基于EEMD-CC和PCA的风电齿轮箱状态监测方法
|
摘要
齿轮箱是风力发电机组的核心传动部件,不仅结构复杂制造成本高,而且故障率高维修费用巨大,对其进行状态监测具有重要意义。针对风电齿轮箱在复杂工况下运行所产生的非线性、非平稳振动信号,提出了一种基于EEMD-CC和PCA的风电齿轮箱状态监测方法。该方法先对含有大量噪声的风电齿轮箱振动信号进行集合经验模态分解和相关系数(EEMD-CC)降噪处理。然后,将降噪后的正常信号数据进行主分量分析(PCA)建模,并以T~2统计量和SPE统计量作为信号异常的评判指标。最后,把降噪后的测试数据带入PCA模型中,分别判断T~2和SPE值是否超出阈值,实现风电齿轮箱的状态监测。试验结果证明,该方法能够有效地监测风电齿轮箱的状态。
As the key part of wind turbine,gearbox consists of a large amount of complex parts and has a high incidence failure,and the costs of production and maintenance are especially expensive. Thus,it is meaningful to monitor the status of it. The vibration signals generated by the wind power gearbox under complex operating conditions are non-stationary and nonlinear. A wind turbine gearbox condition monitoring method based on Ensemble Empirical Mode Decomposition and Correlation Coefficien t(EEMD-CC)and Principal Component Analysi s(PCA)is proposed to analyze the vibration signals.Firstly,the vibration signal of wind power gearbox was denoised by EEMD-CC. Then,a model for normal vibration signal data was established by PCA and using T~2 and SPE statistics as evaluation indexes. Finally,the denoised test data was brought into the PCA model,and determined whether the T~2 and SPE values exceed the threshold. The experimental results show that the proposed method can effectively monitor the status of wind turbine gearbox.
As the key part of wind turbine,gearbox consists of a large amount of complex parts and has a high incidence failure,and the costs of production and maintenance are especially expensive. Thus,it is meaningful to monitor the status of it. The vibration signals generated by the wind power gearbox under complex operating conditions are non-stationary and nonlinear. A wind turbine gearbox condition monitoring method based on Ensemble Empirical Mode Decomposition and Correlation Coefficien t(EEMD-CC)and Principal Component Analysi s(PCA)is proposed to analyze the vibration signals.Firstly,the vibration signal of wind power gearbox was denoised by EEMD-CC. Then,a model for normal vibration signal data was established by PCA and using T~2 and SPE statistics as evaluation indexes. Finally,the denoised test data was brought into the PCA model,and determined whether the T~2 and SPE values exceed the threshold. The experimental results show that the proposed method can effectively monitor the status of wind turbine gearbox.
引文
[1]Soua S,Van Lieshout P,Perera A.Determination of the combined vibrational and acoustic emission signature of a wind turbine gearbox and generator shaft in service as a pre-requisite for effective condition monitoring[J].Renewable Energy,2013(51):175-81.
[2]生伟凯,刘卫,杨怀宇.国内外风电齿轮箱设计技术及主流技术路线综述与展望[J].风能,2012(4):40-44.(Sheng Wei-kai,Liu Wei,Yang Huai-yu.Review and prospect of design technology and main technical route of wind power gearbox at home and abroad[J].Wind Energy,2012(51):175-181.)
[3]郭鹏,David Infield,杨锡运.风电机组齿轮箱温度趋势状态监测及分析方法[J].中国电机工程学报,2011(32):129-136.(Guo Peng,David Infield,Yang Xi-yun.Wind turbine gearbox condition monitoring using temperature trend analysis[J].Proceedings of the CSEE,2011(32):129-136.)
[4]刘长良,闫萧.基于工况辨识的风电机组齿轮箱状态监测[J].热能动力工程,2016(7):41-46+133.(Liu Chang-liang,Yan Xiao.Condition monitoring of wind turbine gearbox based on the working regime identification[J].Journal of Engineering for Thermal Energy and Power,2016(7):41-46+133.)
[5]霍威.风电齿轮箱在线油液磨粒检测系统研究[D].北京:北京交通大学,2014.(Huo Wei.Research on online monitoring system of wear debris in wind turbine gearbox[D].Beijing:Beijing Jiaotong University,2014.)
[6]梁培沛,孙玉彬,陈铁.风电机组齿轮箱油液监测典型案例分析[J].润滑与密封,2016(4):141-145.(Liang Pei-pei,Sun Yu-bin,Chen Tie.Typical cases of oil-monitoring for gearbox of wind turbine[J].Lubrication Engineering,2016(4):141-145.)
[7]李鹏,孔凡让,何清波.齿轮箱状态监测中的振动信号多标度分析[J].计算机工程,2011(14):242-244+247.(Li Peng,Kong Fan-rang,He Qing-bo.Multiscale analysis on vibration signal in gearbox condition monitoring[J].Computer Engineering,2011(14):242-244+247.)
[8]古莹奎,潘高平,朱繁泷.基于邻域属性重要度与主成分分析的齿轮箱故障特征约简[J].中国机械工程,2016(13):1783-1789.(Gu Ying-kui,Pan Gao-ping,Zhu Fan-long.Gearbox fault feature reduction based on neighborhood at tribute importance and PCA[J].China Mechanical Engineering,2016(13):1783-1789.)
[9]杨杰,陈捷,洪荣晶.PCA和自相关包络分析在转盘轴承故障诊断中的应用[J].轴承,2014(10):54-58.(Yang Jie,Chen Jie,Hong Rong-jing.Application of PCA and autocorrelation envelope analysis in fault diagnosis for slewing bearings[J].Bearing,2014(10):54-58.)
[10]Wang F,Sun J,Yan D.A Feature extraction method for fault classification of Rolling Bearing based on PCA[C].//Journal of Physics:Conference Series.IOP Publishing,2015,628(1):012079.
[11]周媛,左洪福.基于主成分分析和线性判别的航空发动机状态监视[J].中国机械工程,2014(11):1433-1437.(Zhou Yuan,Zuo Hong-fu.Condition monitoring of aero-engine based on PCA and LDA[J].China Mechanical Engineering,2014(11):1433-1437.)
[12]Yue H H,Qin S J.Reconstruction-based fault identification using a combined index[J].Industrial&engineering chemistry research,2001,40(20):4403-4414.
[13]Wu Z,Huang N E.Ensemble empirical mode decomposition:a noiseassisted data analysis method[J].Advances in Adaptive Data Analysis,2009,1(1):1-41.
[14]陈仁祥,汤宝平,吕中亮.基于相关系数的EEMD转子振动信号降噪方法[J].振动.测试与诊断,2012(4):542-546+685.(Chen Ren-xiang,Tang Bao-ping,Lv Zhong-liang.Denoising method of rotor vibration signal based on correlation coefficient and EEMD[J].Journal of Vibration,Measurement&Diagnosis,2012(4):542-546+685.)
[2]生伟凯,刘卫,杨怀宇.国内外风电齿轮箱设计技术及主流技术路线综述与展望[J].风能,2012(4):40-44.(Sheng Wei-kai,Liu Wei,Yang Huai-yu.Review and prospect of design technology and main technical route of wind power gearbox at home and abroad[J].Wind Energy,2012(51):175-181.)
[3]郭鹏,David Infield,杨锡运.风电机组齿轮箱温度趋势状态监测及分析方法[J].中国电机工程学报,2011(32):129-136.(Guo Peng,David Infield,Yang Xi-yun.Wind turbine gearbox condition monitoring using temperature trend analysis[J].Proceedings of the CSEE,2011(32):129-136.)
[4]刘长良,闫萧.基于工况辨识的风电机组齿轮箱状态监测[J].热能动力工程,2016(7):41-46+133.(Liu Chang-liang,Yan Xiao.Condition monitoring of wind turbine gearbox based on the working regime identification[J].Journal of Engineering for Thermal Energy and Power,2016(7):41-46+133.)
[5]霍威.风电齿轮箱在线油液磨粒检测系统研究[D].北京:北京交通大学,2014.(Huo Wei.Research on online monitoring system of wear debris in wind turbine gearbox[D].Beijing:Beijing Jiaotong University,2014.)
[6]梁培沛,孙玉彬,陈铁.风电机组齿轮箱油液监测典型案例分析[J].润滑与密封,2016(4):141-145.(Liang Pei-pei,Sun Yu-bin,Chen Tie.Typical cases of oil-monitoring for gearbox of wind turbine[J].Lubrication Engineering,2016(4):141-145.)
[7]李鹏,孔凡让,何清波.齿轮箱状态监测中的振动信号多标度分析[J].计算机工程,2011(14):242-244+247.(Li Peng,Kong Fan-rang,He Qing-bo.Multiscale analysis on vibration signal in gearbox condition monitoring[J].Computer Engineering,2011(14):242-244+247.)
[8]古莹奎,潘高平,朱繁泷.基于邻域属性重要度与主成分分析的齿轮箱故障特征约简[J].中国机械工程,2016(13):1783-1789.(Gu Ying-kui,Pan Gao-ping,Zhu Fan-long.Gearbox fault feature reduction based on neighborhood at tribute importance and PCA[J].China Mechanical Engineering,2016(13):1783-1789.)
[9]杨杰,陈捷,洪荣晶.PCA和自相关包络分析在转盘轴承故障诊断中的应用[J].轴承,2014(10):54-58.(Yang Jie,Chen Jie,Hong Rong-jing.Application of PCA and autocorrelation envelope analysis in fault diagnosis for slewing bearings[J].Bearing,2014(10):54-58.)
[10]Wang F,Sun J,Yan D.A Feature extraction method for fault classification of Rolling Bearing based on PCA[C].//Journal of Physics:Conference Series.IOP Publishing,2015,628(1):012079.
[11]周媛,左洪福.基于主成分分析和线性判别的航空发动机状态监视[J].中国机械工程,2014(11):1433-1437.(Zhou Yuan,Zuo Hong-fu.Condition monitoring of aero-engine based on PCA and LDA[J].China Mechanical Engineering,2014(11):1433-1437.)
[12]Yue H H,Qin S J.Reconstruction-based fault identification using a combined index[J].Industrial&engineering chemistry research,2001,40(20):4403-4414.
[13]Wu Z,Huang N E.Ensemble empirical mode decomposition:a noiseassisted data analysis method[J].Advances in Adaptive Data Analysis,2009,1(1):1-41.
[14]陈仁祥,汤宝平,吕中亮.基于相关系数的EEMD转子振动信号降噪方法[J].振动.测试与诊断,2012(4):542-546+685.(Chen Ren-xiang,Tang Bao-ping,Lv Zhong-liang.Denoising method of rotor vibration signal based on correlation coefficient and EEMD[J].Journal of Vibration,Measurement&Diagnosis,2012(4):542-546+685.)