基于Hilbert-Huang变换的滚动轴承故障诊断方法研究
|
摘要
滚动轴承是大部分旋转机械的基本组成部件,也是易损零件,许多旋转机械中的故障都与滚动轴承有关。因此,研究滚动轴承的故障诊断方法具有十分重要的意义。本文以提取滚动轴承振动信号的特征及其故障诊断为主线,在分析了传统滚动轴承的振动信号处理方法的基础上,以Hilbert-Huang变换的时频分析方法为重点,对滚动轴承故障特征提取和故障诊断方法进行探索和研究,并对滚动轴承典型故障的振动信号进行分析和验证。最后,在此基础上开发出滚动轴承振动信号分析软件。
论文首先研究了滚动轴承的常见失效形式及其特征,并对滚动轴承的振动机理和振动信号进行了深入研究。针对几种常用的滚动轴承故障特征提取的时频分析方法：短时傅里叶变换、Wigner-Ville分布、小波变换等进行了研究。
对Hilbert-Huang算法中涉及的采样数据延拓技术开展了研究,以减弱端点效应；利用中值滤波技术来克服模态裂解问题；通过仿真比较,发现Rilling的IMF筛选终止准则具有较好的数据处理特性。
论文利用多种时频分析方法对仿真信号进行特征提取,再把基于Hilbert-Huang的时频分析方法与典型的时频分析方法进行对比研究,指出该方法对于分析非平稳信号具有更好的有效性和优越性。
论文利用基于Hilbert-Huang的时频分析方法对实测的滚动轴承故障振动信号进行了分析。实验研究表明,Hilbert-Huang方法对滚动轴承故障诊断具有很好的有效性和可行性。
最后,本文开发出交互式用户界面,便于分析程序的实际应用。
论文的研究内容为滚动轴承故障特征提取及分析提供了新的思路。
The rolling bearing is basic component part of the most rotating machinery, and it is damaged easily. On account of this, many faults of rotating machinery have great relationship with it. Therefore, it is of importance to study fault diagnosis methods. In this paper, main attention has been put on extracting the feature of vibration signal and investigation on fault diagnosis for rolling bearing. Based on the analysis of the traditional processing methods of vibration signal, focusing on the time-frequency analysis method based on Hilbert-Huang transform, the feature extraction and fault diagnosis methods for rolling bearing have been investigated. And analysis and validation for the vibration signal of typical fault for roller bearing have been done. Finally the vibration signal analysis software for rolling bearing was developed.
First of all, the common failure mode and characteristics of rolling bearing were analyzed in this paper, the vibration mechanism and characteristics of rolling bearing were deeply studied. Several commonly used time-frequency analysis methods have been researched too, such as Short-time Fourier transform, Wigner-Ville distribution and Wavelet transform.
The extending technology of sampled data related in Hilbert-Huang algorithm were studied to diminish the end effects; The median filtering was used to overcome the problem of mode decomposition; According simulation and comparison, it was found that the Rilling termination criteria for the sifting of the IMF has better properties in data-processing.
Features of simulated signal were extracted through a variety of time-frequency analysis methods, compared with the typical time-frequency analysis methods, we find Hilbert-Huang transform has more validity and superiority in fields of analyzing non-stationary signal.
The actual fault vibration signals of rolling bearing were analiyed with the time-frequency method based on Hilbert-Huang. It has been showed that Hilbert-Huang method was more effective and feasible.
Finally, an interactive user interface was developed in this paper, which is used for practical application of analyzing programs.
This paper provides a new way for fault feature extraction and analysis for rolling bearing.
论文首先研究了滚动轴承的常见失效形式及其特征,并对滚动轴承的振动机理和振动信号进行了深入研究。针对几种常用的滚动轴承故障特征提取的时频分析方法：短时傅里叶变换、Wigner-Ville分布、小波变换等进行了研究。
对Hilbert-Huang算法中涉及的采样数据延拓技术开展了研究,以减弱端点效应；利用中值滤波技术来克服模态裂解问题；通过仿真比较,发现Rilling的IMF筛选终止准则具有较好的数据处理特性。
论文利用多种时频分析方法对仿真信号进行特征提取,再把基于Hilbert-Huang的时频分析方法与典型的时频分析方法进行对比研究,指出该方法对于分析非平稳信号具有更好的有效性和优越性。
论文利用基于Hilbert-Huang的时频分析方法对实测的滚动轴承故障振动信号进行了分析。实验研究表明,Hilbert-Huang方法对滚动轴承故障诊断具有很好的有效性和可行性。
最后,本文开发出交互式用户界面,便于分析程序的实际应用。
论文的研究内容为滚动轴承故障特征提取及分析提供了新的思路。
The rolling bearing is basic component part of the most rotating machinery, and it is damaged easily. On account of this, many faults of rotating machinery have great relationship with it. Therefore, it is of importance to study fault diagnosis methods. In this paper, main attention has been put on extracting the feature of vibration signal and investigation on fault diagnosis for rolling bearing. Based on the analysis of the traditional processing methods of vibration signal, focusing on the time-frequency analysis method based on Hilbert-Huang transform, the feature extraction and fault diagnosis methods for rolling bearing have been investigated. And analysis and validation for the vibration signal of typical fault for roller bearing have been done. Finally the vibration signal analysis software for rolling bearing was developed.
First of all, the common failure mode and characteristics of rolling bearing were analyzed in this paper, the vibration mechanism and characteristics of rolling bearing were deeply studied. Several commonly used time-frequency analysis methods have been researched too, such as Short-time Fourier transform, Wigner-Ville distribution and Wavelet transform.
The extending technology of sampled data related in Hilbert-Huang algorithm were studied to diminish the end effects; The median filtering was used to overcome the problem of mode decomposition; According simulation and comparison, it was found that the Rilling termination criteria for the sifting of the IMF has better properties in data-processing.
Features of simulated signal were extracted through a variety of time-frequency analysis methods, compared with the typical time-frequency analysis methods, we find Hilbert-Huang transform has more validity and superiority in fields of analyzing non-stationary signal.
The actual fault vibration signals of rolling bearing were analiyed with the time-frequency method based on Hilbert-Huang. It has been showed that Hilbert-Huang method was more effective and feasible.
Finally, an interactive user interface was developed in this paper, which is used for practical application of analyzing programs.
This paper provides a new way for fault feature extraction and analysis for rolling bearing.
引文
[1]韩捷.旋转机械故障机理及诊断技术[M].北京：机械工业出版社,2002：35-36
[2]O.G.Gustantsson, T.Tallian.Detection of Damage in Assembled Rolling Bearing. Transactions of the ASLE,1962,5:197-209
[3]陈进.机械设备振动监测与故障诊断[M].上海：上海交通大学出版社,1999：48-51
[4]王肇琪,付勤毅.滚动轴承故障的振动监测方法[J].有色矿山,1999,8(1)：6-40
[5](美)L.科恩著,自居宪译.时-频分析：理论与应用[M].西安：西安交通大学出版社,2001：50-52
[6]殷晓中,于盛林.信号的时频分析理论与应用评述[J].现代电子技术,2006,29(21)：118-120
[7]N.E. Huang, Z. Shen, S.R. Long,etc.The empirical mode decomposition and Hilbert spectrum for nonlinear and non-stationary time series analysis[J]. Proc. Roy. Soc. London. A,1998,454:903-995
[8]N.E.Huang. A new method for nonlinear and non-stationary time series analysis:Empirical mode decomposition and Hilbert spectral analysis. Proc. of SPIE,2000 (4056):197-209
[9]N.E.Huang. Review of Empirical mode decomposition. Proc.of SPIE,2001(4391):71-79
[10]钟佑明,秦树人,汤宝平.Hilbert-Huang变换中的理论研究[J].振动与冲击,2002,21(4)：13-17
[11]Gravier B.M, N.N.J, Pelstring J. A. An assessment of the application of the Hilbert spectrum to the fatigue analysis of marine risers. Proceedings of the International Offshore and Polar Engineering Conference,2001.2:268-275
[12]钟佑明,秦树人,汤宝平.一种振动信号新变换法的研究[J].振动工程学报,2002,15(2)：233-38
[13]于德介,程军圣.EMD方法在齿轮故障诊断中的应用[J].湖南大学学报(自然科学版),2002,29(6)：47-51
[14]胡劲松.面向旋转机械故障诊断的经验模态分解时频分析方法及实验研究[D].浙江：浙江大学,2003：35-38
[15]王宏宇.非平稳随机信号分析与处理[M].北京：国防工业出版社,2001,45-50
[16]L.Cohen.Time-frequency analysis. NJ:Prentice Hall Englewood Cliffs,1995:15-50
[17]N.E.Huang.A New View of Non-linear Waves:The Hilbert Spectrum. Annual Review of Fluid Mechanics,1999,5(31):417-457
[18]S.O.Rice. Mathematical analysis of random noise. Bell syst.Tech,1994(23):21-32
[19]谭善文.基于多分辨分析的实验模态分解方法[D].重庆：重庆大学,2002：44-51
[20]于德介,程军圣,杨宇.机械故障诊断的Hilbert-Huang变换方法[M].北京：科学出版社,2006：58-63
[21]钟佑明,秦树人,汤宝平.希尔伯特—黄变换中边际谱的研究[J].系统工程与电子技术,2004,26(9)：1324-1325
[22]熊学军,郭炳火等.EMD方法和Hilbert谱分析法的应用与探讨[J].黄渤海海洋,2002,6：12-14
[23]李琳.HHT时频分析方法的研究与应用[D].吉林：吉林大学,2006：35-40
[24]Huang.N.E, C.C.Chern, etal.A new spectral representation of earthquake data:Hilbert spectral analysis of station TUC129.Bulletin of the Seismological Society of America.1999,91(5): 1310-1338.
[25]陈忠,郑时雄.EMD信号分析方法边缘效应的分析[J].数据采集与处理,2003,18(1)：94-97
[26]赵进平.异常事件对EMD方法的影响及其解决方法研究[J].青岛海洋大学学报,2001,31(6)：805-814
[27]邓拥军,王伟,钱成春等.EMD方法及Hilbert变换中边界问题的处理[J].科学通报,2001,46(3)：257-263
[28]陈娟.Hilbert-Huang变换及其在信号处理中的应用[D].大连：大连理工大学,2006：51-52
[29]张晏,程家兴.利用Hilbert变换提取信号瞬时特征参数的问题研究[J].电讯技术,2003,14(4)：45-47
[30]Zhao Jin-ping.Mirror extending and circular spline function for empirical mode decomposition method [J]. Journal of Zejiang University,2000,54(3):247-252
[31]黄大吉,赵进平,苏纪兰.希尔伯特一黄变换的端点延拓[J].海洋学报,2003,25(1)：1-11
[32]胡劲松,杨世锡,吴昭同等.滤波技术对旋转机械振动信号EMD的影响[J].风机技术,2002,4：57-59
[33]刘丽梅,孙玉荣,李莉.中值滤波技术发展研究[J].云南师范大学学报,2004,24(1)：23-27
[34]程军圣,于德介,杨宇.经典模态分解方法中内禀模态函数判据问题研究[J].中国机械：工程,2004,15(20)：1861-1864
[35]Quek S T, Tua P S, Wang Q. Detecting anomalies in beams and plate based on the Hilbert-Huang transform of real signals [J]. Smart Materials andStructures,2003,12(3):447-460
[36]Gabriel Rilling, Patrick Flandrin and Paulo Goncalves. On Empirical Mode Decomposition And Its Algorithms.IEEE-EURASIP Workshop on Nonlinear Signal and Image Processing, NSIP-03,2003 (in press)
[37]胡昌华,周涛,夏启兵等.基于MATLAB的系统分析与设计一时频分析[M].西安：西安电子科技大学出版社,2002
[38]胡劲松,杨世锡,吴昭同等.旋转机械振动信号的HHT和STFT时频谱图比较研究[J].汽轮机技术,2002,44(6)：336-338
[39]葛哲学,陈仲生.Matlab时频分析技术及其应用[M].北京：人民邮电出版社,2006
[40]邹红星,戴琼海.不含交叉项干扰且具有WVD聚集性的时频分布之不存在性[J].中国科学,2002,31(4)：349-354
[41]刘贵忠,邸双亮.小波分析及其应用[M].西安：西安电子科技大学出版社,2001
[42]杨世锡,胡劲松,吴昭同等.旋转机械振动信号基于EMD的希尔伯特变换和小波变换时频分析比较[J].中国电机工：程学报,2003,23(6)：102-107
[43]范德功.基于希尔伯特_黄变换机械故障诊断的研究[D].北京：华北电力大学,2006：54-56
[44]何正嘉等.机械故障诊断案例选编[M].西安：西安交通大学出版社,1991：62-65
[45]刘泽九.滚动轴承应用手册[M].北京：北京机械工业出版社,1996：20-50
[46]虞和济,韩庆大等.设备故障诊断工程[M].北京：冶金工业出版社,2001：32-35
[47]丁玉兰,石来德.机械设备故障诊断技术[M].上海：上海科学技术文献出版社,1994：52-58
[48]曾海平.基于经验模态分解法的滚动轴承故障诊断系统研究[D].浙江：浙江大学,2005：26-30
[49]滚动轴承数据中心(Case Westem Reserve University Bearing Data Center) http://www.eecs.case.edu/laboratory/bearing/download.htm
[50]张智星.MATLAB程序设计与应用[M].北京：清华大学出版社,2002
[51]苏金明,刘宏,刘波MATLAB高级编程[M].北京：电子工业出版社,2005.8
[2]O.G.Gustantsson, T.Tallian.Detection of Damage in Assembled Rolling Bearing. Transactions of the ASLE,1962,5:197-209
[3]陈进.机械设备振动监测与故障诊断[M].上海：上海交通大学出版社,1999：48-51
[4]王肇琪,付勤毅.滚动轴承故障的振动监测方法[J].有色矿山,1999,8(1)：6-40
[5](美)L.科恩著,自居宪译.时-频分析：理论与应用[M].西安：西安交通大学出版社,2001：50-52
[6]殷晓中,于盛林.信号的时频分析理论与应用评述[J].现代电子技术,2006,29(21)：118-120
[7]N.E. Huang, Z. Shen, S.R. Long,etc.The empirical mode decomposition and Hilbert spectrum for nonlinear and non-stationary time series analysis[J]. Proc. Roy. Soc. London. A,1998,454:903-995
[8]N.E.Huang. A new method for nonlinear and non-stationary time series analysis:Empirical mode decomposition and Hilbert spectral analysis. Proc. of SPIE,2000 (4056):197-209
[9]N.E.Huang. Review of Empirical mode decomposition. Proc.of SPIE,2001(4391):71-79
[10]钟佑明,秦树人,汤宝平.Hilbert-Huang变换中的理论研究[J].振动与冲击,2002,21(4)：13-17
[11]Gravier B.M, N.N.J, Pelstring J. A. An assessment of the application of the Hilbert spectrum to the fatigue analysis of marine risers. Proceedings of the International Offshore and Polar Engineering Conference,2001.2:268-275
[12]钟佑明,秦树人,汤宝平.一种振动信号新变换法的研究[J].振动工程学报,2002,15(2)：233-38
[13]于德介,程军圣.EMD方法在齿轮故障诊断中的应用[J].湖南大学学报(自然科学版),2002,29(6)：47-51
[14]胡劲松.面向旋转机械故障诊断的经验模态分解时频分析方法及实验研究[D].浙江：浙江大学,2003：35-38
[15]王宏宇.非平稳随机信号分析与处理[M].北京：国防工业出版社,2001,45-50
[16]L.Cohen.Time-frequency analysis. NJ:Prentice Hall Englewood Cliffs,1995:15-50
[17]N.E.Huang.A New View of Non-linear Waves:The Hilbert Spectrum. Annual Review of Fluid Mechanics,1999,5(31):417-457
[18]S.O.Rice. Mathematical analysis of random noise. Bell syst.Tech,1994(23):21-32
[19]谭善文.基于多分辨分析的实验模态分解方法[D].重庆：重庆大学,2002：44-51
[20]于德介,程军圣,杨宇.机械故障诊断的Hilbert-Huang变换方法[M].北京：科学出版社,2006：58-63
[21]钟佑明,秦树人,汤宝平.希尔伯特—黄变换中边际谱的研究[J].系统工程与电子技术,2004,26(9)：1324-1325
[22]熊学军,郭炳火等.EMD方法和Hilbert谱分析法的应用与探讨[J].黄渤海海洋,2002,6：12-14
[23]李琳.HHT时频分析方法的研究与应用[D].吉林：吉林大学,2006：35-40
[24]Huang.N.E, C.C.Chern, etal.A new spectral representation of earthquake data:Hilbert spectral analysis of station TUC129.Bulletin of the Seismological Society of America.1999,91(5): 1310-1338.
[25]陈忠,郑时雄.EMD信号分析方法边缘效应的分析[J].数据采集与处理,2003,18(1)：94-97
[26]赵进平.异常事件对EMD方法的影响及其解决方法研究[J].青岛海洋大学学报,2001,31(6)：805-814
[27]邓拥军,王伟,钱成春等.EMD方法及Hilbert变换中边界问题的处理[J].科学通报,2001,46(3)：257-263
[28]陈娟.Hilbert-Huang变换及其在信号处理中的应用[D].大连：大连理工大学,2006：51-52
[29]张晏,程家兴.利用Hilbert变换提取信号瞬时特征参数的问题研究[J].电讯技术,2003,14(4)：45-47
[30]Zhao Jin-ping.Mirror extending and circular spline function for empirical mode decomposition method [J]. Journal of Zejiang University,2000,54(3):247-252
[31]黄大吉,赵进平,苏纪兰.希尔伯特一黄变换的端点延拓[J].海洋学报,2003,25(1)：1-11
[32]胡劲松,杨世锡,吴昭同等.滤波技术对旋转机械振动信号EMD的影响[J].风机技术,2002,4：57-59
[33]刘丽梅,孙玉荣,李莉.中值滤波技术发展研究[J].云南师范大学学报,2004,24(1)：23-27
[34]程军圣,于德介,杨宇.经典模态分解方法中内禀模态函数判据问题研究[J].中国机械：工程,2004,15(20)：1861-1864
[35]Quek S T, Tua P S, Wang Q. Detecting anomalies in beams and plate based on the Hilbert-Huang transform of real signals [J]. Smart Materials andStructures,2003,12(3):447-460
[36]Gabriel Rilling, Patrick Flandrin and Paulo Goncalves. On Empirical Mode Decomposition And Its Algorithms.IEEE-EURASIP Workshop on Nonlinear Signal and Image Processing, NSIP-03,2003 (in press)
[37]胡昌华,周涛,夏启兵等.基于MATLAB的系统分析与设计一时频分析[M].西安：西安电子科技大学出版社,2002
[38]胡劲松,杨世锡,吴昭同等.旋转机械振动信号的HHT和STFT时频谱图比较研究[J].汽轮机技术,2002,44(6)：336-338
[39]葛哲学,陈仲生.Matlab时频分析技术及其应用[M].北京：人民邮电出版社,2006
[40]邹红星,戴琼海.不含交叉项干扰且具有WVD聚集性的时频分布之不存在性[J].中国科学,2002,31(4)：349-354
[41]刘贵忠,邸双亮.小波分析及其应用[M].西安：西安电子科技大学出版社,2001
[42]杨世锡,胡劲松,吴昭同等.旋转机械振动信号基于EMD的希尔伯特变换和小波变换时频分析比较[J].中国电机工：程学报,2003,23(6)：102-107
[43]范德功.基于希尔伯特_黄变换机械故障诊断的研究[D].北京：华北电力大学,2006：54-56
[44]何正嘉等.机械故障诊断案例选编[M].西安：西安交通大学出版社,1991：62-65
[45]刘泽九.滚动轴承应用手册[M].北京：北京机械工业出版社,1996：20-50
[46]虞和济,韩庆大等.设备故障诊断工程[M].北京：冶金工业出版社,2001：32-35
[47]丁玉兰,石来德.机械设备故障诊断技术[M].上海：上海科学技术文献出版社,1994：52-58
[48]曾海平.基于经验模态分解法的滚动轴承故障诊断系统研究[D].浙江：浙江大学,2005：26-30
[49]滚动轴承数据中心(Case Westem Reserve University Bearing Data Center) http://www.eecs.case.edu/laboratory/bearing/download.htm
[50]张智星.MATLAB程序设计与应用[M].北京：清华大学出版社,2002
[51]苏金明,刘宏,刘波MATLAB高级编程[M].北京：电子工业出版社,2005.8