一种数据驱动的旋转机械早期故障检测模型构建和应用研究

doi:10.3901/JME.2020.16.022

机械工程学报 ›› 2020, Vol. 56 ›› Issue (16): 22-32.doi: 10.3901/JME.2020.16.022

扫码分享

一种数据驱动的旋转机械早期故障检测模型构建和应用研究

王庆锋^1,2, 卫炳坤^1,2, 刘家赫^1,2, 马文生¹, 许述剑³

1. 北京化工大学高端机械设备健康监控及自愈化北京市重点实验室北京 100029;
2. 北京化工大学发动机健康监控及网络化教育部重点实验室北京 100029;
3. 中国石油化工股份有限公司青岛安全工程研究院青岛 266000

收稿日期:2020-02-25 修回日期:2020-04-29 出版日期:2020-08-20 发布日期:2020-10-19
通讯作者: 王庆锋(通信作者),男,1972年出生,博士,副研究员。主要研究方向为设备动态监测、诊断与维护;故障诊断与自愈;在役再制造;装置可靠性与风险评估。E-mail:wqf2422@163.com
作者简介:卫炳坤,男,1995年出生。主要研究方向为故障诊断与健康管理。E-mail:weibingkun0417@163.com
基金资助:
中国工程院咨询（2020-XY-1）、中国石化科技部研发（319022-1）和重庆市科委技术创新与应用示范（cstc2019jszx-cyzdX0167）资助项目。

Research on Construction and Application of Data-driven Incipient Fault Detection Model for Rotating Machinery

WANG Qingfeng^1,2, WEI Bingkun^1,2, LIU Jiahe^1,2, MA Wensheng¹, XU Shujian³

1. Beijing key laboratory of Health monitoring and Self-recovery of High-end Machinery Equipment, Beijing University of Chemical Technology, Beijing 100029;
2. Key Laboratory of Engine Health Monitoring and Networking Ministry of Education, Beijing University of Chemical Technology, Beijing 100029;
3. Sinopec Qingdao Research Institute of Safety Engineering, Qingdao 266000

Received:2020-02-25 Revised:2020-04-29 Online:2020-08-20 Published:2020-10-19

摘要/Abstract

摘要： 传统在线监测系统未能实现早期故障预警，旋转机械状态劣化评估采用固定阈值分级报警方法，存在较多的误报警和漏报警现象，难以指导企业设备预测性维修开展，设备运行安全性、可靠性、利用率难以保障。立足于工程应用，研究基于小波包分解、动态核主成分分析、T²统计分析、Beta分布预警控制限自学习等技术，构建数据驱动基于振动信号分析的旋转机械早期故障检测模型。应用辛辛那提大学智能维修系统中心滚动轴承试验数据和中国某石化公司加氢裂化装置P3409A离心泵轴承“运转到坏”的在线监测振动数据，对构建的设备早期故障检测模型进行验证，结果表明，构建的设备早期故障检测模型，相比传统固定阈值分级报警方法，能够检测滚动轴承早期故障并实现早期故障准确告警，能够有效降低错误报警率和漏报警率。构建的基于振动信号的旋转机械早期故障检测模型，只需要知道监测部件正常运行状态历史数据，无需外部专家支持，实时数据驱动即可实现早期故障检测预警。

关键词: 小波包分解, 动态核主成分分析, 监测统计量, 早期故障检测, 预测性维修

Abstract: The traditional online monitoring system could not realize incipient fault warning, and the fixed threshold grading alarm method which is used to evaluate the machine degradation status exist many false alarms and missed alarms. Excessive false alarm rate and missed alarm rate are difficult to guide the enterprises to carry out predictive maintenance of rotating machinery, and are difficult to guarantee its running safety, reliability and utilization. In order to meet the needs of engineering applications, a data-driven incipient fault detection and warning model has been built based on the technologies such as wavelet packet decomposition (WPD), dynamic kernel principal component analysis (DKPCA), T² statistical analysis, Beta distribution control limit and so on. The incipient fault detection model has been validated by the rolling bearing vibration data from Center for Intelligent Maintenance Systems (IMS) of University of Cincinnati and by the "run to failure" online monitoring vibration data from P3409A centrifugal pump bearing of PetroChina certain hydrocracking unit. Compared with the traditional fixed threshold grading alarm method, the verified results show that the model can detect the incipient fault of rolling bearings and can realize accurate incipient fault warning, and can reduce the false alarm rate and missed alarm rate effectively. The incipient fault detection and warning model is driven by real-time vibration signals, and it works only need the historical data collected under normal operating status of key components of rotating machinery.

Key words: wavelet packet decomposition, dynamic kernel principal component analysis, monitoring statistics, incipient fault detection and warning, predictive maintenance

中图分类号:

TH17

王庆锋, 卫炳坤, 刘家赫, 马文生, 许述剑. 一种数据驱动的旋转机械早期故障检测模型构建和应用研究[J]. 机械工程学报, 2020, 56(16): 22-32.

WANG Qingfeng, WEI Bingkun, LIU Jiahe, MA Wensheng, XU Shujian. Research on Construction and Application of Data-driven Incipient Fault Detection Model for Rotating Machinery[J]. Journal of Mechanical Engineering, 2020, 56(16): 22-32.

参考文献

[1] MIN J H, KIM D W,PARK C Y. Demonstration of the validity of the early warning in online monitoring system for nuclear power plants[J]. Nuclear Engineering and Design,2019,349:56-62.
[2] ISO 13381-2015,Condition monitoring and diagnostics of machines-prognostics(Part 1):General Guidelines:International Standards Organization[S]. 2015-09.
[3] 张学延. 汽轮发电机组振动诊断[M]. 北京:中国电力出版社,2008. ZHANG Xueyan. Vibration diagnosis of turbo generator sets[M]. Beijing:China Electric Power Press,2008.
[4] SIKORSKA J Z,HODKIEWICZ M,MA L. Prognostic modelling options for remaining useful life estimation by industry[J]. Mechanical Systems & Signal Processing,2011,25(5):1803-1836.
[5] VELDMAN J,KLINGENBERG W,WORTMANN H. Managing condition-based maintenance technology:A multiple case study in the process industry[J]. Journal of Quality in Maintenance Engineering,2011,17(1):40-62.
[6] KATIPAMULA S,BRAMBLEY M. Methods for fault detection,diagnostics,and prognostics for building systems:A review,Part I[J]. HVAC&R Research,2005,11(1):3-25.
[7] BARALDI P,DIMAIO F,TURATI P,et al. Robust signal reconstruction for condition monitoring of indu-strial components via a modified auto associative kernel regression method[J]. Mechanical Systems and Signal Processing,2015,60-61:29-44.
[8] TOBON-MEJIA D A,MEDJAHER K,ZERHOUNI N,et al. A data-driven failure prognostics method based on mixture of gaussians hidden markov models[J]. IEEE Transactions on Reliability,2012,61(2):491-503.
[9] MOSALLAM A,KAMAL M. Nonparametric time series modelling for industrial prognostics and health management[J]. International Journal of Advanced Manufacturing Technology,2013,69(5-8):1685-1699.
[10] LOUTAS T H,ROULIAS D,GEORGOULAS G. Remai-ning useful life estimation in rolling bearings utilizing data-driven probabilistic e-support vectors regression[J]. IEEE Transactions on Reliability,2013,62(4):821-832.
[11] MOSALLAM A,MEDJAHER K,ZERHOUNI N. Data-driven prognostic method based on Bayesian approaches for direct remaining useful life prediction[J]. Journal of Intelligent Manufacturing,2014,27(5):1-12.
[12] MEDJAHER K,ZERHOUNI N,JIHÈNE B. Data-driven prognostics based on health indicator construction:Application to PRONOSTIA's data[C]//European Control Conference,ECC'13,Jan. 2013,Switzerland,2013:1451-1456.
[13] CHOI S W,LEE I B. Nonlinear dynamic process monitoring based on dynamic kernel PCA[J]. Chemical Engineering Science,2004,59(24):5897-5908.
[14] 冯雄峰,阳宪惠,徐用懋. 多元统计过程控制方法平方预测误差分析[J]. 清华大学学报,1999,39(7):41-45. FENG Xiongfeng,YANG Xianhui,XU Yongmao. Squared prediction error analysis of multivariate statistical process control[J]. Tsinghua Science and Technology,1999,39(7):41-45.
[15] BENKEDJOUH T,MEDJAHER K,ZERHOUNI N,et al. Remaining useful life estimation based on nonlinear feature reduction and support vector regression[J]. Engin-eering Applications of Artificial Intelligence,2013,26(7):1751-1760.
[16] CERRADA M,SÁNCHEZ R V,LI C,et al. A review on data-driven fault severity assessment in rolling bearings[J]. Mechanical Systems and Signal Processing,2018,99:169-196.
[17] ZHANG Nanhua,GAO Xuejin,LI Yafen,et al. Fault detection of chiller based on improved KPCA[C/CD]//2016 Chinese Control and Decision Conference (CCDC),28-30 May,2016,Yinchuan,China. DOI:10. 1109/CCDC. 2016. 7531487.
[18] SAXENA M,BANNETT O O,SHARMA V. Bearing fault evaluation for structural health monitoring,fault detection,failure prevention and prognosis[J]. Procedia Engineering,2016,144:208-214.
[19] CHEN X,SHEN Z,HE Z,et al. Remaining life prognostics of rolling bearing based on relative features and multivariable support vector machine[J]. Proceedings of the Institution of Mechanical Engineers,Part C:Journal of Mechanical Engineering Science,2013,227(12):2849-2860.
[20] MA L,KANG J S,ZHAO C Y. Research on condition monitoring of bearing health using vibration data[J]. Applied Mechanics and Materials,2012,226-228:340-344.
[21] PALIWAL D,CHOUDHURY A,TINGARIKAR G. Wavelet and scalar indicator based fault assessment app-roach for rolling element bearings[J]. Procedia Materials Science,2014,5:2347-2355.
[22] WU Z. Rolling bearing fault evolution based on vibration time-domain parameters[J]. Key Engineering Materials,2016,693:1412-1418.
[23] HONG S,ZHOU Z,ZIO E,et al. Condition assessment for the performance degradation of bearing based on a combinatorial feature extraction method[J]. Digital Signal Processing,2014,27:159-166.
[24] GU H,ZHAO J,ZHANG X. Hybrid methodology of degradation feature extraction for bearing prognostics[J]. Maintain. Reliability,2013,15(2):195-201.
[25] HONG S,ZHOU Z,ZIO E,et al. An adaptive method for health trend prediction of rotating bearings[J]. Digital Signal Processing. 2014,35:117-123.
[26] HE W,MIAO Q,AZARIAN M,et al. Health monitoring of cooling fan bearings based on wavelet filter[J]. Mech-anical Systems and Signal Processing,2015,64-65:149-161.
[27] ZHANG Y,ZUO H,BAI F. Classification of fault location and performance degradation of a roller bearing[J]. Measurement,2013,46(3):1178-1189.
[28] LI D,LIU S,ZHANG H. Negative selection algorithm with constant detectors for anomaly detection[J]. Applied Soft Computing,2015,36:618-632.
[29] CHEN F F,YANG Y P,TANG B P,et al. Performance degradation prediction of mechanical equipment based on optimized multi-kernel relevant vector machine and fuzzy information granulation[J]. Measurement,2020,151:1-12.
[30] 杨蕊,李宏坤,贺长波,等. 利用最优小波尺度循环谱的滚动轴承早期故障特征提取[J]. 机械工程学报,2018,54(17):208-217. YANG Rui,LI Hongkun,HE Changbo,et al. Rolling element bearing incipient fault feature extraction based on optimal wavelet scales cyclic spectrum[J],Journal of Mechanical Engineering,2018,54(17):208-217.
[31] 孙鲜明,刘欢,赵新光,等. 基于瞬时包络尺度谱熵的滚动轴承早期故障奇异点识别及特征提取[J]. 机械工程学报,2017,53(3):73-80. SUN Xianming,LIU Huan,ZHAO Xinguang,et al. Singular point recognition and feature extraction for incipient bearing fault based on instantaneous envelope scalogram entropy[J],Journal of Mechanical Engineering,2017,53(3):73-80.
[32] 唐贵基,王晓龙. 参数优化变分模态分解方法在滚动轴承早期故障诊断中的应用[J]. 西安交通大学学报,2015,49(5):73-81. TANG Guiji,WANG Xiaolong. Parameter optimized variational mode decomposition method with application to incipient fault diagnosis of rolling bearing[J]. Journal of Xi'an Jiaotong University,2015,49(5):73-81.
[33] 张云强,张培林,王怀光,等. 基于双时域微弱故障特征增强的轴承早期故障智能识别[J]. 机械工程学报,2016,52(21):96-103. ZHANG Yunqiang,ZHANG Peilin,WANG Huaiguang,et al. Rolling bearing early fault intelligence recognition based on weak fault feature enhancement in time-time domain[J]. Journal of Mechanical Engineering,2016,52(21):96-103.
[34] 严保康,周凤星. 基于相干累积量分段正交匹配追踪方法的轴承早期故障稀疏特征提取[J]. 机械工程学报,2014,50(13):88-96. YAN Baokang,ZHOU Fengxing. Initial fault identification of bearing based on coherent cumulant stagewise orthogonal matching pursuit[J],Journal of Mechanical Engineering,2015,34(9):149-153.
[35] 李宏坤,刘洪轶,徐福健,等. 连续小波最优重构尺度确定方法与故障早期识别[J]. 机械工程学报,2014,50(17):69-76. LI Hongkun,LIU Hongyi,XU Fujian,et al. Method for the optimal continuous wavelet reconstruction scale determination and early fault classification[J]. Journal of Mechanical Engineering,2014,50(17):69-76.
[36] 冯辅周,司爱威,饶国强,等. 基于小波相关排列熵的轴承早期故障诊断技术[J]. 机械工程学报,2012,48(13):73-79. FENG Fuzhou,SI Aiwei,RAO Guoqiang,et al. Early fault diagnosis technology for bearing based on wavelet correlation permutation entropy[J]. Journal of Mechanical Engineering,2012,48(13):73-79.
[37] 孟玲霞,徐小力,徐杨梅,等. 变工况时频脊流形早期故障预警方法研究[J]. 北京理工大学学报,2017,37(9):942-947. MENG Lingxia,XU Xiaoli,XU Yangmei,et al. Time-frequency ridge manifold incipient fault warning on variable conditions[J]. Transactions of Beijing Institute of Technology,2017,37(9):942-947.
[38] 李鲲鹏,李衍达,张学工. 基于小波包分解的地层吸收补偿[J]. 地球物理学报,2000,43(4):542-549. LI Kunpeng,LI Yanda,ZHANG Xuegong. Stratum absorption compensation based on wavelet packet decom-position[J]. Chinese Journal of Geophysics,2000,43(4):542-549.
[39] 闫晓玲,董世运,徐滨士. 基于最优小波包Shannon熵的再制造电机转子缺陷诊断技术[J]. 机械工程学报,2016,52(4):7-12. YAN Xiaoling,DONG Shiyun,XU Binshi. Flaw diagnosis technology for remanufactured motor rotor based on optimal wavelet packet Shannon entropy[J]. Journal of Mechanical Engineering,2016,52(4):7-12.
[40] CRAIGMILE P F,PERCIVAL D B. Asymptotic decorrelation of between-scale wavelet coefficients[J]. IEEE Transac Tions on Information Theory,2005,51(3):1039-1048.
[41] BIANCHI D,MAYRHOFER E,GROSCHL M,et al. Wavelet packet transform for detection of single events in acoustic emission signals[J]. Mechanical Systems and Signal Processing,2015,64-65:441-451.
[42] 刘韬,陈进,董广明. KPCA和耦合隐马尔科夫模型在轴承故障诊断中的应用[J]. 振动与冲击,2014,33(21):85-89. LIU Tao,CHEN Jin,DONG Guangming. Application of KPCA and coupled hidden Markov model in bearing fault diagnosis[J]. Journal of Vibration and Shock,2014,33(21):85-89.
[43] 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(6):3-7. WANG Haiqing SONG Zhihuan,LI Ping. Improved PCA with application to process monitoring and fault diagnosis[J]. CIESC Journal,2001(6):3-7.
[44] 张明,冯坤,江志农. 基于动态自学习阈值和趋势滤波的机械故障智能预警方法[J]. 振动与冲击,2014,33(24):8-14. ZHANG Ming,FENG Kun,JIANG Zhinong. A mechanical fault early warning methodology based on dynamic self-learning threshold and trend filtering techniques[J]. Journal of Vibration and Shock,2014,33(24):8-14.
[45] LEE J,QIU H,YU G,et al. Rexnord technical services,bearing data set IMS[D]. Cincinnati:University of Cincinnati,2007.
[46] WANG D,TSUI K L,MIAO Q. Prognostics and Health management:A review of vibration based bearing and gear health indicators[J]. IEEE Access,2018,6(99):665-676.
[47] WANG B,LEI Y,LI N,et al. A hybrid prognostics approach for estimating remaining useful life of rolling element bearings[J]. IEEE Transactions on Reliability,2018:1-12.

一种数据驱动的旋转机械早期故障检测模型构建和应用研究

Research on Construction and Application of Data-driven Incipient Fault Detection Model for Rotating Machinery

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 6

编辑推荐

Metrics

本文评价

[1]	王庆锋, 李中, 许述剑, 陈文武. 基于故障案例学习的设备健康评价方法研究[J]. 机械工程学报, 2020, 56(20): 28-37.
[2]	张智, 刘成颖, 刘辛军, 张洁. 采用小波包能量熵的铣削振动状态分析方法研究[J]. 机械工程学报, 2018, 54(21): 57-62.
[3]	陈保家;陈雪峰;李兵;曹宏瑞;蔡改改;何正嘉. Logistic回归模型在机床刀具可靠性评估中的应用[J]. , 2011, 47(18): 158-164.
[4]	敖银辉;汪宝生. 钻头磨损检测与剩余寿命评估[J]. , 2011, 47(1): 177-181.
[5]	高英杰;孔祥东;ZHANG Qin. 基于小波包分析的液压泵状态监测方法[J]. , 2009, 45(8): 80-88.
[6]	褚福磊;王庆禹;卢文秀. 用声发射技术与小波包分解确定转子系统的碰摩位置[J]. , 2002, 38(3): 139-143.