基于HMM与AIS混合模型的核动力设备故障诊断系统开发
|
摘要
本文以国防基础科研项目“核动力关键设备动态检测与故障分析技术研究”(项目编号:B0120060585)为基础,针对核动力设备具有系统复杂、可靠性要求极高、所积累的资料与故障样本少等特点,引入隐马尔可夫模型与人工免疫系统作为基本建模和识别工具,综合两者的优点,进行核动力设备状态监测与故障诊断,应用Visual C++开发出相应的核动力设备故障诊断系统。后期进行的大量实验验证了所开发系统的有效性、鲁棒性以及预报的实时性。本文主要研究内容如下:
第一章介绍了课题的研究背景和核动力设备故障诊断研究的意义,对核动力设备故障诊断技术的研究现状进行了概括,介绍了论文的特色与创新;最后给出了本论文的主要研究内容,并给出了论文的总体框架。
第二章对HMM和AIS的基本理论做了介绍,着重介绍了HMM和AIS的几种基本算法。研究了HMM和AIS在核动力设备故障诊断中的应用,并提出了一种改进的AIS算法。
第三章在介绍基于HMM与AIS的核动力设备故障诊断系统架构的基础上,应用Visual C++开发了基于HMM与AIS混合模型的核动力设备故障诊断系统,并对系统软件结构和功能模块以及软件设计时的一些关键技术进行了说明。
第四章通过实验对所开发的基于HMM与AIS的核动力设备故障诊断系统进行了验证,介绍了实验所用的装置及仪器,给出了具体的实验方案和一些典型的实验结果,并对实验结果进行了分析。
第五章总结了全文的研究成果,并对今后的工作提出了展望。本项研究对于促进核动力设备状态监测与故障诊断技术的进步,对于保证核动力设备安全可靠地运行、提高国防战斗力都具有重要的理论意义和实际应用价值。
Nuclear-powered system is a complicated system, which only accumulated less data and fault samples. In view of these characteristics, the Hidden Markov Models (HMMs) and Artificial Immune System (AIS), as the basic modeling and recognizing tools in this thesis, are introduced to implement condition monitoring and fault diagnosis technologies for key equipment of nuclear power system, based on the“Research on Condition Monitoring and Fault Diagnosis Technologies for Key Equipment of Nuclear Power System”(National Defense Basic Science Fund Project, No. B0120060585). Fault diagnosis software for key equipment of nuclear power system has been developed with Visual C++. The experiments testified the validation of the software, the robust of the system, real time forecasting. The main contents in this thesis includes:
Chapter one briefly introduced the meaning of studying the faults diagnosis technologies in nuclear power plant and summarized the developing and the current situations of it. Finally, given the research substance of this dissertation basing on this National Defense Basic Science Fund Project, and proposed the total frame and innovations of this dissertation.
Chapter two introduces the basic theories and algorithms of HMM and AIS in detail. The application methods for faults diagnosis of nuclear power equipment based HMM/AIS are given. Finally, a new algorithm by improving basic AIS algorithm is presented.
Chapter three introduces the fault diagnosis system of nuclear power equipment based on HMM and AIS. Visual C++ is used to develop this fault diagnosis software. Chapter four proposes experimental methods, and verifies the effect of this faults diagnosis system. The experiments are carried out on IS50-32-200 centrifugal pump. Chapter five sums up the main achievements and innovation in this thesis, and presents prospect.
This research can promote the progress of condition monitoring and fault diagnosis technologies, has important theoretical and practical value in guaranteeing the nuclear power system’s safe and reliable operations and increasing the combat effectiveness of the national defence.
第一章介绍了课题的研究背景和核动力设备故障诊断研究的意义,对核动力设备故障诊断技术的研究现状进行了概括,介绍了论文的特色与创新;最后给出了本论文的主要研究内容,并给出了论文的总体框架。
第二章对HMM和AIS的基本理论做了介绍,着重介绍了HMM和AIS的几种基本算法。研究了HMM和AIS在核动力设备故障诊断中的应用,并提出了一种改进的AIS算法。
第三章在介绍基于HMM与AIS的核动力设备故障诊断系统架构的基础上,应用Visual C++开发了基于HMM与AIS混合模型的核动力设备故障诊断系统,并对系统软件结构和功能模块以及软件设计时的一些关键技术进行了说明。
第四章通过实验对所开发的基于HMM与AIS的核动力设备故障诊断系统进行了验证,介绍了实验所用的装置及仪器,给出了具体的实验方案和一些典型的实验结果,并对实验结果进行了分析。
第五章总结了全文的研究成果,并对今后的工作提出了展望。本项研究对于促进核动力设备状态监测与故障诊断技术的进步,对于保证核动力设备安全可靠地运行、提高国防战斗力都具有重要的理论意义和实际应用价值。
Nuclear-powered system is a complicated system, which only accumulated less data and fault samples. In view of these characteristics, the Hidden Markov Models (HMMs) and Artificial Immune System (AIS), as the basic modeling and recognizing tools in this thesis, are introduced to implement condition monitoring and fault diagnosis technologies for key equipment of nuclear power system, based on the“Research on Condition Monitoring and Fault Diagnosis Technologies for Key Equipment of Nuclear Power System”(National Defense Basic Science Fund Project, No. B0120060585). Fault diagnosis software for key equipment of nuclear power system has been developed with Visual C++. The experiments testified the validation of the software, the robust of the system, real time forecasting. The main contents in this thesis includes:
Chapter one briefly introduced the meaning of studying the faults diagnosis technologies in nuclear power plant and summarized the developing and the current situations of it. Finally, given the research substance of this dissertation basing on this National Defense Basic Science Fund Project, and proposed the total frame and innovations of this dissertation.
Chapter two introduces the basic theories and algorithms of HMM and AIS in detail. The application methods for faults diagnosis of nuclear power equipment based HMM/AIS are given. Finally, a new algorithm by improving basic AIS algorithm is presented.
Chapter three introduces the fault diagnosis system of nuclear power equipment based on HMM and AIS. Visual C++ is used to develop this fault diagnosis software. Chapter four proposes experimental methods, and verifies the effect of this faults diagnosis system. The experiments are carried out on IS50-32-200 centrifugal pump. Chapter five sums up the main achievements and innovation in this thesis, and presents prospect.
This research can promote the progress of condition monitoring and fault diagnosis technologies, has important theoretical and practical value in guaranteeing the nuclear power system’s safe and reliable operations and increasing the combat effectiveness of the national defence.
引文
[1]高永钰.核电业的发展机遇.第一财经日报,2008,2,19
[2]王乃彦.核电站安全性分析.世界科技研究与发展,2002,24(6):14-17
[3] Kim M C. Seong P H. A computational model for knowledge-driven monitoring of nuclear power plant operators based on information theory. Reliability Engineering & System Safety, 2006, 91(3), 283–291.
[4] Jyoti K. Sinha . Vibration-based diagnosis techniques used in nuclear power plants: An overview of experiences. Nuclear Engineering and Design, 2008,Volume 238, Issue 9, Pages 2439-2452
[5]张松,庄文翠,臧希年.压水堆核电厂严重事故对策.核动力工程, 2007,28(6):101- 105
[6]刘永阔,夏虹等.核电设备状态监测与故障诊断系统的研究.原子能科学技术, 2008,42(3):200-205
[7] Nabeshima K, Suzudo T, Ohno T, Kudo K. Nuclear reactor monitoring with the combination of neural network and expert system. Mathematics and Computers in Simulation, 2002, 60(3-5):233-244
[8] Papadopoulos Y. Model-based system monitoring and diagnosis of failures using state charts and fault trees. Reliability Engineering and System Safety, 2003, 81(3): 325-341
[9] Seker S, Ayaz E, Turkcan E. Elman's recurrent neural network applications to condition monitoring in nuclear power plant and rotating machinery. Engineering Applications of Artificial Intelligence, 2003, 16(7-8): 647-656
[10] Vinod S, Babar A K, etc. Symptom based diagnostic system for nuclear power plant operations using artificial neural networks. Reliability Engineering and System Safety, 2003, 82(1):33-40
[11] Majumdar D. Artificial intelligence applications in the nuclear industry: An international view. Nuclear Engineering and Design, 1989, 113(2): 181-193
[12]陈志辉.船用核动力故障诊断专家系统.哈尔滨工程大学硕士学位论文.2005
[13]刘永阔.基于模糊神经网络的船用核动力装置故障诊断方法研究.哈尔滨工程大学硕士学文论文.2004
[14]于宏丽.旋转机械故障数据库与故障精密诊断系统开发.华北电力大学硕士学文论文.2004
[15]庞凤阁,彭敏俊.船舶核动力装置.哈尔滨:哈尔滨工程大学出版社,2000
[16]苏杰.核动力装置运行支持数据库技术的研究.哈尔滨工程大学硕士学位论文.2004
[17] Levinson S E, etc. An introduction to the application of the theory of probabilistic functions of a Markov process to automatic speech recognition. Bell Syst Tech. Apr. 1983:1035-1074
[18] Rabiner L R, etc. An introduction to the hidden Markov models. IEEE ASSP Mag. Jan. 1986:4-16
[19] Galindo P L, etc. Competitive forward-backward algorithm(CFB). IEE Conference Publication, 1995,409:82-85
[20] Patel S.Lower-complexity viterbi algorithm. Speech ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing-Proceedings, 1995,1:592-595
[21]战普明.语音识别隐马尔可夫模型的改进.电子学报,1994,22(4):9-15
[22] Wallhoff F, Eickeler S, Rigoll G. A comparison of discrete and continuous output modeling techniques for a pseudo-2D hidden Markov model face recongnition system.Proceedings of International Conference on Image Processing.2001,2:685-688
[23]谢锦辉.隐Markov模型(HMM)及其在语音处理中的应用.武汉:华中理工大学出版社,1995
[24] Hariharan R, Viikki O. An integrated study of speaker normalisation and HMM adaptation for noise robust speaker-independent speech recognition. Speech Communication, 2002, 37(3-4): 349-361
[25] Bengio S. Multimodal speech processing using asynchronous Hidden Markov Models. Information Fusion, 2004, 5(2): 81-89
[26]王实,高文,李锦涛.基于隐马尔可夫模型的兴趣迁移模式发现.计算机学报,2001,24(2):152-157
[27] Stein D W, etc. Applying hidden Markov models to radar detection in clutter. IEE Conference Publication, 1997,449:589-590
[28] Lee J M, Kim S J, etc. Diagnosis of mechanical fault signals using continuous hidden Markov model . Journal of Sound and Vibration, 2004, 276(3-5): 1065-1080
[29] Khar H C, etc. Recent applications of hidden markov models in computational biology. Genomics Proteomics Bioinformatics, 2004,2(2):84-96
[30] Ocak H, Loparo K A. HMM-based fault detection and diagnosis scheme for rolling element bearings. Journal of Vibration and Acoustics, Transactions of the ASME, 2005,127(4):299-306
[31]赵辉,张春良.基于HMM的核动力旋转机械故障诊断的研究.噪声与振动控制,2007,27(6):73-79
[32]肖人彬,王磊.人工免疫系统:原理、模型、分析及展望.计算机学报,2002,25(12):1281-1293
[33] Farmer J D, etc. The immune system, adaptation, and machine learning. Physica 22D,1986: 187-204
[34]焦李成,杜海峰.人工免疫系统进展与展望.电子学报, 2003, 31(10): 1540-1548
[35] E Hart,J Timmis.Application Areas of AIS:The Past,the Present and the Future. In Proc. The 4rd International Conference on Artificial Immune Systems (ICARIS-2005),2005,3267:483-497
[36] Luh G C, Cheng W C. Immune model-based fault diagnosis. Mathematics and Computers in Simulation, 2005, 67(6) : 515-539
[37]孙勇智,韦巍.人工免疫系统及其应用.计算机工程, 2003, 29(15):1-2,62
[38]辛长安,梅林编著. VC++编程技术与难点剖析.北京:清华大学出版社, 2002.4
[39]何强,何英编著. MATLAB扩展编程.北京:清华大学出版社, 2002.6
[40]辛菁,刘丁,杨延西,刘涵.基于matcom的VC与MATLAB集成应用研究.现代电子技术. 2002(7):52-54
[2]王乃彦.核电站安全性分析.世界科技研究与发展,2002,24(6):14-17
[3] Kim M C. Seong P H. A computational model for knowledge-driven monitoring of nuclear power plant operators based on information theory. Reliability Engineering & System Safety, 2006, 91(3), 283–291.
[4] Jyoti K. Sinha . Vibration-based diagnosis techniques used in nuclear power plants: An overview of experiences. Nuclear Engineering and Design, 2008,Volume 238, Issue 9, Pages 2439-2452
[5]张松,庄文翠,臧希年.压水堆核电厂严重事故对策.核动力工程, 2007,28(6):101- 105
[6]刘永阔,夏虹等.核电设备状态监测与故障诊断系统的研究.原子能科学技术, 2008,42(3):200-205
[7] Nabeshima K, Suzudo T, Ohno T, Kudo K. Nuclear reactor monitoring with the combination of neural network and expert system. Mathematics and Computers in Simulation, 2002, 60(3-5):233-244
[8] Papadopoulos Y. Model-based system monitoring and diagnosis of failures using state charts and fault trees. Reliability Engineering and System Safety, 2003, 81(3): 325-341
[9] Seker S, Ayaz E, Turkcan E. Elman's recurrent neural network applications to condition monitoring in nuclear power plant and rotating machinery. Engineering Applications of Artificial Intelligence, 2003, 16(7-8): 647-656
[10] Vinod S, Babar A K, etc. Symptom based diagnostic system for nuclear power plant operations using artificial neural networks. Reliability Engineering and System Safety, 2003, 82(1):33-40
[11] Majumdar D. Artificial intelligence applications in the nuclear industry: An international view. Nuclear Engineering and Design, 1989, 113(2): 181-193
[12]陈志辉.船用核动力故障诊断专家系统.哈尔滨工程大学硕士学位论文.2005
[13]刘永阔.基于模糊神经网络的船用核动力装置故障诊断方法研究.哈尔滨工程大学硕士学文论文.2004
[14]于宏丽.旋转机械故障数据库与故障精密诊断系统开发.华北电力大学硕士学文论文.2004
[15]庞凤阁,彭敏俊.船舶核动力装置.哈尔滨:哈尔滨工程大学出版社,2000
[16]苏杰.核动力装置运行支持数据库技术的研究.哈尔滨工程大学硕士学位论文.2004
[17] Levinson S E, etc. An introduction to the application of the theory of probabilistic functions of a Markov process to automatic speech recognition. Bell Syst Tech. Apr. 1983:1035-1074
[18] Rabiner L R, etc. An introduction to the hidden Markov models. IEEE ASSP Mag. Jan. 1986:4-16
[19] Galindo P L, etc. Competitive forward-backward algorithm(CFB). IEE Conference Publication, 1995,409:82-85
[20] Patel S.Lower-complexity viterbi algorithm. Speech ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing-Proceedings, 1995,1:592-595
[21]战普明.语音识别隐马尔可夫模型的改进.电子学报,1994,22(4):9-15
[22] Wallhoff F, Eickeler S, Rigoll G. A comparison of discrete and continuous output modeling techniques for a pseudo-2D hidden Markov model face recongnition system.Proceedings of International Conference on Image Processing.2001,2:685-688
[23]谢锦辉.隐Markov模型(HMM)及其在语音处理中的应用.武汉:华中理工大学出版社,1995
[24] Hariharan R, Viikki O. An integrated study of speaker normalisation and HMM adaptation for noise robust speaker-independent speech recognition. Speech Communication, 2002, 37(3-4): 349-361
[25] Bengio S. Multimodal speech processing using asynchronous Hidden Markov Models. Information Fusion, 2004, 5(2): 81-89
[26]王实,高文,李锦涛.基于隐马尔可夫模型的兴趣迁移模式发现.计算机学报,2001,24(2):152-157
[27] Stein D W, etc. Applying hidden Markov models to radar detection in clutter. IEE Conference Publication, 1997,449:589-590
[28] Lee J M, Kim S J, etc. Diagnosis of mechanical fault signals using continuous hidden Markov model . Journal of Sound and Vibration, 2004, 276(3-5): 1065-1080
[29] Khar H C, etc. Recent applications of hidden markov models in computational biology. Genomics Proteomics Bioinformatics, 2004,2(2):84-96
[30] Ocak H, Loparo K A. HMM-based fault detection and diagnosis scheme for rolling element bearings. Journal of Vibration and Acoustics, Transactions of the ASME, 2005,127(4):299-306
[31]赵辉,张春良.基于HMM的核动力旋转机械故障诊断的研究.噪声与振动控制,2007,27(6):73-79
[32]肖人彬,王磊.人工免疫系统:原理、模型、分析及展望.计算机学报,2002,25(12):1281-1293
[33] Farmer J D, etc. The immune system, adaptation, and machine learning. Physica 22D,1986: 187-204
[34]焦李成,杜海峰.人工免疫系统进展与展望.电子学报, 2003, 31(10): 1540-1548
[35] E Hart,J Timmis.Application Areas of AIS:The Past,the Present and the Future. In Proc. The 4rd International Conference on Artificial Immune Systems (ICARIS-2005),2005,3267:483-497
[36] Luh G C, Cheng W C. Immune model-based fault diagnosis. Mathematics and Computers in Simulation, 2005, 67(6) : 515-539
[37]孙勇智,韦巍.人工免疫系统及其应用.计算机工程, 2003, 29(15):1-2,62
[38]辛长安,梅林编著. VC++编程技术与难点剖析.北京:清华大学出版社, 2002.4
[39]何强,何英编著. MATLAB扩展编程.北京:清华大学出版社, 2002.6
[40]辛菁,刘丁,杨延西,刘涵.基于matcom的VC与MATLAB集成应用研究.现代电子技术. 2002(7):52-54