轧辊磨床状态监测与故障预测技术进展
|
摘要
轧辊磨床是冶金行业的关键装备,针对其运行状态缺乏有效监测等问题,特提出了应用于轧辊磨床的状态监测与故障预测技术。简要介绍了轧辊磨床状态监测技术的方式与方法,综述了国内外学者在信号特征提取及轧辊磨床状态监测系统开发上所做的工作。重点对故障预测的方法进行了分类和介绍,总结归纳了21种故障预测的工具方法并分别做了评价。最后提出当前轧辊磨床状态监测与故障预测所面临的问题与挑战。
The roller grinder is the key equipment in metallurgical industry,which lacks effective monitoring for its running state. Thus,the state monitoring and fault prediction technology applied to the roller grinder is put forward. This paper briefly introduces the ways and methods of the status monitoring technology for roller grinder,and summarizes the work done by both domestic and abroad scholars on the characteristics extraction signal and the development of the status monitoring system for the roller grinder. The methods of fault prediction are classified and introduced,and 21 kinds of fault prediction methods are summarized and evaluated respectively. Finally,the problems and challenges of current roll mill condition monitoring and fault prediction are put forward.
The roller grinder is the key equipment in metallurgical industry,which lacks effective monitoring for its running state. Thus,the state monitoring and fault prediction technology applied to the roller grinder is put forward. This paper briefly introduces the ways and methods of the status monitoring technology for roller grinder,and summarizes the work done by both domestic and abroad scholars on the characteristics extraction signal and the development of the status monitoring system for the roller grinder. The methods of fault prediction are classified and introduced,and 21 kinds of fault prediction methods are summarized and evaluated respectively. Finally,the problems and challenges of current roll mill condition monitoring and fault prediction are put forward.
引文
[1]LEE J,WU F,ZHAO W,et al. Prognostics and healthmanagement design for rotary machinery systems-re-views,methodology and applications[J]. MechanicalSystems and Signal Processing,2014,42(1):314.
[2]初光勇.数控轧辊磨床振动信号检测系统研究[D].贵阳:贵州大学,2016.
[3]徐小力,梁福平,许宝杰,等.旋转机械状态监测及预测技术的发展与研究[J]. 2003,16(7):20.
[4]刘海宁.基于稀疏编码的设备状态识别及其重型轧辊磨床监测应用[D].上海:上海交通大学,2011.
[5]RIAZ S,ELAHI H,JAVAID K,et al. Vibration featureextraction and analysis for fault diagnosis of rotating ma-chinery-a literature survey[J]. Asia Pacific Journal ofMultidisciplinary Research,2017,51(5):103.
[6]TOUTOUNTZAKIS T,MBA D. Observations of acousticemission activity during gear defect diagnosis[J]. NDT&E International,2003,36(7):471.
[7]CHEN J,YANG Y,HUANG J. On-line monitoring anddiagnosis of membrane fouling using ultrasonic tech-niques[J]. Chemometrics and Intelligent Laboratory Sys-tems,2013(127):147.
[8]ZHOU W,FEI M,ZHOU H,et al. A sparse representa-tion based fast detection method for surface defect detec-tion of bottle caps[J]. Neurocomputing,2014(123):406.
[9]谭兴强,陈新德.精密轧辊磨床故障振动分析及消除[J].机床与液压,2014(10):64.
[10]李凤轶,周国荣.精密轧辊磨床磨削缺陷诊断方法的研究[J].铝加工,2001,24(3):42.
[11]张海南,胡学雄.振动分析技术在轧辊磨床故障判断中的应用[J].中国设备工程,2011(8):51.
[12]李楠.基于单类学习的异常检测方法及其重型装备状态识别应用[D].上海:上海交通大学,2014.
[13]毛彬,周国荣,蒋复岱.基于小波变换的轧辊磨床振动信号分析研究[J].噪声与振动控制,2007,27(1):48.
[14]温仲元.设备监测和诊断系统的传感器选择[J].中国设备工程,2012(8):5.
[15]胡长海.轧机主传动齿轮箱振动传感器的选型分析[J].机械工程与自动化,2008(5):56.
[16]焦慧锋.基于贝叶斯网络的数控平面磨床磨削状态智能监测研究[D].杭州:浙江大学,2011.
[17]李子枫,杜文辽,李彦明,等.基于牛顿迭代盲源分离方法的磨床故障诊断[J].振动与冲击,2013,32(24):44.
[18]KWAK J,HA M. Neural network approach for diagnosisof grinding operation by acoustic emission and powersignals[J]. Journal of Materials Processing Technology,2004,147(1):65.
[19]LIN J,QU L. Feature extraction based on Morlet waveletand its application for mechanical fault diagnosis[J].Journal of Sound and Vibration,2000,234(1):135.
[20]WIDODO A,YANG B S. Application of nonlinear fea-ture extraction and support vector machines for fault di-agnosis of induction motors[J]. Expert Systems withApplications,2007,33(1):241.
[21]LOU X,LOPARO K A. Bearing fault diagnosis based onwavelet transform and fuzzy inference[J]. MechanicalSystems&Signal Processing,2004,18(5):1077.
[22]LEI Y,HE Z,ZI Y. A new approach to intelligent faultdiagnosis of rotating machinery[J]. Expert Systems withApplications,2008,35(4):1593.
[23]LI B,ZHANG P L,LIU D S,et al. Feature extraction forrolling element bearing fault diagnosis utilizing general-ized S transform and two-dimensional non-negative ma-trix factorization[J]. Journal of Sound&Vibration,2011,330(10):2388.
[24]WU J D,LIU C H. An expert system for fault diagnosisin internal combustion engines using wavelet packettransform and neural network[J]. Expert Systems withApplications,2009,36(3):4278.
[25]吕会娟.数控磨床远程监控与诊断现场信息处理关键技术研究[D].上海:上海交通大学,2010.
[26]张颖颖.数控磨床控制系统故障自诊断及抗干扰设计[J].安阳工学院学报,2012,11(4):5.
[27]赵荣泳,张浩,樊留群,等. ARM嵌入式系统在数控磨床故障诊断中的应用研究[J].机电一体化,2004,10(3):87.
[28]温熙森.模式识别与状态监控[M].北京:机械工业出版社,2007.
[29]徐小力.设备运行状态监测与故障诊断的现代技术[C]//全国设备监测与诊断学术会议.三亚:中国机械工程学会,2005:7.
[30]邱立军,吴明辉. PHM技术框架及其关键技术综述[J].国外电子测量技术,2018(2):10.
[31]左宪章,康健,李浩,等.故障预测技术综述[J].火力与指挥控制,2010(1):1.
[32]彭宇,刘大同,彭喜元.故障预测与健康管理技术综述[J].电子测量与仪器学报,2010,24(1):1.
[33]马洁,徐小力,周东华.旋转机械的故障预测方法综述[J].自动化仪表,2011,32(8):1.
[34]王子玲,许爱强,杨智勇.装备故障诊断和预测技术综述[J].火力与指挥控制,2008(S2):8.
[35]JARDINE A K S,LIN D,BANJEVIC D. A review onmachinery diagnostics and prognostics implementingcondition-based maintenance[J]. Mechanical Systemsand Signal Processing,2006,20(7):1483.
[36]LIU Z,CHEN Y,LEE J. Review and recent advances inbattery health monitoring and prognostics technologiesfor electric vehicle(EV)safety and mobility[J]. Jour-nal of Power Sources,2014,256(12):110.
[37]孙晓丽,张来斌,段礼祥.机械设备故障预警方法综述[C]//全国设备监测与诊断学术会议.北京:中国机械工程学会,2012:681.
[38]HANSEN R,HALL D,KURTZ S. New approach to thechallenge of machinery prognostics[J]. Journal of Engi-neering for Gas Turbines&Power,1995,117(2):320.
[39]薛子云,杨江天,朱衡君.机械故障预测模型综述[J].机械强度,2006(S1):60.
[40]王金福,李富才.机械故障诊断技术中的信号处理方法:时域分析[J].噪声与振动控制,2013(2):128.
[41]上海航空测控技术研究所.航空故障诊断与健康管理技术[M].北京:航空工业出版社,2013.
[42]LIU Z,YIN X,ZHANG Z,et al. Online rotor mixed faultdiagnosis way based on spectrum analysis of instantane-ous power in squirrel cage induction motors[J]. Pro-ceedings of the CSEE,2003,19(3):485.
[43]边海龙,陈光福.基于短时傅里叶变换检测非平稳信号的频域内插优化抗混叠算法[J].仪器仪表学报,2008,29(2):284.
[44]叶昊,王桂增.小波变换在故障检测中的应用[J].自动化学报,1997,23(6):736.
[45]RENAUD O,STARCK J L,MURTAGH F. Wavelet-based combined signal filtering and prediction[J].IEEE Transactions on Systems Man&Cybernetics PartB,2005,35(6):1241.
[46]HUANG N E,SHEN S S P. Hilbert-Huang transformand its applications[M]. Singapore:World Scientific,2005.
[47]李风民.希尔伯特变换及其应用研究[J].中国无线电,2006(7):53.
[48]JOLLIFFE I T. Principal component analysis[M]. Ber-lin:Springer-Verlag,2005.
[49]ABDI H,WILLIAMS L J. Principal component analysis[J]. Wiley Interdisciplinary Reviews ComputationalStatistics,2010,2(4):433.
[50]杨梁军. Fisher线性判别对电缆故障的判别[J].价值工程,2016(22):136.
[51]乔少杰,金琨,韩楠,等.一种基于高斯混合模型的轨迹预测算法[J].软件学报,2015,26(5):1048.
[52]孙旭升,周刚,于洋,等.机械设备故障预测与健康管理综述[J].兵工自动化,2016(1):30.
[53]YAN J,LEE J. Degradation assessment and fault modesclassification using logistic regression[J]. Journal ofManufacturing Science&Engineering,2005,127(4):912.
[54]STEYERBERG E,EIJKEMANS M F J,HABBEMA J.Prognostic modelling with logistic regression analysis:acomparison of selection and estimation methods in smalldata sets[J]. Statistics in Medicine,2000,19(8):1059.
[55]SOHN H,FARRAR C R,HUNTER N F,et al. Structur-al health monitoring using statistical pattern recognitiontechniques[J]. Journal of Dynamic Systems Measure-ment&Control,2001,123(4):706.
[56]JAIN A K,DUIN R P W,MAO J. Statistical pattern rec-ognition:a review[J]. IEEE Transactions on Pattern A-nalysis&Machine Intelligence,2002,27(11):1502.
[57]何志昆,刘光斌,赵曦晶,等.高斯过程回归方法综述[J].控制与决策,2013(8):1121.
[58]SEEGER M. Gaussian process for machine learning[M]. Boston:MIT Press,2006.
[59]杨小军,潘泉,王睿,等.粒子滤波进展与展望[J].控制理论与应用,2006,23(2):261.
[60]CARPENTER J,CLIFFORD P,FEARNHEAD P. Im-proved particle filter for nonlinear problems[J]. IEEProceedings-Radar,Sonar and Navigation,2002,146(1):2.
[61]JENSEN F V. An Introduction to Bayesian networks[M]. Berlin:Springer-Verlag,1996.
[62]黄影平.贝叶斯网络发展及其应用综述[J].北京理工大学学报,2013,33(12):1211.
[63]QUINLAN J R. Induction of decision trees[J]. MachineLearning,1986,1(1):81.
[64]董明.基于组合决策树的油浸式电力变压器故障诊断[J].中国电机工程学报,2005,25(16):35.
[65]KANTARDZIC M. Fuzzy sets and fuzzy logic[M]. Lon-don:Prentice Hall PTR,1995.
[66]EBERSBACH S,PENG Z. Expert system developmentfor vibration analysis in machine condition monitoring[M]. Oxford:Pergamon Press,Inc.,2008.
[67]WIDODO A,YANG B. Support vector machine in ma-chine condition monitoring and fault diagnosis[J]. Me-chanical Systems and Signal Processing,2007,21(6):2560.
[68]KOHLSCHEIN C. An introduction to hidden Markovmodels[J]. Current Protocols in Bioinformatics,2013(Appendix 3):3A.
[69]RABINER L R. A tutorial on hidden Markov model andselected applications in speech recognition[J]. Read-ings in Speech Recognition,1989,77(2):267.
[70]MAN H,BAE J,KANG S,et al. Real time and an in-process measuring system for the grinding process cylin-drical workpieces using Kalman filtering[J]. IEEETransactions on Industrial Electronics,2002,47(6):1326.
[2]初光勇.数控轧辊磨床振动信号检测系统研究[D].贵阳:贵州大学,2016.
[3]徐小力,梁福平,许宝杰,等.旋转机械状态监测及预测技术的发展与研究[J]. 2003,16(7):20.
[4]刘海宁.基于稀疏编码的设备状态识别及其重型轧辊磨床监测应用[D].上海:上海交通大学,2011.
[5]RIAZ S,ELAHI H,JAVAID K,et al. Vibration featureextraction and analysis for fault diagnosis of rotating ma-chinery-a literature survey[J]. Asia Pacific Journal ofMultidisciplinary Research,2017,51(5):103.
[6]TOUTOUNTZAKIS T,MBA D. Observations of acousticemission activity during gear defect diagnosis[J]. NDT&E International,2003,36(7):471.
[7]CHEN J,YANG Y,HUANG J. On-line monitoring anddiagnosis of membrane fouling using ultrasonic tech-niques[J]. Chemometrics and Intelligent Laboratory Sys-tems,2013(127):147.
[8]ZHOU W,FEI M,ZHOU H,et al. A sparse representa-tion based fast detection method for surface defect detec-tion of bottle caps[J]. Neurocomputing,2014(123):406.
[9]谭兴强,陈新德.精密轧辊磨床故障振动分析及消除[J].机床与液压,2014(10):64.
[10]李凤轶,周国荣.精密轧辊磨床磨削缺陷诊断方法的研究[J].铝加工,2001,24(3):42.
[11]张海南,胡学雄.振动分析技术在轧辊磨床故障判断中的应用[J].中国设备工程,2011(8):51.
[12]李楠.基于单类学习的异常检测方法及其重型装备状态识别应用[D].上海:上海交通大学,2014.
[13]毛彬,周国荣,蒋复岱.基于小波变换的轧辊磨床振动信号分析研究[J].噪声与振动控制,2007,27(1):48.
[14]温仲元.设备监测和诊断系统的传感器选择[J].中国设备工程,2012(8):5.
[15]胡长海.轧机主传动齿轮箱振动传感器的选型分析[J].机械工程与自动化,2008(5):56.
[16]焦慧锋.基于贝叶斯网络的数控平面磨床磨削状态智能监测研究[D].杭州:浙江大学,2011.
[17]李子枫,杜文辽,李彦明,等.基于牛顿迭代盲源分离方法的磨床故障诊断[J].振动与冲击,2013,32(24):44.
[18]KWAK J,HA M. Neural network approach for diagnosisof grinding operation by acoustic emission and powersignals[J]. Journal of Materials Processing Technology,2004,147(1):65.
[19]LIN J,QU L. Feature extraction based on Morlet waveletand its application for mechanical fault diagnosis[J].Journal of Sound and Vibration,2000,234(1):135.
[20]WIDODO A,YANG B S. Application of nonlinear fea-ture extraction and support vector machines for fault di-agnosis of induction motors[J]. Expert Systems withApplications,2007,33(1):241.
[21]LOU X,LOPARO K A. Bearing fault diagnosis based onwavelet transform and fuzzy inference[J]. MechanicalSystems&Signal Processing,2004,18(5):1077.
[22]LEI Y,HE Z,ZI Y. A new approach to intelligent faultdiagnosis of rotating machinery[J]. Expert Systems withApplications,2008,35(4):1593.
[23]LI B,ZHANG P L,LIU D S,et al. Feature extraction forrolling element bearing fault diagnosis utilizing general-ized S transform and two-dimensional non-negative ma-trix factorization[J]. Journal of Sound&Vibration,2011,330(10):2388.
[24]WU J D,LIU C H. An expert system for fault diagnosisin internal combustion engines using wavelet packettransform and neural network[J]. Expert Systems withApplications,2009,36(3):4278.
[25]吕会娟.数控磨床远程监控与诊断现场信息处理关键技术研究[D].上海:上海交通大学,2010.
[26]张颖颖.数控磨床控制系统故障自诊断及抗干扰设计[J].安阳工学院学报,2012,11(4):5.
[27]赵荣泳,张浩,樊留群,等. ARM嵌入式系统在数控磨床故障诊断中的应用研究[J].机电一体化,2004,10(3):87.
[28]温熙森.模式识别与状态监控[M].北京:机械工业出版社,2007.
[29]徐小力.设备运行状态监测与故障诊断的现代技术[C]//全国设备监测与诊断学术会议.三亚:中国机械工程学会,2005:7.
[30]邱立军,吴明辉. PHM技术框架及其关键技术综述[J].国外电子测量技术,2018(2):10.
[31]左宪章,康健,李浩,等.故障预测技术综述[J].火力与指挥控制,2010(1):1.
[32]彭宇,刘大同,彭喜元.故障预测与健康管理技术综述[J].电子测量与仪器学报,2010,24(1):1.
[33]马洁,徐小力,周东华.旋转机械的故障预测方法综述[J].自动化仪表,2011,32(8):1.
[34]王子玲,许爱强,杨智勇.装备故障诊断和预测技术综述[J].火力与指挥控制,2008(S2):8.
[35]JARDINE A K S,LIN D,BANJEVIC D. A review onmachinery diagnostics and prognostics implementingcondition-based maintenance[J]. Mechanical Systemsand Signal Processing,2006,20(7):1483.
[36]LIU Z,CHEN Y,LEE J. Review and recent advances inbattery health monitoring and prognostics technologiesfor electric vehicle(EV)safety and mobility[J]. Jour-nal of Power Sources,2014,256(12):110.
[37]孙晓丽,张来斌,段礼祥.机械设备故障预警方法综述[C]//全国设备监测与诊断学术会议.北京:中国机械工程学会,2012:681.
[38]HANSEN R,HALL D,KURTZ S. New approach to thechallenge of machinery prognostics[J]. Journal of Engi-neering for Gas Turbines&Power,1995,117(2):320.
[39]薛子云,杨江天,朱衡君.机械故障预测模型综述[J].机械强度,2006(S1):60.
[40]王金福,李富才.机械故障诊断技术中的信号处理方法:时域分析[J].噪声与振动控制,2013(2):128.
[41]上海航空测控技术研究所.航空故障诊断与健康管理技术[M].北京:航空工业出版社,2013.
[42]LIU Z,YIN X,ZHANG Z,et al. Online rotor mixed faultdiagnosis way based on spectrum analysis of instantane-ous power in squirrel cage induction motors[J]. Pro-ceedings of the CSEE,2003,19(3):485.
[43]边海龙,陈光福.基于短时傅里叶变换检测非平稳信号的频域内插优化抗混叠算法[J].仪器仪表学报,2008,29(2):284.
[44]叶昊,王桂增.小波变换在故障检测中的应用[J].自动化学报,1997,23(6):736.
[45]RENAUD O,STARCK J L,MURTAGH F. Wavelet-based combined signal filtering and prediction[J].IEEE Transactions on Systems Man&Cybernetics PartB,2005,35(6):1241.
[46]HUANG N E,SHEN S S P. Hilbert-Huang transformand its applications[M]. Singapore:World Scientific,2005.
[47]李风民.希尔伯特变换及其应用研究[J].中国无线电,2006(7):53.
[48]JOLLIFFE I T. Principal component analysis[M]. Ber-lin:Springer-Verlag,2005.
[49]ABDI H,WILLIAMS L J. Principal component analysis[J]. Wiley Interdisciplinary Reviews ComputationalStatistics,2010,2(4):433.
[50]杨梁军. Fisher线性判别对电缆故障的判别[J].价值工程,2016(22):136.
[51]乔少杰,金琨,韩楠,等.一种基于高斯混合模型的轨迹预测算法[J].软件学报,2015,26(5):1048.
[52]孙旭升,周刚,于洋,等.机械设备故障预测与健康管理综述[J].兵工自动化,2016(1):30.
[53]YAN J,LEE J. Degradation assessment and fault modesclassification using logistic regression[J]. Journal ofManufacturing Science&Engineering,2005,127(4):912.
[54]STEYERBERG E,EIJKEMANS M F J,HABBEMA J.Prognostic modelling with logistic regression analysis:acomparison of selection and estimation methods in smalldata sets[J]. Statistics in Medicine,2000,19(8):1059.
[55]SOHN H,FARRAR C R,HUNTER N F,et al. Structur-al health monitoring using statistical pattern recognitiontechniques[J]. Journal of Dynamic Systems Measure-ment&Control,2001,123(4):706.
[56]JAIN A K,DUIN R P W,MAO J. Statistical pattern rec-ognition:a review[J]. IEEE Transactions on Pattern A-nalysis&Machine Intelligence,2002,27(11):1502.
[57]何志昆,刘光斌,赵曦晶,等.高斯过程回归方法综述[J].控制与决策,2013(8):1121.
[58]SEEGER M. Gaussian process for machine learning[M]. Boston:MIT Press,2006.
[59]杨小军,潘泉,王睿,等.粒子滤波进展与展望[J].控制理论与应用,2006,23(2):261.
[60]CARPENTER J,CLIFFORD P,FEARNHEAD P. Im-proved particle filter for nonlinear problems[J]. IEEProceedings-Radar,Sonar and Navigation,2002,146(1):2.
[61]JENSEN F V. An Introduction to Bayesian networks[M]. Berlin:Springer-Verlag,1996.
[62]黄影平.贝叶斯网络发展及其应用综述[J].北京理工大学学报,2013,33(12):1211.
[63]QUINLAN J R. Induction of decision trees[J]. MachineLearning,1986,1(1):81.
[64]董明.基于组合决策树的油浸式电力变压器故障诊断[J].中国电机工程学报,2005,25(16):35.
[65]KANTARDZIC M. Fuzzy sets and fuzzy logic[M]. Lon-don:Prentice Hall PTR,1995.
[66]EBERSBACH S,PENG Z. Expert system developmentfor vibration analysis in machine condition monitoring[M]. Oxford:Pergamon Press,Inc.,2008.
[67]WIDODO A,YANG B. Support vector machine in ma-chine condition monitoring and fault diagnosis[J]. Me-chanical Systems and Signal Processing,2007,21(6):2560.
[68]KOHLSCHEIN C. An introduction to hidden Markovmodels[J]. Current Protocols in Bioinformatics,2013(Appendix 3):3A.
[69]RABINER L R. A tutorial on hidden Markov model andselected applications in speech recognition[J]. Read-ings in Speech Recognition,1989,77(2):267.
[70]MAN H,BAE J,KANG S,et al. Real time and an in-process measuring system for the grinding process cylin-drical workpieces using Kalman filtering[J]. IEEETransactions on Industrial Electronics,2002,47(6):1326.