民用航空器基于案例推理的故障诊断系统设计
|
摘要
我国民航机务维护的发展与民航机队规模和航班运营量相比表现出较大的滞后性。目前我国民用航空器维修采用以零部件完备理论为基础的定时维修,而不是以监控设备技术状态为基础的视情维护。由于经验的抽象性以及不连续性,使得机务工作仍然停留在原先的“师徒”模式以及“传帮带”的传统工作模式中。同时,由于经验的不可存储以及无法共享,使得部分经验以及数据随着专家的退休或者各类原因而趋向消亡,因此民航维修系统内的经验无法得到螺旋式的上升、从而大大的制约了中国民航维修企业技术发展与进步。针对上述情况,本文对民用航空维修的故障诊断做了如下研究:
1.通过对民航维修企业针对航空器的排故过程进行总结,得到了民用航空器故障诊断流程以及民用航空维修领域经验的表述形式。并针对民航维修企业所面对的航空器故障进行了简单的分类。在结合了手册以及经验的基础上,提出基于案例推理为主,故障树的分析为辅的智能故障诊断系统。利用案例推理技术可以加快对于相同案例的搜索以及匹配速度,针对案例库中无法进行适合匹配的故障案例,可以利用故障树的分析进行故障诊断。
2.由于基于案例推理技术同知识表示密切相关,在不同的领域会有不同的方法,本文主要结合民用航空器维修领域,对基于案例推理的实际应用做了详细的分析。主要研究并分析有关案例在数据库中的保存形式,案例的特征表示以及各类案例的分类方式以及数据库存储结构。在此基础上,简要分析了关于案例搜索算法以及案例匹配算法。提出了基于关键字符的比较算法以及匹配算法,最终完成智能故障诊断系统的诊断算法。
3.为保证故障诊断系统在航空器故障诊断上尽可能的全面,本文在基于案例的推理技术的基础上,结合基于故障树分析的方法对系统进行补充。主要讨论了有关数据库技术在故障树的保存上的结构形式,在数据库的基础上对故障树的遍历算法进行讨论。
4.本文在上述各类算法的基础上,利用VB6.0编程语言以及SQL Service 2000数据库系统对案例库以及故障树的数据库进行构建。通过VBA以及OBV技术完成VB程序对数据库进行读取,通过VB“所见即所得”完成程序的主要人机交互界面的设计。最终通过实验完成程序的调试。验证搜索算法以及匹配算法的正确性以及可靠性。
In China, compared with the Airlines fleet size and the business volume, the development of civil aviation locomotive maintenance show much more hysteresis . At present our civil aircraft maintenance based on components complete theory as a foundation of regular maintenance. Instead of condition-based maintenance which is based on monitoring equipment technical status. Because of the abstractness and discontinuity of experience, aircraft maintenance work has still stayed in the original "master" mode and "ChuanBangDai" traditional working mode. At the same time, because of the experience of storage, and cannot be shared, while the experts is going to be retired, some part of the experience and data is becoming extinct. Therefore the experience of civil aviation maintenance systems can't spirals up into the high level. It dramatically limits the Chinese civil aviation maintenance technology development and progress of enterprises. In such circumstances, this paper makes following research to civil aviation maintenance fault diagnosis:
1.Through to making the summary of the civil aviation maintenance enterprises for aircraft troubleshooting process, it obtained the expressions of the civil aircraft fault diagnosis process and civil aviation maintenance field experience. And it made a simple classification of the failure, which the civil aircraft maintenance enterprises are facing. On the Basis of the combination of the experience and the manual, it suggested a intelligent fault diagnosis system, which is given priority to CRB and fault tree analysis. Use of the case for the same reasoning skills will speed up the case of a search speed and matching. When the fault case will match out in the Case Library, the system can diagnose the fault by the fault tree analysis.
2.Because CBR technology is closely related with the knowledge representation, there are different ways in the different areas. Reasoning based on the case for the practical application has been minutely dissected with civil aircraft maintenance field in this article. And it prevailingly analysed preservation forms of relevant cases in the database, case feature expression and all kinds of case classification method and database storage structure. On this basis, this paper analyses the case search and examples about matching algorithms. It put forward based on key character comparison algorithm and matching algorithm. Finally it completed intelligent fault diagnosis system diagnostic algorithm.
3.To ensure that the fault diagnosis system in aircraft fault diagnosis on the comprehensive as possible, In this paper, based on case reasoning technique in the foundation, the union based on fault tree analysis method of system and supplement. it mainly discussed the database technology in fault tree structure form and preserving in database based on fault tree traversing algorithm is discussed.
4.Based on the above all kinds of algorithms, and on the basis of using VB6.0 programming languages and SQL database system for Service 2000 putted forward and fault tree of database construction. Through OBV and VBA technology, the VB program the database accesses, by VB "what you see is what you get" complete program mainly the man-machine interface design. Finally through debugging experiment complete program, this article to verify the accuracy of the fundamental algorithm and a match and reliability.
1.通过对民航维修企业针对航空器的排故过程进行总结,得到了民用航空器故障诊断流程以及民用航空维修领域经验的表述形式。并针对民航维修企业所面对的航空器故障进行了简单的分类。在结合了手册以及经验的基础上,提出基于案例推理为主,故障树的分析为辅的智能故障诊断系统。利用案例推理技术可以加快对于相同案例的搜索以及匹配速度,针对案例库中无法进行适合匹配的故障案例,可以利用故障树的分析进行故障诊断。
2.由于基于案例推理技术同知识表示密切相关,在不同的领域会有不同的方法,本文主要结合民用航空器维修领域,对基于案例推理的实际应用做了详细的分析。主要研究并分析有关案例在数据库中的保存形式,案例的特征表示以及各类案例的分类方式以及数据库存储结构。在此基础上,简要分析了关于案例搜索算法以及案例匹配算法。提出了基于关键字符的比较算法以及匹配算法,最终完成智能故障诊断系统的诊断算法。
3.为保证故障诊断系统在航空器故障诊断上尽可能的全面,本文在基于案例的推理技术的基础上,结合基于故障树分析的方法对系统进行补充。主要讨论了有关数据库技术在故障树的保存上的结构形式,在数据库的基础上对故障树的遍历算法进行讨论。
4.本文在上述各类算法的基础上,利用VB6.0编程语言以及SQL Service 2000数据库系统对案例库以及故障树的数据库进行构建。通过VBA以及OBV技术完成VB程序对数据库进行读取,通过VB“所见即所得”完成程序的主要人机交互界面的设计。最终通过实验完成程序的调试。验证搜索算法以及匹配算法的正确性以及可靠性。
In China, compared with the Airlines fleet size and the business volume, the development of civil aviation locomotive maintenance show much more hysteresis . At present our civil aircraft maintenance based on components complete theory as a foundation of regular maintenance. Instead of condition-based maintenance which is based on monitoring equipment technical status. Because of the abstractness and discontinuity of experience, aircraft maintenance work has still stayed in the original "master" mode and "ChuanBangDai" traditional working mode. At the same time, because of the experience of storage, and cannot be shared, while the experts is going to be retired, some part of the experience and data is becoming extinct. Therefore the experience of civil aviation maintenance systems can't spirals up into the high level. It dramatically limits the Chinese civil aviation maintenance technology development and progress of enterprises. In such circumstances, this paper makes following research to civil aviation maintenance fault diagnosis:
1.Through to making the summary of the civil aviation maintenance enterprises for aircraft troubleshooting process, it obtained the expressions of the civil aircraft fault diagnosis process and civil aviation maintenance field experience. And it made a simple classification of the failure, which the civil aircraft maintenance enterprises are facing. On the Basis of the combination of the experience and the manual, it suggested a intelligent fault diagnosis system, which is given priority to CRB and fault tree analysis. Use of the case for the same reasoning skills will speed up the case of a search speed and matching. When the fault case will match out in the Case Library, the system can diagnose the fault by the fault tree analysis.
2.Because CBR technology is closely related with the knowledge representation, there are different ways in the different areas. Reasoning based on the case for the practical application has been minutely dissected with civil aircraft maintenance field in this article. And it prevailingly analysed preservation forms of relevant cases in the database, case feature expression and all kinds of case classification method and database storage structure. On this basis, this paper analyses the case search and examples about matching algorithms. It put forward based on key character comparison algorithm and matching algorithm. Finally it completed intelligent fault diagnosis system diagnostic algorithm.
3.To ensure that the fault diagnosis system in aircraft fault diagnosis on the comprehensive as possible, In this paper, based on case reasoning technique in the foundation, the union based on fault tree analysis method of system and supplement. it mainly discussed the database technology in fault tree structure form and preserving in database based on fault tree traversing algorithm is discussed.
4.Based on the above all kinds of algorithms, and on the basis of using VB6.0 programming languages and SQL database system for Service 2000 putted forward and fault tree of database construction. Through OBV and VBA technology, the VB program the database accesses, by VB "what you see is what you get" complete program mainly the man-machine interface design. Finally through debugging experiment complete program, this article to verify the accuracy of the fundamental algorithm and a match and reliability.
引文
[1]潘茂庆,惠克翔.基于神经网络的飞机故障诊断专家系统.航空工程与维修.2002(5):34~35
[2]曲明,郝红卫. CBR与RBR相结合的实时专家系统设计与实现.计算机工程.2004(18):144~145
[3]史定华,王松瑞.故障树分析技术方法和理论.北京:北京师范大学出版社.1993.
[4]季会媛.动态故障树分析方法研究[硕士论文].国防科技大学.2002.
[5] Fussell J B. A Formal Methodology for Fault Tree Construction . Nuclear Science & engineering, 1973.
[6] A320飞机故障历史记录(内部资料).
[7]章立民. SQL Server 2000完全实战——数据转换服务(DTS).北京:中国铁道出版社.2002.
[8]王晟,马里杰.SQL Server数据库开发经典案例解析,北京:清华大学出版社.2006.
[9]邹建.中文版SQL Server 2000开发与管理应用实例,北京:人民邮电出版社.2005.
[10]李保利,陈玉忠,俞士汶.信息抽取研究综述.计算机工程与应用.2003(10):1~5.
[11]朱明.互联网信息智能搜索与获取方法研究[博士论文].中国科学技术大学.2000.
[12] S.H.Lin,J.M.Ho.Discovering Informative Content Blocks from Web Documents.In Proceedings of ACM.2002:387~396.
[13]瞿有利,于浩,徐国伟等. Web页面信息块的自动分割.中文信息学报.2003(1):6~13.
[14] Aidan Finn,Nicholas Kushmerick,Barry Smyth.“Fact or fiction:Content Classification for digital libraries”.In Joint DELOS-NSF Workshop on Personalisation and Recommender Systems in Digital Libraries,Dublin.2001:134~141.
[15]张晓滨,宋擒豹,沈钧毅.一种基于语义特征的Web文档检索方法.计算机工程与应用.2001(20):7~9.
[16]雷继尧,何世德编著.机械故障诊断基础知识,西安:西安交通大学出版社.1989.
[17]吴今培,肖健华.智能故障诊断与专家系统.北京:科学出版社.1997.
[18]黄可鸣.专家系统导论.南京:东南大学出版社.1988.
[19]杨炳儒.知识工程与知识发现.北京:冶金工业出版社.2000.
[20]武波,马玉祥.专家系统(修订版).北京:北京理工大学出版社.2001.
[21]陈玮.故障诊断专家系统的学习机制.上海交通大学学报.1999(9)
[22]邓志江,自行火炮故障智能诊断与预测系统设计[硕士学位论文].南京理工大学.2007.
[23]张恒喜.数字化维修理论与技术.北京:国防工业出版社.2006.
[24]孙卫祥.基于数据挖掘与信息融合的故障诊断方法研究[博士学位论文].上海交通大学.2006.
[25]李美华,朝大明,陆怀民.人工智能在机械故障诊断中的应用.森林工程.2006,3(22).
[26] Davies J,Goel A.k,Nersessian N J.Transfer in Visual Case-Based Problem Solving[C].Anon Intemational Conference on Case-Based Reasoning 2005.Berlin:Springer.2005.
[27] R.C.Schank,D.B.Leake.CreativityandLearning in a Case-Based Explainer. Artificial Intelligence. 1989 (1-3):353~385.
[28] I.Watson,F.Marin Case-Based Reasoning::a review.The Knowledge Engineering Review,1994,9(4):327~354.
[29] B.P.Allen.Case-Based Reasoning:Business Applications.Communications of the ACM,1994,37(3):40~42.
[30] J.Hunt.Case Based Diagnosis and Repari of SoftWare Faults.Experts Systems,1997,14(1):15~20.
[31]房文娟,李绍稳,袁媛等.基于案例推理技术的研究与应用.农业网络信息.2005(1).
[32]白士红,李海平.基于实例推理的柴油机故障诊断与维修.机械设计与制造.2005.
[33]谷建光,张为华,王中伟等.基于案例推理的固体火箭发动机总体方案设计技术研究.弹箭与制导学报.2006.
[34]苏群星,娄建安,王丽.案例智能维修训练系统的设计与实现.火力与指挥控制.2005(7).
[35]胡良明,徐诚,李万平.基于案例推理的自行火炮故障诊断专家系统.火炮发射与控制学报.2006(2).
[36]张文修,仇国芳,吴伟志.粗糙集属性约简的一般理论.中国科学.2005 (12):1304~1313.
[37]周平,丁进良,岳恒等.基于相似粗糙集的案例特征权值确定新方法.信息与控制.2005(3):1002~1011.
[38]李希亮,贾希胜.粗集理论在CBR系统案例评价中的应用.军械工程学院.
[39]杨健,杨晓光,刘晓彬等.一种基于k-NN的案例相似度权重调整算法.计算机工程与应用.2007 (23)
[40] Jiawei Han,数据挖掘:概念与技术.北京:高等教育出版社.2001:223~262
[41]庞彦军,贾文国,刘开第.粗集中属性权重的概念及应用.第八届中国青年运筹信息管理学者大会论文集.2006(8),805~807.
[42]周平,丁进良,岳恒等.基于相似粗糙集的案例特征权值确定新方法.信息与控制.2006 (35).
[43] Park C S,Han I.A case-based reasoning with the feature weights derived by analytichierarchy process for bankruptey prediction[J].Expert Systems with Applications.2002,23(3):255~264.
[44]刘清,Rough集及Rough推理.北京:科学出版社.2001.
[45]史忠植,王文杰.人工智能.北京:国防工业出版社.2007.
[46]邱永庆,刘毅等.基于事例优化的故障树诊断[J].中国民航学院学报.2003.7,21:37~39.
[47]东方航空公司. Airbus A300-600 Trouble Shooting Manual, revision NO.33 Jun 01/03.
[48]梁胜利,李学仁等.基于融合诊断的某型飞机故障诊断应用研究.航空计算技术.2003 (3):86~88.
[49]薄海涛,白振兴.基于故障树和神经网络的飞机电源系统故障诊断.自动化与仪表.2005 (4):65~67.
[50]张若青,沈现军,裘丽华.飞机余度舵机系统鲁棒故障诊断[J].控制理论与应用.2000 (2):205~209.
[51]祁树胜.飞机电源GCU BITE系统研究[J].西安航空技术高等专科学校学报.2005 (5):17~18.
[52] J.Kolodner. An Introduction to Case-Based Reasoning. Artificial Intelligence Review. 1992 (1):3~34.
[2]曲明,郝红卫. CBR与RBR相结合的实时专家系统设计与实现.计算机工程.2004(18):144~145
[3]史定华,王松瑞.故障树分析技术方法和理论.北京:北京师范大学出版社.1993.
[4]季会媛.动态故障树分析方法研究[硕士论文].国防科技大学.2002.
[5] Fussell J B. A Formal Methodology for Fault Tree Construction . Nuclear Science & engineering, 1973.
[6] A320飞机故障历史记录(内部资料).
[7]章立民. SQL Server 2000完全实战——数据转换服务(DTS).北京:中国铁道出版社.2002.
[8]王晟,马里杰.SQL Server数据库开发经典案例解析,北京:清华大学出版社.2006.
[9]邹建.中文版SQL Server 2000开发与管理应用实例,北京:人民邮电出版社.2005.
[10]李保利,陈玉忠,俞士汶.信息抽取研究综述.计算机工程与应用.2003(10):1~5.
[11]朱明.互联网信息智能搜索与获取方法研究[博士论文].中国科学技术大学.2000.
[12] S.H.Lin,J.M.Ho.Discovering Informative Content Blocks from Web Documents.In Proceedings of ACM.2002:387~396.
[13]瞿有利,于浩,徐国伟等. Web页面信息块的自动分割.中文信息学报.2003(1):6~13.
[14] Aidan Finn,Nicholas Kushmerick,Barry Smyth.“Fact or fiction:Content Classification for digital libraries”.In Joint DELOS-NSF Workshop on Personalisation and Recommender Systems in Digital Libraries,Dublin.2001:134~141.
[15]张晓滨,宋擒豹,沈钧毅.一种基于语义特征的Web文档检索方法.计算机工程与应用.2001(20):7~9.
[16]雷继尧,何世德编著.机械故障诊断基础知识,西安:西安交通大学出版社.1989.
[17]吴今培,肖健华.智能故障诊断与专家系统.北京:科学出版社.1997.
[18]黄可鸣.专家系统导论.南京:东南大学出版社.1988.
[19]杨炳儒.知识工程与知识发现.北京:冶金工业出版社.2000.
[20]武波,马玉祥.专家系统(修订版).北京:北京理工大学出版社.2001.
[21]陈玮.故障诊断专家系统的学习机制.上海交通大学学报.1999(9)
[22]邓志江,自行火炮故障智能诊断与预测系统设计[硕士学位论文].南京理工大学.2007.
[23]张恒喜.数字化维修理论与技术.北京:国防工业出版社.2006.
[24]孙卫祥.基于数据挖掘与信息融合的故障诊断方法研究[博士学位论文].上海交通大学.2006.
[25]李美华,朝大明,陆怀民.人工智能在机械故障诊断中的应用.森林工程.2006,3(22).
[26] Davies J,Goel A.k,Nersessian N J.Transfer in Visual Case-Based Problem Solving[C].Anon Intemational Conference on Case-Based Reasoning 2005.Berlin:Springer.2005.
[27] R.C.Schank,D.B.Leake.CreativityandLearning in a Case-Based Explainer. Artificial Intelligence. 1989 (1-3):353~385.
[28] I.Watson,F.Marin Case-Based Reasoning::a review.The Knowledge Engineering Review,1994,9(4):327~354.
[29] B.P.Allen.Case-Based Reasoning:Business Applications.Communications of the ACM,1994,37(3):40~42.
[30] J.Hunt.Case Based Diagnosis and Repari of SoftWare Faults.Experts Systems,1997,14(1):15~20.
[31]房文娟,李绍稳,袁媛等.基于案例推理技术的研究与应用.农业网络信息.2005(1).
[32]白士红,李海平.基于实例推理的柴油机故障诊断与维修.机械设计与制造.2005.
[33]谷建光,张为华,王中伟等.基于案例推理的固体火箭发动机总体方案设计技术研究.弹箭与制导学报.2006.
[34]苏群星,娄建安,王丽.案例智能维修训练系统的设计与实现.火力与指挥控制.2005(7).
[35]胡良明,徐诚,李万平.基于案例推理的自行火炮故障诊断专家系统.火炮发射与控制学报.2006(2).
[36]张文修,仇国芳,吴伟志.粗糙集属性约简的一般理论.中国科学.2005 (12):1304~1313.
[37]周平,丁进良,岳恒等.基于相似粗糙集的案例特征权值确定新方法.信息与控制.2005(3):1002~1011.
[38]李希亮,贾希胜.粗集理论在CBR系统案例评价中的应用.军械工程学院.
[39]杨健,杨晓光,刘晓彬等.一种基于k-NN的案例相似度权重调整算法.计算机工程与应用.2007 (23)
[40] Jiawei Han,数据挖掘:概念与技术.北京:高等教育出版社.2001:223~262
[41]庞彦军,贾文国,刘开第.粗集中属性权重的概念及应用.第八届中国青年运筹信息管理学者大会论文集.2006(8),805~807.
[42]周平,丁进良,岳恒等.基于相似粗糙集的案例特征权值确定新方法.信息与控制.2006 (35).
[43] Park C S,Han I.A case-based reasoning with the feature weights derived by analytichierarchy process for bankruptey prediction[J].Expert Systems with Applications.2002,23(3):255~264.
[44]刘清,Rough集及Rough推理.北京:科学出版社.2001.
[45]史忠植,王文杰.人工智能.北京:国防工业出版社.2007.
[46]邱永庆,刘毅等.基于事例优化的故障树诊断[J].中国民航学院学报.2003.7,21:37~39.
[47]东方航空公司. Airbus A300-600 Trouble Shooting Manual, revision NO.33 Jun 01/03.
[48]梁胜利,李学仁等.基于融合诊断的某型飞机故障诊断应用研究.航空计算技术.2003 (3):86~88.
[49]薄海涛,白振兴.基于故障树和神经网络的飞机电源系统故障诊断.自动化与仪表.2005 (4):65~67.
[50]张若青,沈现军,裘丽华.飞机余度舵机系统鲁棒故障诊断[J].控制理论与应用.2000 (2):205~209.
[51]祁树胜.飞机电源GCU BITE系统研究[J].西安航空技术高等专科学校学报.2005 (5):17~18.
[52] J.Kolodner. An Introduction to Case-Based Reasoning. Artificial Intelligence Review. 1992 (1):3~34.