高速列车双通道速度传感器故障检测与隔离研究
|
摘要
高速列车双通道速度传感器在随列车高速运行时受粉尘、振动、温度变化等因素影响,易出现车轴故障检测虚漏警或误诊等问题。现阶段工程中常利用测试对比等方法检测速度传感器状态,但无法及时监测且易产生虚报警。因此,利用改进主元分析法(PCA)和改进重构贡献图法(IRBCP)建立全新的速度传感器故障检测隔离系统。首先,利用可检测故障幅值选择合适的故障检测统计量并细化,再利用基于组合最大化思想的改进重构贡献图法,确保单故障、多故障均可隔离。以高速列车双通道速度传感器为例,实验验证了所提的策略能够满足常见速度传感器故障的检测隔离需求,准确有效。
The dual channel speed sensors installed on the axes of high-speed train are often influenced by the factors like dusts, vibration and temperature changes in high speed operation, which easily causes the problems of false or miss alarm, even wrong isolation of the axis fault. At present engineering, the working states of the speed sensors are mainly detected by testing and comparing, which cannot monitor the faults in time and false alarms occur easily. In this paper, a novel FDI system for speed sensors is proposed using the improved principal component analysis(PCA) and the improved reconstruction based contribution plots(IRBCP). Firstly, the detectable fault amplitude is used to select appropriate fault detection statistic features, which is then refined. Then the modified reconstruction based contribution plot based on the idea of combination maximization is adopted to ensure the fault detection and isolation of single or multiple faults. Taking the dual-channel speed sensors on high-speed train as an example, experiments were conducted, and experiment results show that the proposed strategy can satisfy the FDI requirements of the commonly used speed sensors, and is accurate and effective.
The dual channel speed sensors installed on the axes of high-speed train are often influenced by the factors like dusts, vibration and temperature changes in high speed operation, which easily causes the problems of false or miss alarm, even wrong isolation of the axis fault. At present engineering, the working states of the speed sensors are mainly detected by testing and comparing, which cannot monitor the faults in time and false alarms occur easily. In this paper, a novel FDI system for speed sensors is proposed using the improved principal component analysis(PCA) and the improved reconstruction based contribution plots(IRBCP). Firstly, the detectable fault amplitude is used to select appropriate fault detection statistic features, which is then refined. Then the modified reconstruction based contribution plot based on the idea of combination maximization is adopted to ensure the fault detection and isolation of single or multiple faults. Taking the dual-channel speed sensors on high-speed train as an example, experiments were conducted, and experiment results show that the proposed strategy can satisfy the FDI requirements of the commonly used speed sensors, and is accurate and effective.
引文
[1] 宋云,朱明年.基于神经网络的地铁列车速度传感器故障诊断方法[J].城市轨道交通研究,2014,17(5):94- 97.SONG Y, ZHU M N. Subway sensor fault diagnosis based on radial neural network[J]. Urban Mass Transit, 2014,17(5):94- 97.
[2] 严加斌,朱峰,李军,等.高速动车组速度传感器的电磁干扰测试与分析[J].电子测量与仪器学报2015,29(3):433- 438.YAN J B, ZHU F, LI J, et al. Electromagnetic interference measurement and analysis of high-speed electric multiple units speed sensor[J]. Journal of Electronic Measurement and Instrumentation, 2015,29(3):433- 438.
[3] 常玉清,王姝,王福利,等.基于多PCA模型的过程监测方法[J].仪器仪表学报,2014,35(4):901- 908.CHANG Y Q, WANG SH, WANG F L, et al. Process monitoring method based on multiple PCA models[J]. Chinese Journal of Scientific Instrument, 2014, 35(4): 901- 908.
[4] 王前,于嘉成,宁永杰.基于MFCC与PCA的滚动轴承故障诊断[J].组合机床与自动化加工技术,2017(12):103- 105.WANG Q, YU J CH, NING Y J. Fault diagnosis of rolling bearing based on MFCC and PCA[J]. Modular Machine Tool & Automatic Manufacturing Technique, 2017(12):103- 105.
[5] 马峻,赵飞乐,徐潇,等. MRA-PCA-PSO组合优化BP神经网络模拟电路故障诊断研究[J].电子测量与仪器学报,2018,32(3):73- 79.MA J, ZHAO F Y, XU X, et al. Fault diagnosis of analog circuit of BP neural network optimized by MRA-PCA-PSO combination[J]. Journal of Electronic Measurement and Instrumentation, 2018,32(3):73- 79.
[6] 吴良国,魏书宁,周棒棒,等.基于PCA降维的分层超限学习机手势识别方法[J].电子测量技术,2017,40(3):82- 88.WU L G, WEI SH N, ZHOU B B, et al. Hierarchical extreme learning machine gesture recognition method based on PCA dimension reduction[J]. Electronic Measurement Technology, 2017,40(3):82- 88.
[7] 马云鹏,李庆武,何飞佳.金属表面缺陷自适应分割算法[J].仪器仪表学报,2017,38(1):245- 251.MA Y P, LI Q W, HE F J. Adaptive segmentation algorithm for metal surface defects[J]. Chinese Journal of Scientific Instrument, 2017, 38(1):245- 251.
[8] ALCALA C F. JOE QIN S. Analysis and generalization of fault diagnosis methods for process monitoring[J]. Journal of Process Control, 2011,21(3):322- 330.
[9] MACGREGOR J F. YOON S Y. Fault diagnosis with multivariate statistical models, part I: Using steady state fault signatures[J]. Journal of Process Control, 2001, 11(4):387- 400.
[10] ALCALA CF. JOE QIN S. Reconstruction-based contribution for process monitoring[J]. Automatica, 2009,45(7):1593- 1600.
[11] 唐永波,桂伟华,彭涛,等.基于重构贡献和灰度联熵的变压器诊断方法[J].仪器仪表学报,2012,33(1):132- 138.TANG Y B, GUI W H, PENG T, et al. Transformer fault diagnosis method based on reconstruction-based contribution and grey relation entropy[J]. Chinese Journal of Scientific Instrument, 2012,33(1):132- 138.
[12] 周桂法,王坚.基于神经网络的机车速度传感器故障诊断方法研究[J].机车电传动,2010(1):36- 38.ZHOU G F, WANG J. Research on locomotive speed sensor fault diagnosis method based on neural network[J]. Electric Drive for Locomotives, 2010(1):36- 38.
[13] 徐进学,李元,谢植.基于PCA的故障传感器重构的理论研究[J].仪器仪表学报,2003,24(增刊1):187- 188.XU J X, LI Y, XIE ZH. A study on faulty sensor reconstruction based on PCA[J]. Chinese Journal of Scientific Instrument, 2003,24(Suppl.1):187- 188.
[14] 叶昊,徐海鹏.基于重构传感器故障诊断贡献分析[J].清华大学学报(自然科学版),2012,52(1):36- 39.YE H, XU H P. Reconstruction-based contribution analysis for sensor fault diagnostics[J]. Journal of Tsinghua University (Science and Technology), 2012,52(1):36- 39.
[15] 高翔,白丽娜.基于限定记忆模式的递归PCA故障诊断法[J].仪器仪表学报,2006,27(增刊2):1726- 1727.GAO X, BAI L N. Fault diagnosis based on recursive PCA with restricted memory of patterns[J]. Chinese Journal of Scientific Instrument, 2006,27(Suppl.2):1726- 1727.
[16] 邱天,丁艳军,吴占松.基于主元分析的故障可检测性的统计指标比较[J].清华大学学报(自然科学版), 2006,46(8): 1447- 1450.QIU T, DING Y J, WU ZH S. Sensor fault detection statistics based on principal component analysis[J]. Journal of Tsinghua University (Science and Technology), 2006,46(8):1447- 1450.
[17] 郭小萍,杨猛,李元.基于改进重构贡献图的故障定位方法[J].仪器仪表学报,2015,36(5):1193- 1200.GUO X P, YANG M, LI Y. Modified reconstruction-based contribution plots for fault isolation[J]. Chinese Journal of Scientific Instrument, 2015,36(5):1193- 1200.
[18] JOE QIN S. Statistical process monitoring: Basics and beyond[J]. Journal of Chemometrics, 2003,17(8- 9):480- 502.
[19] DING S, ZHANG P, DING E, et al. On the application of PCA technique to fault diagnosis[J]. Tsinghua Science and Technology, 2010,15(2):138- 144.
[20] DUNIA R. JOE QIN S. A unified geometric approach to process and sensor fault identification and reconstruction: the unidimensional fault case[J]. Computers and Chemical Engineering, 1998,22(7- 8):927- 943.
[21] 王海清,宋执环,李平.改进PCA及其在过程监测与故障诊断中的应用[J].化工学报,2001,52(6):471- 475.WANG H Q, SONG ZH H, LI P. Improved PCA with application to process monitoring and fault diagnosis[J]. Journal of Chemical Industry and Engineering (China), 2001,52(6):471- 475.
[22] LIU J L, CHEN D S. Fault isolation using modified contribution plots[J]. Computers & Chemistry Engineering, 2014(61):9- 19.
[23] YUE H H,JOE QIN S. Reconstruction-based fault identification using combined index[J]. Industrial & Engineering Chemistry Research, 2001,40(20):4403- 4414.
[24] LI G,ALCALA C F, JOE QIN S, et al. Generalized reconstruction-based contributions for output relevant fault diagnosis with application to the Tennessee Eastman process[J]. IEEE Transaction on Control System Technology, 2011,19(5):1114- 1127.
[25] HE B, ZHANG J, CHEN T, et al. Penalized reconstruction-based multivariate contribution analysis for fault isolation[J]. Industrial & Engineering Chemistry Research, 2013,52(23):7784- 7794.
[26] 王世阳,毕祥军,王平.基于LabVIEW多通道应变采集系统设计[J].国外电子测量技术,2017,36(8):83- 87.WANG SH Y, BI X J, WANG P. Multichannel strain acquisition system based on LabVIEW[J]. Foreign Electronic Measurement Technology, 2017,36(8):83- 87.
[2] 严加斌,朱峰,李军,等.高速动车组速度传感器的电磁干扰测试与分析[J].电子测量与仪器学报2015,29(3):433- 438.YAN J B, ZHU F, LI J, et al. Electromagnetic interference measurement and analysis of high-speed electric multiple units speed sensor[J]. Journal of Electronic Measurement and Instrumentation, 2015,29(3):433- 438.
[3] 常玉清,王姝,王福利,等.基于多PCA模型的过程监测方法[J].仪器仪表学报,2014,35(4):901- 908.CHANG Y Q, WANG SH, WANG F L, et al. Process monitoring method based on multiple PCA models[J]. Chinese Journal of Scientific Instrument, 2014, 35(4): 901- 908.
[4] 王前,于嘉成,宁永杰.基于MFCC与PCA的滚动轴承故障诊断[J].组合机床与自动化加工技术,2017(12):103- 105.WANG Q, YU J CH, NING Y J. Fault diagnosis of rolling bearing based on MFCC and PCA[J]. Modular Machine Tool & Automatic Manufacturing Technique, 2017(12):103- 105.
[5] 马峻,赵飞乐,徐潇,等. MRA-PCA-PSO组合优化BP神经网络模拟电路故障诊断研究[J].电子测量与仪器学报,2018,32(3):73- 79.MA J, ZHAO F Y, XU X, et al. Fault diagnosis of analog circuit of BP neural network optimized by MRA-PCA-PSO combination[J]. Journal of Electronic Measurement and Instrumentation, 2018,32(3):73- 79.
[6] 吴良国,魏书宁,周棒棒,等.基于PCA降维的分层超限学习机手势识别方法[J].电子测量技术,2017,40(3):82- 88.WU L G, WEI SH N, ZHOU B B, et al. Hierarchical extreme learning machine gesture recognition method based on PCA dimension reduction[J]. Electronic Measurement Technology, 2017,40(3):82- 88.
[7] 马云鹏,李庆武,何飞佳.金属表面缺陷自适应分割算法[J].仪器仪表学报,2017,38(1):245- 251.MA Y P, LI Q W, HE F J. Adaptive segmentation algorithm for metal surface defects[J]. Chinese Journal of Scientific Instrument, 2017, 38(1):245- 251.
[8] ALCALA C F. JOE QIN S. Analysis and generalization of fault diagnosis methods for process monitoring[J]. Journal of Process Control, 2011,21(3):322- 330.
[9] MACGREGOR J F. YOON S Y. Fault diagnosis with multivariate statistical models, part I: Using steady state fault signatures[J]. Journal of Process Control, 2001, 11(4):387- 400.
[10] ALCALA CF. JOE QIN S. Reconstruction-based contribution for process monitoring[J]. Automatica, 2009,45(7):1593- 1600.
[11] 唐永波,桂伟华,彭涛,等.基于重构贡献和灰度联熵的变压器诊断方法[J].仪器仪表学报,2012,33(1):132- 138.TANG Y B, GUI W H, PENG T, et al. Transformer fault diagnosis method based on reconstruction-based contribution and grey relation entropy[J]. Chinese Journal of Scientific Instrument, 2012,33(1):132- 138.
[12] 周桂法,王坚.基于神经网络的机车速度传感器故障诊断方法研究[J].机车电传动,2010(1):36- 38.ZHOU G F, WANG J. Research on locomotive speed sensor fault diagnosis method based on neural network[J]. Electric Drive for Locomotives, 2010(1):36- 38.
[13] 徐进学,李元,谢植.基于PCA的故障传感器重构的理论研究[J].仪器仪表学报,2003,24(增刊1):187- 188.XU J X, LI Y, XIE ZH. A study on faulty sensor reconstruction based on PCA[J]. Chinese Journal of Scientific Instrument, 2003,24(Suppl.1):187- 188.
[14] 叶昊,徐海鹏.基于重构传感器故障诊断贡献分析[J].清华大学学报(自然科学版),2012,52(1):36- 39.YE H, XU H P. Reconstruction-based contribution analysis for sensor fault diagnostics[J]. Journal of Tsinghua University (Science and Technology), 2012,52(1):36- 39.
[15] 高翔,白丽娜.基于限定记忆模式的递归PCA故障诊断法[J].仪器仪表学报,2006,27(增刊2):1726- 1727.GAO X, BAI L N. Fault diagnosis based on recursive PCA with restricted memory of patterns[J]. Chinese Journal of Scientific Instrument, 2006,27(Suppl.2):1726- 1727.
[16] 邱天,丁艳军,吴占松.基于主元分析的故障可检测性的统计指标比较[J].清华大学学报(自然科学版), 2006,46(8): 1447- 1450.QIU T, DING Y J, WU ZH S. Sensor fault detection statistics based on principal component analysis[J]. Journal of Tsinghua University (Science and Technology), 2006,46(8):1447- 1450.
[17] 郭小萍,杨猛,李元.基于改进重构贡献图的故障定位方法[J].仪器仪表学报,2015,36(5):1193- 1200.GUO X P, YANG M, LI Y. Modified reconstruction-based contribution plots for fault isolation[J]. Chinese Journal of Scientific Instrument, 2015,36(5):1193- 1200.
[18] JOE QIN S. Statistical process monitoring: Basics and beyond[J]. Journal of Chemometrics, 2003,17(8- 9):480- 502.
[19] DING S, ZHANG P, DING E, et al. On the application of PCA technique to fault diagnosis[J]. Tsinghua Science and Technology, 2010,15(2):138- 144.
[20] DUNIA R. JOE QIN S. A unified geometric approach to process and sensor fault identification and reconstruction: the unidimensional fault case[J]. Computers and Chemical Engineering, 1998,22(7- 8):927- 943.
[21] 王海清,宋执环,李平.改进PCA及其在过程监测与故障诊断中的应用[J].化工学报,2001,52(6):471- 475.WANG H Q, SONG ZH H, LI P. Improved PCA with application to process monitoring and fault diagnosis[J]. Journal of Chemical Industry and Engineering (China), 2001,52(6):471- 475.
[22] LIU J L, CHEN D S. Fault isolation using modified contribution plots[J]. Computers & Chemistry Engineering, 2014(61):9- 19.
[23] YUE H H,JOE QIN S. Reconstruction-based fault identification using combined index[J]. Industrial & Engineering Chemistry Research, 2001,40(20):4403- 4414.
[24] LI G,ALCALA C F, JOE QIN S, et al. Generalized reconstruction-based contributions for output relevant fault diagnosis with application to the Tennessee Eastman process[J]. IEEE Transaction on Control System Technology, 2011,19(5):1114- 1127.
[25] HE B, ZHANG J, CHEN T, et al. Penalized reconstruction-based multivariate contribution analysis for fault isolation[J]. Industrial & Engineering Chemistry Research, 2013,52(23):7784- 7794.
[26] 王世阳,毕祥军,王平.基于LabVIEW多通道应变采集系统设计[J].国外电子测量技术,2017,36(8):83- 87.WANG SH Y, BI X J, WANG P. Multichannel strain acquisition system based on LabVIEW[J]. Foreign Electronic Measurement Technology, 2017,36(8):83- 87.