Comcepts and Technique.Morgan Kaufmann Publishers,2000
[15] 阮跃.基于案例、规则和模型推理的电站智能诊断系统 [J].电站系统工程,2006,14(3):25-28
[16] Slowinski R.Rough set approach to decision analysis.AlExPert,1995:19-25
[17] 刘清.Rough 集及 Rough 推理[M].北京:科学出版社,2001 年,第一版
[18] Usama Fayyad, Paul Stolorz, Data mining and KDD: Promise and challenges, F uture Generation Computer Systems.1997,13:99-115
[19] 史忠植.高级人工智能[M].北京:科学出版社,1998
[20] 范明等.数据挖掘概念与技术[M].北京:机械工业出版社,200l
[21] Han J.数据挖掘:概念与技术[M].北京:高等教育出版社,2001
[22] 朱明.数据挖掘[M].合肥:中国科学技术大学出版社,2002
[23] 王国胤.Rough 集理论与知识获取[M].西安:西安交通大学出版社,2001
[24] 邵峰晶,等.数据挖掘原理与算法[M].北京:中国水利水电出版社,2003
[25] D.Cheung,J.Han,V.Ng and C.Y.Wong.Maintenance of discovered association rules in large databases: An incremental updating technique. Stanley Y.W.Su. Proceeings of the 12th International Conference on Data Engineering(ICDE'96),New Orleans, IEE Computer Society,1996:1 06-114
[26] 伍星.基于数据挖掘的设备远程监测和故障诊断系统的研究:[上海交通大学博士学位论文].上海:上海交通大学,2005:41-43
[27] 李宗杰.基于粗糙集的飞机远程诊断知识获取模型研究:[中国民航大学硕士学位论文].北京:中国民航大学,2007
[28] 曹倬瑝,许南山.基于粗集理论的压缩机故障诊断方法研究[J].计算机与数字工程,2005,34(2):36-37
[29] 曾黄麟.粗集理论及其应用-关于数据推理的新方法[M].重庆大学出版社
[30] J W Guan,D A Bell . Rough Computatlonal Methods for information Systems[J].Artifica1 Intelligence,1998,35(105):77-103
[31] Skowron A,Crausze.The Diseernlbility Matrix,and functions in Information Systems,Handbook of Applications,and Advances of the Rough Set Theory.Kluwer Academic Publlshers,1992:331-362
[32] Brown P,Della Pietra V.Class-based n-gram models of natural language [J]. Computational Linguistics,2005,28 (4) 477-480
[33] 王艳梅,胡小平,李舟军.基于粗糙集理论的液体火箭发动机故障诊断[J].导弹与航天运载技术,2006,(2):51-53
[34] Barletta, R.. An intreduction to case-based reasoning.AI Expert,1991:42-49
[35] Helton.T.The hottest new AI techno1ogy-case-based reasoning.The spang robinson report on artificial intelligence,1991,7(8)
[36] Slade.S. Case-basedreasoni: A research Paradigm.AI Magazine,1991,42-55
[37] 史忠值.知识发现[M].北京:清华大学出版社,2002
[38] 周平,柴天佑.基于案例推理的磨矿分级系统智能设定控制[J].东北大学学报,2007,28(5):613-616
[39] D.Alia,K.Branting.Stratified Case-Based Reasoning:Reusing Hierar-chical Problem Solving Episodes. IJCAI-95. 1995,384-390
[40] I.Watson,S.Perera.A Hierarchical Case Representation Using Context Guided Retrieval.Knowledge-Based Systems,1998,11:285-292
[41] D.Alia,K.Branting.Stratified Case-Based Reasoning:Reusing Hierarchical Problem Solving Episodes.IJCAI-95.1995,384-390
[42] 杨健,杨晓光.一种基于 K-NN 的案例相似度权重调整算法[J].计算机工程与应用,2007,43(23):8-11
[43] 王日新.基于案例推理的智能诊断技术研究及应用:[哈尔滨工业大学博士学位论文].哈尔滨:哈尔滨工业大学,2002
[44] 安淑芝等编著.数据仓库与数据挖掘.北京:清华大学出版社,2005.6
[45] Inmon WH.数据仓库[M].北京:机械工业出版社,2000
[46] 关惠玲,韩捷.设备故障诊断专家系统原理及实践[M].北京:机械工业出版社,2000
[47] 涂又光.数据挖掘在网络故障诊断中的应用研究:[重庆大学硕士学位论文].重庆:重庆大学,2004,5-6
[48] 曹倬瑝.基于数据挖掘的故障诊断:[北京化工大学硕士学位论文].北京:北京化工大学,2005,4-5