结合C-均值聚类的自适应共振神经网络在风电机组齿轮箱故障诊断中的应用
|
摘要
提出了一种结合C-均值聚类的自适应共振(Adaptive Resonance Theory 2,ART2)神经网络无监督学习分类方法,用于风电机组齿轮箱设备群的故障诊断.利用某风电场齿轮箱运行数据,采用ART2神经网络对样本数据进行初步分类,再采用C-均值聚类算法对神经网络分类结果进行修正,得到最终诊断结果,并与ART2神经网络分类结果进行了比较.结果表明:所提出的方法解决了原始神经网络算法存在"硬竞争"导致分类精度下降的问题,准确度高于传统的ART2神经网络,可以准确识别出故障齿轮箱.
An unsupervised learning classification method by using adaptive resonance theory 2(ART2)neural network combined with C-means clustering was proposed for fault diagnosis of wind turbine gearbox group.For operation data of gearboxes in a wind farm,ART2 neural network was used to preliminarily classify the samples,and subsequently C-means clustering algorithm was used to modify the classification results of ART2 neural network,and the final classification results were compared with those of ART2 neural network.Results show that the proposed method has higher accuracy than traditional ART2 neural network,since it solves the problem of low classification accuracy caused by hard competition of ART2,which therefore can identify the faulted gearbox accurately.
An unsupervised learning classification method by using adaptive resonance theory 2(ART2)neural network combined with C-means clustering was proposed for fault diagnosis of wind turbine gearbox group.For operation data of gearboxes in a wind farm,ART2 neural network was used to preliminarily classify the samples,and subsequently C-means clustering algorithm was used to modify the classification results of ART2 neural network,and the final classification results were compared with those of ART2 neural network.Results show that the proposed method has higher accuracy than traditional ART2 neural network,since it solves the problem of low classification accuracy caused by hard competition of ART2,which therefore can identify the faulted gearbox accurately.
引文
[1]The Association of German Engineers.VDI 3834—2014Part 1:measurement and evaluation of the mechanical vibration of wind turbines and their components—wind turbines with gearbox[S].Germany:VDI Society Product and Process Design,2014:2-3.
[2]国家能源局.NB/T 31004—2011风力发电机组振动状态监测导则[S].北京:中国电力出版社,2011:8-9.
[3]周雁冰,柳亦兵,赵秋丽,等.基于非高斯性强度的风电齿轮箱故障特征提取[J].动力工程学报,2013,33(11):865-870.ZHOU Yanbing,LIU Yibing,ZHAO Qiuli,et al.Fault feature extraction from wind turbine gearbox based on non-Gaussian intensity[J].Journal of Chinese Society of Power Engineering,2013,33(11):865-870.
[4]JIANG Yonghua,TANG Baoping,QIN Yi,et al.Feature extraction method of wind turbine based on adaptive Morlet wavelet and SVD[J].Renewable Energy,2011,36(8):2146-2153.
[5]BAFROUIH H,OHADI A.Application of wavelet energy and Shannon entropy for feature extraction in gearbox fault detection under varying speed conditions[J].Neurocomputing,2014,133:437-445.
[6]龙泉,刘永前,杨勇平.基于粒子群优化BP神经网络的风电机组齿轮箱故障诊断方法[J].太阳能学报,2012,33(1):120-125.LONG Quan,LIU Yongqian,YANG Yongping.Fault diagnosis method of wind turbine gearbox based on BP neural network trained by particle swarm optimization algorithm[J].Acta Energiae Solaris Sinica,2012,33(1):120-125.
[7]张栾英,欧丽君,王亚顺.基于BP神经网络的解耦方法在直流锅炉控制系统中的应用[J].动力工程,2009,29(9):823-826.ZHANG Luanying,OU Lijun,WANG Yashun.Application of a decoupling method based on BP neural network in once-through boiler control system[J].Journal of Power Engineering,2009,29(9):823-826.
[8]TANG Baoping,SONG Tao,LI Feng,et al.Fault diagnosis for a wind turbine transmission system based on manifold learning and Shannon wavelet support vector machine[J].Renewable Energy,2014,62:1-9.
[9]YANG Zhiling,WANG Bin,DONG Xinghui,et al.Expert system of fault diagnosis for gear box in wind turbine[J].Systems Engineering Procedia,2012,4:189-195.
[10]马纪明,万蔚,王法岩.基于模型诊断技术的神经网络实现方法[J].北京航空航天大学学报,2013,39(2):178-183.MA Jiming,WAN Wei,WANG Fayan.Realization of model-based fault diagnosis with artificial neural network[J].Journal of Beijing University of Aeronautics and Astronautics,2013,39(2):178-183.
[11]彭治国,耿涛,刘海浪.自适应谐振理论在基于行为的故障诊断中的应用研究[J].制造技术与机床,2013(3):140-144.PENG Zhiguo,GENG Tao,LIU Hailang.Research of ART2in behavior-based faults diagnosis[J].Manufacturing Technology&Machine Tool,2013(3):140-144.
[12]LEE I S,LEE S J,KIM Y W.Fault diagnosis based on discrete wavelet transform and ART2neural network[C]//Proceedings of the SICE Annual Conference.Taiwan:IEEE,2010:3365-3370.
[13]周丽芳,李伟生,黄颖.模式识别原理及工程应用[M].北京:机械工业出版社,2013:122-125.
[14]SHAO Renping,HU Wentao,WANG Yayun,et al.The fault feature extraction and classification of gear using principal component analysis and kernel principal component analysis based on the wavelet packet transform[J].Measurement,2014,54:118-132.
[15]齐敏芳,付忠广,景源,等.基于信息熵与主成分分析的火电机组综合评价方法[J].中国电机工程学报,2013,33(2):58-64.QI Minfang,FU Zhongguang,JING Yuan,et al.A comprehensive evaluation method of power plant units based on information entropy and principal component analysis[J].Proceedings of the CSEE,2013,33(2):58-64.
[16]刘积芬.网络入侵检测关键技术研究[D].上海:东华大学,2013.
[2]国家能源局.NB/T 31004—2011风力发电机组振动状态监测导则[S].北京:中国电力出版社,2011:8-9.
[3]周雁冰,柳亦兵,赵秋丽,等.基于非高斯性强度的风电齿轮箱故障特征提取[J].动力工程学报,2013,33(11):865-870.ZHOU Yanbing,LIU Yibing,ZHAO Qiuli,et al.Fault feature extraction from wind turbine gearbox based on non-Gaussian intensity[J].Journal of Chinese Society of Power Engineering,2013,33(11):865-870.
[4]JIANG Yonghua,TANG Baoping,QIN Yi,et al.Feature extraction method of wind turbine based on adaptive Morlet wavelet and SVD[J].Renewable Energy,2011,36(8):2146-2153.
[5]BAFROUIH H,OHADI A.Application of wavelet energy and Shannon entropy for feature extraction in gearbox fault detection under varying speed conditions[J].Neurocomputing,2014,133:437-445.
[6]龙泉,刘永前,杨勇平.基于粒子群优化BP神经网络的风电机组齿轮箱故障诊断方法[J].太阳能学报,2012,33(1):120-125.LONG Quan,LIU Yongqian,YANG Yongping.Fault diagnosis method of wind turbine gearbox based on BP neural network trained by particle swarm optimization algorithm[J].Acta Energiae Solaris Sinica,2012,33(1):120-125.
[7]张栾英,欧丽君,王亚顺.基于BP神经网络的解耦方法在直流锅炉控制系统中的应用[J].动力工程,2009,29(9):823-826.ZHANG Luanying,OU Lijun,WANG Yashun.Application of a decoupling method based on BP neural network in once-through boiler control system[J].Journal of Power Engineering,2009,29(9):823-826.
[8]TANG Baoping,SONG Tao,LI Feng,et al.Fault diagnosis for a wind turbine transmission system based on manifold learning and Shannon wavelet support vector machine[J].Renewable Energy,2014,62:1-9.
[9]YANG Zhiling,WANG Bin,DONG Xinghui,et al.Expert system of fault diagnosis for gear box in wind turbine[J].Systems Engineering Procedia,2012,4:189-195.
[10]马纪明,万蔚,王法岩.基于模型诊断技术的神经网络实现方法[J].北京航空航天大学学报,2013,39(2):178-183.MA Jiming,WAN Wei,WANG Fayan.Realization of model-based fault diagnosis with artificial neural network[J].Journal of Beijing University of Aeronautics and Astronautics,2013,39(2):178-183.
[11]彭治国,耿涛,刘海浪.自适应谐振理论在基于行为的故障诊断中的应用研究[J].制造技术与机床,2013(3):140-144.PENG Zhiguo,GENG Tao,LIU Hailang.Research of ART2in behavior-based faults diagnosis[J].Manufacturing Technology&Machine Tool,2013(3):140-144.
[12]LEE I S,LEE S J,KIM Y W.Fault diagnosis based on discrete wavelet transform and ART2neural network[C]//Proceedings of the SICE Annual Conference.Taiwan:IEEE,2010:3365-3370.
[13]周丽芳,李伟生,黄颖.模式识别原理及工程应用[M].北京:机械工业出版社,2013:122-125.
[14]SHAO Renping,HU Wentao,WANG Yayun,et al.The fault feature extraction and classification of gear using principal component analysis and kernel principal component analysis based on the wavelet packet transform[J].Measurement,2014,54:118-132.
[15]齐敏芳,付忠广,景源,等.基于信息熵与主成分分析的火电机组综合评价方法[J].中国电机工程学报,2013,33(2):58-64.QI Minfang,FU Zhongguang,JING Yuan,et al.A comprehensive evaluation method of power plant units based on information entropy and principal component analysis[J].Proceedings of the CSEE,2013,33(2):58-64.
[16]刘积芬.网络入侵检测关键技术研究[D].上海:东华大学,2013.