摘要
输变电设备是电网的重要组成部分,输变电设备的可用性与稳定性直接影响到电网的安全运行。及时发现并排除输变电设备的潜伏性故障是电网企业关注的一项重要课题。随着我国电力工业的发展,一方面,电网规模不断发展,输变电设备数量激增,用户对供电可靠性要求不断提高；另一方面,设备的信息化程度越来越高,设备状态监测技术日益成熟,设备运行数据与测试数据激增,借助信息技术对设备进行故障诊断势在必行。
本文在学习和借鉴国内外相关研究成果的基础之上,建立基于范例推理的输变电设备状态智能诊断模型；以输变电设备的在线监测数据、历史运行数据等为基础,应用智能诊断模型对数据资源进行深度挖掘与分析,建立输变电设备在线状态分析与智能诊断系统。论文的研究主要工作体现在以下五个方面：
(1)在深入研究设备故障诊断和基于数据驱动的设备故障诊断理论与技术体系的基础上,结合范例推理的理论,提出基于范例推理的故障诊断模型,并结合输变电设备故障诊断的实际情况,分析输变电设备故障诊断应用中需要解决的问题。模型以设备的各种数据信息为核心,为既缺乏明确的因果关系又需要大量经验的复杂设备诊断提供了新的思路。
(2)针对基于范例推理的故障诊断模型的关键环节,重点研究解决模型范例库的建立和模型推理过程的设计问题。将核函数技术应用到模型中,构造对局部数据敏感、对数据提取完备的新的核函数,并将核函数应用到支持向量机的分类器中,为输变电设备状态智能诊断模型的提出奠定理论基础。
(3)研究解决输变电设备在线状态分析与智能诊断系统的数据模型问题。以输变电设备数据为中心,建立设备信息模型；在此基础上,建立设备状况范例库,并利用支持向量机分类器,对范例库进行分类学习,建立设备故障分类器与设备指纹识别器,生成设备故障诊断树与设备故障指纹；最后,建立基于范例推理的输变电设备状态智能诊断模型与算法,为输变电设备故障诊断提供方法指导。
(4)针对目前众多输变电设备在线监测系统存在的局限性,应用输变电设备状态智能诊断模型,将现有输变电设备相关的各系统数据进行深度整合,建立输变电设备在线状态分析与智能诊断系统,综合设备的运行巡视、离线试验、带电检测等信息对设备故障进行实时、综合、智能诊断,进一步完善设备故障诊断能力,提高设备故障诊断的准确性。
(5)以某省电力公司额定电压为500KV的变压器作为实证研究对象,将变压器油中溶解气体分析结果组织成范例,应用输变电设备状态智能诊断模型,融合来自不同系统的设备基础和设备运行信息,实现变压器状态实时智能诊断,及时查找出变压器的潜伏性故障,排除可能导致变压器故障的潜在原因,验证输变电设备状态智能诊断模型的有效性。
Power transmission and transformation equipment is an important part of the power grid, its availability and stability has a direct impact on the safe operation of power system. To detect and exclude latent faults is an issue of concern for power grid enterprises. With the rapid development of power industry, on the one hand, the structure of power grid continues to expand, in the meantime, the number of transmission and transformation equipment and user demand for the reliability of power supply are increasing greatly; on the other hand, equipment operating data and test data surge with more informational, so the on-line monitoring and fault diagnosis of their insulation is of great practical for realization of state maintenance.
An intelligent fault diagnosis model of transmission and transformation equipment based on case-based reasoning is proposed in this dissertation on the basis of learning from relevant researches home and abroad. Based on the historical operating and on-line monitoring data, then applying the model to mining and analysis the data resources in depth, a transmission and transformation equipment online status analysis and intelligent diagnosis system is established. The major works are summarized as follows.
This dissertation proposes an intelligent fault diagnosis model of transmission and transformation equipment based on the deep study of fault diagnosis data-driven theory and technology combined with case-based reasoning. The model analyses problems which need to be resolved in the transmission and distribution equipment fault diagnosis applications. The model makes the variety of data inforrmation of equipments as the core, and provides a new way of thinking of the diagnosis of complex equipment which requires a lot of experience for both the lack of a clear causal relationship.
Aimed at the key link for fault diagnosis model based on case-based reasoning, study and solve the problem of model case for the establishment and design of the model of the reasoning process. Applying tne nuclear function technique to the model, construct new nuclear function wnich is sensitive to the local data structure and completely data extraction, and apply the nuclear function to the support vector machine classifier. To lay the theoretical foundation for the proposal of the transmission and distribution equipment fault diagnosis model.
Study and solve the problem of power transmission equipment state analysis and intelligent diagnosis system data model. Make the power transmission equipment data as the center, then establish the device information model. On this basis, establish the device status sample library and using support vector machine classifier, classification learning of the case base, establish the equipment fault classification device fingerprint identification, generate fault diagnosis tree and equipment failure fingerprint. Finally, the establishment of the state of case-based reasoning transmission and distribution equipment intelligent diagnosis model and algorithm, provide the method guide for the transmission and distribution equipment fault diagnosis.
Against the limitations of a number of transmission and distribution equipment online monitoring system, apply the power transmission equipment status intelligent diagnosis model, then use existing transmission and distribution equipment system data for deep integration, and establish the power transmission equipment online status analysis and intelligent diagnosis system, with integrating the information of operation of equipment inspections, off-line test, live detection to diagnose equipment failures in real-time to improve the fault diagnosis capability and accuracy further. Design the overall framework of the system, then discusses the key technologies of the system, and complete the development of a prototype system.
Take a500kV transformer in one province electric power company as the research object, transformer oil dissolved gas analysis results are organized into an example application of power transmission equipment status intelligent diagnosis model, the integration of the base of equipment from different systems and equipment operation, real-time intelligent diagnosis of transformer condition, time to find out the potential failure of the transformer, exclusion may lead to a potential cause of transformer failure to verify the validity of the model of the power transmission device status intelligent diagnosis.
引文
[1]屈靖,郭剑波.“九五”期间我国电网事故统计分析.电网技术,2004,28(21)：60～62
[2]孙才新.输变电设备状态在线监测与诊断技术现状和前景[J].中国电力,2005,38(2)：1-7
[3]刘海鹏.大中型水电站设备健康状态评估体系研究[D].重庆大学,2010.
[4]赵登福,段小峰,张磊等.考虑设备状态和系统风险的输电设备检修计划[J].西安交通大学学报,2012,46(3)：84-89
[5]张怀宇.输变电设备状态检修技术体系研究与实施[J].电网技术.2009
[6]李杰.江苏电网输变电设备状态检修管理研究[D].东南大学,2010
[7]龚李伟.电容型设备绝缘在线监测与智能化故障诊断研究[D].华中科技大学,2008
[8]肖燕.输变电设备的状态检修与可靠性分析[A].2010年全国输变电设备状态检修技术交流研讨会论文集[C].2010：151-156
[9]袁泉.电力设备健康状态评估算法的设计与实现[D].华北电力大学(北京),2003
[10]李明,韩学山,杨明等.电网状态检修概念与理论基础研究[J].中国电机工程学报,2011,31(34)：43-52
[11]王佳明,刘文颖,魏帆等.基于寿命周期成本管理的输变电设备状态检修策略研究[J].电力系统保护与控制,2011,39(5)：77-80
[12]徐淑文.对输变电设备状态检修辅助决策系统的研究[A].2010年全国输变电设备状态检修技术交流研讨会论文集[C].2010：779-784
[13]吴烽,陈斌,方立等.供电企业实施输变电设备状态检修的探讨[A].2010年全国输变电设备状态检修技术交流研讨会论文集[C].2010：69-74
[14]Solomon R, Sandborn P A, Pecht M G. Electronic part life cycle concepts and obsolescence forecasting[J]. Components and Packaging Technologies,2000, 23(4):707-717
[15]张扬,楼其民.北美电力企业输变电设备状态检修实施调研与分析限[R].2004
[16]Wu Xing, Chen Jin, Li Ruqiang. Modeling a Web-based Remote Monitoring and Fault Diagnosis System with UML and Component Technology[J]. Journal of Intelligent Information Systems,2006,27(1):5-19
[17]谢达境.输编电设备状态检修工作在国内外的开展现状[J].企业技术开友：下,2011,(12)：105-106
[18]许婧,王晶,高峰等.电力设备状态检修技术研究综述[J].电网技术,2000,24(8)：48-52
[19]Ogawa T, Umeda Y, Inamura T. Framework for systematic evaluation of life cycle strategy by means of life cycle simulation[J]. Environmentally Conscious Design and Inverse Manufacturing,2003,1B(8):198-205
[20]郑爱霞.江苏电网输变电设备的状态检修策略研究[D].东南大学,2006
[21]刘建敏.基于生产MIS的变电设备状态检修智能分析系统[D].天津工业大学,2004
[22]李明华,严璋,刘春文等.我国供电系统状态检修开展状况统计[J].中国电力,2005,38(12)：33-36
[23]赵嵩正.设备检修管理培训教材.中国设备管理培训中心,1998
[24]Liao Zhiwei, Wen Fushuan, Guo Wenxin, etc. An analytic model and optimization technique based methods for fault diagnosis in power systems[A].3rd International Conference on Deregulation and Restructuring and Power Technologies[C],2008. DRPT 2008
[25]Ming Ge, R.Du, Guicai Zhang, et al. Fault diagnosis using support vector machine with an application in sheet metal stamping operations[A]. Mechanical Systems and Signal Processing[C].2004, (18):143-159
[26]Jain M.Babita, Jain Amit, Srinivas M.B.. A web based expert system shell for fault diagnosis and control of power system equipment[A]. Proceedings of 2008 International Conference on Condition Monitoring and Diagnosis[C], 2008. CMD 2008
[27]周东华,叶银忠.现代故障诊断与容错控制[M].北京,清华大学出版社,2000
[28]Eitel J M, Lauriapeter J.Duehessi. A methodology for developing baysian networks:an application to information technology(IT) implementation [J]. Eropean Joural of Operational Researeh,2007,179:234-252
[29]WEI Liuhong, GUO Wenxin, WEN Fushuan, et al. Waveform matching approach for fault diagnosis of a high-vohage transmission line employing harmony search algorithm [J]. IET Generation, Transmission & Distribution, 2010,4(7):801-809
[30]吴洪艳,高国宏.基于数据驱动的工业过程故障诊断综述[J].科技广场, 2010,(6)：67-69
[31]HUANG Jingbo, MU Longhua. Fault diagnosis of substation based on Petri nets technology[A]. Proceedings of 2006 International Conference on Power System Technology[C], October 22-26,2006, Chongqing, China:1-5
[32]应宇乾.基于数据驱动的烧结处理过程建模和控制[D].浙江大学,2012
[33]Darema F. Grid computing and beyond:the context of dynamic data driven applications systems [J]. Proceedings of the IEEE,2005,93(3):692-697
[34]柴天佑,刘德荣.基于数据的控制、决策、调度与故障诊断[C].第33期双清论坛-基于数据的控制、决策、调度与故障诊断论文集.2008：Ⅲ-Ⅳ
[35]侯忠生,许建新.数据驱动控制理论及方法的回顾和展望[J].自动化学报,2009,35(6)：650-667
[36]HOU Z-S, XU J-X. On Data-driven Control Theory:the State of the Art and Perspective[J]. Acta Automatica Sinica,2009,35(6):650-67
[37]陈琳.基于数据驱动的多源信息融合技术研究[D].杭州电子科技大学,2011
[38]赵文浩,阎威武.基于数据驱动的故障诊断研究[J].微计算机信,2010,(28)：10-17
[39]吕宁.基于数据驱动的故障诊断模型及算法研究[D].哈尔滨理工大学,2009
[40]胡友强.数据驱动的多元统计故障诊断及应用[D].重庆大学,2010
[41]王汝杰,石博强.现代设备管理[M].北京：冶金工业出版社,2007
[42]王慧芳,杨荷娟,何奔腾等.输变电设备状态故障率模型改进分析[J].电力系统自动化,2011,35(16)：27-31,43
[43]夏虹,刘永阔,谢春丽.设备故障诊断技术[M].哈尔滨市：哈尔滨工业大学出版社.2010
[44]Nemeth B., Benyo T., Jager A., Csepes G., Woynarovich G. Complex diagnostic methods for lifetime extension of power transformers[C].2008 IEEE International Symposium on Electrical Insulation,2008. DRPT 2008
[45]周东华,胡艳艳.动态系统的故障诊断技术[J].自动化学报,2009,35(6)：748-758
[46]张斌,张薇薇.机械设备故障诊断技术概述[J].建筑机械化,2005,8(15)：14-15
[47]孙秋野,张化光,李钟旭等.数据驱动的不确定信息下智能电网故障诊断[C].2011年中国智能自动化会议论文集.2011：905-911
[48]杜运成,石红瑞,杨晓波等.控制系统故障诊断方法综述[J].工业仪表与 自动化装置,2008,(5)：9-13,68
[49]李晗,萧德云.基于数据驱动的故障诊断方法综述[J].控制与决策,2011,01：1-9+16
[50]周福娜,汤天浩,文成林等.一种数据驱动的多故障诊断方法研究[J].控制工程,2008,15(4)：470-473
[51]Camil Fuchs, Ron S Kenett. Multivariate quality control:Theory and application [M]. New York:Marcel Dekker,1998
[52]Necip Doganaksoy, Frederick W Faltin, William T Tucker. Identification of out of control quality characteristics ina multivariate manufacturing environment[J]. Communi-cations in Statistics-Theory and Methods,1991, 20(9):2775-2790
[53]Rolf Isermann. Model based fault detection and diagnosis methods[J]. Annual Reviews in Control,2005,29(1):71-85
[54]Necip Doganaksoy, Frederick W Faltin, William T Tucker. Identification of out of control quality characteristics in a multivariate manufacturing environment[J]. Communi-cations in Statistics-Theory and Methods,1991, 20(9):2775-2790
[55]杨廷方.变压器在线监测与故障诊断新技术的研究[D].华中科技大学,2008.
[56]Drif M, Benouzza N, Kraloua B, Bendiabdellah A, Dente JA. Squirrel cage rotor faults detection in induction motorutilizing stator power spectrum approach. In:Proceedings of International Conference on Power Electronics, Machines, and Drives. London, UK:IEEE,2002.133-138
[57]Rolf Isermann. Model based fault detection and diagnosis methods[J]. Annual Reviews in Control,2005,29(1):71-85
[58]Venkatasubramanian V, Rengaswamy R, Yin K, et al. Areview of process fault detection and diagnosis:Part Ⅲ[J]. Computers and Chemical Engineering,2003, 27(3):327-346
[59]LEE H J, AHN B S, PARK Y M. A fault diagnosis expert system for distribution substations[J]. IEEE Trans on Power Delivery,2000,15(1):92-97
[60]闫如忠.基于Multi-agent分布式监控和智能诊断模型研究与应用[D].上海大学,2004
[61]H Ishibuchi, T Yamamoto. Fuzzy rule selection by multi-objective genetic local search algorithms and rule evaluation measures in data mining[J]. Fuzzy Sets and Systems,2004,141(1):59-88
[62]R.Schank, R.P.Abelson. Scripts, Plans, Goals and Understanding[M]. Edbaum, Hillsdale, New Jersey, US.1977
[63]R. Schank. Dynamicmemory:a theory of reminding and learning in computers and people[M]. Cambridge University Press, Cambridge, UK.1982
[64]J. Kolodner, Case-Based Reasoning[M].. Morgan Kaufmann Publishers, Inc, ISBN 1-55860-237-2,1993
[65]HONG Sheng, ZHAO Sufeng. A novel substation fault diagnosis approach based on RS and ANN and ES[C]. Proceedings of 2006 International Conference on Communications, Circuits and Systems Proceedings, June 25-28, 2006, Guilin, China:2124-2127
[66]Ibrahim W. R. A., Morcos M. M. An adaptive fuzzy self-learning technique for prediction of abnormal operation of electrical systems[J]. IEEE Transactions on Power Delivery,2006,21(4):1770-1777
[67]B. Smyth, M. T. Keane and P. Cunninghain. Hierarchical Case-Based Reasoning: Integrating Case-Based and Decompositional Problem-Solving Techniques for Plant-Control SoRware Design[M], IEEE Transactions on Knowledge and Data Engineering,2001,13(5):793-812
[68]Yi D, Peichao Z, Weiyong Y. A substation fault diagnosis system based on case-based reasoning and rule-based reasoning[J]. POWER SYSTEM TECHNOLOGY-BEIJING-,2004,28(1):34-37
[69]史忠植.知识发现[M].北京：清华大学出版社,2002
[70]成睿星.基于范例推理技术的服务组合研究[D].北京邮电大学,2007
[71]Redmond M. Distributed cases for case-based reasoning:Facilitating use of multiple cases[C]. Proceedings of A AAI.1990,90:304-309
[72]Ballesteros M, Martin R, Diaz-Agudo B. Jadaweb. A cbr system for cooking recipes[C] Proceedings of 18th International Conference on Case-based Reasoning workshop on Computer Cooking Contest. ICCBR, Alessandria, Italy.2010:179-188
[73]Michel Manago, Ralph Bergrnann,& Stefan Wessand Ralph Traphoener: CASUEL:A Common Case Representation Language-Version 2.0 ESPRIT-Project INRECA,1994, available from http://www.Agr.informatik.unikl.de/ bergmann/casuel/
[74]Enric Plaza:An Overview of the Noos Representation Language,1997, available from http://www.iiia.csic.es/enric/noos/Overview/
[75]J.Shlens.(2005,April) A tutorial on principal component analysis. [Online]. Available:http://www.cs.cmu.edu/elaw/papers/pca.pdf
[76]王国胜.核函数的性质及其构造方法[J].计算机科学,2006,33(6)：172-174,178
[77]北京大学数学系.高等代数[M].北京：高等教育出版社,2004
[78]Scholkopf B, Smola AJ. Learning with Kernel[M]. MA:PressCambridge,2000
[79]S. Poyhonen, A. Arkkio, P. Jover, et al. Coupling pairwise support vector machines for fault classification[C]. Control Engineering Practice.2005, (13): 759-769
[80]Solbing F S. Multi-class pattern recognition problems with tree structured support vector machines[C]. LNCS 2191:DAGM,2001:283-290
[81]Hsu Chih-Wei, Lin Chih-Jen. A comparison of methods for multi-class suppot vector machines[J]. IEEE Transactions on Neural Net works.2002.13(2): 415-425
[82]Angulo C, Xavier P, Catala A. K-SVCR a support vector machine fur multi-class classification[J]. Neuro computing,2003,55(1-2):57-77
[83]Platt J C, Cristianini N, Shawe-Taylor J, Large margin DAGs for multi-class classification[C]. Advances in Neural Information Processing Systems,2000, 12:547-553
[84]Francis E.H.Tay, Lijuan Cao. Application of Support Vector Machines in Financial Time Series Forecasting[J]. Omega.2001, Vol.29:309-317
[85]Kyong-jae Kim. Financial Time Series Forecasting Using Support Vector Machines [J]. Neuro computing.2003, Vol.55:307-319
[86]Ma Gang, Wu Kehe, Zhang Tong, Li Wei. A Flexible Policy-Based Access Control Model for Workflow. PRZEGLAD ELEKTROTECHNICZNY,2012, 88,3b:67-71
[87]Vlada Maten, Sanjeev Krishnan, Linda DeMichiel.基于组件的J2EE平台开发[M].北京：清华大学出版社,2004：45-88
[88]马廷准,赵亚伟,刘忠.用EJB开发J2EE应用[J].计算机应用,2002,22(4)：109-111
[89]Cheong S. Support vector machines with Binary tree architecture for multi-class classification[J]. Neural Information Processing-Letters and Reviews,2004
[90]孙禄,卢潇,张强。基于Java Bean组件技术的模型表示研究[J].电子设计工程,2010,18(10)：9-12
[91]孙莹,陈松乔.基于Web和组件技术DSS模型设计与实现[J].计算机工程,2004,30(10)：74-76
[92]童恒庆,聂会琴,李锡喆.CORBA/COM/EJB三种组件模型的分析与比较[J].计算机应用研究,2004,21(4)：66-67,94
[93]Sun Microsystems, "JSR 220:Enterprise JavaBeans, Version 3.0, "Early draft 2 edition (2005), http://java.sun.com/products/ejb/docs.html
[94]Gergely Varr'o. Implementing an EJB3-specific graph transformation plugin by database independent queries[C]. In Proc. of the Fifth International Workshop on Graph Transformation and Visual Modelling Techniques, ENTCS, pages 115-126. Elsevier,2006
[95]Scott Stark, JBoss, Inc. JBoss Administration and Development Third Edition[M]. Beijing:Electronic Industry Press,2004:55-100
[96]Eric Jendrock. Java EE 5 Definitive Guide, Beijing:Machinery Industry Press, 2008:44-98
[97]Wu Kehe, Ma Gang, Zhang Tong Design and implementation of processes mandatory running control[J]. Tongxin Xuebao/Journal on Communications, 2009,30(10A):150-153
[98]Vlada Maten, Sanjeev Krishnan, Linda DeMichiel. Design of J2EE Platform based on Componet[M]. Beijing:Tsinghua University Press,2004:45-88
[99]Kehe Wu, Gang Ma, Tong Zhang. Research and design of runtime software monitoring and control model [C]. Software Technology and Engineering (ICSTE),2010, v1,146-149
[100]Roman, E. et al. Mastering Enterprise JavaBeans 3. Wiley,2004:112-189
[101]王芳.基于数据挖掘的变压器故障诊断研究[D].2006
[102]孙才新,陈伟根,李俭等.电器设备油中气体在线监测与故障诊断技术[M].北京：科学出版社.2003：239-241
[103]高文胜,严璋,谈克雄.基于油中溶解气体分析的电力变压器绝缘故障诊断方法[M].电工电能新技术,2000,19(1)：22-26
[104]GBT7252-2001,标准变压器油中溶解气体分析和判断导则[s],中华人民共和国：国家质量监督检验检疫总局,2002
[105]廖怀东.变压器油色谱分析及故障判断[J].高电压技术,2006,32(1)：112-113
[106]赖永标,乔春生,刘开云等.支持向量机在围岩稳定性分类中的应用[J].水利学报,2006,37(9)：1092-1096