应用深度自编码网络和XGBoost的风电机组发电机故障诊断
|
摘要
针对风电机组现场故障样本难获取的问题,为实现风电机组发电机部件的故障诊断,通过分析风机监控与采集(SCADA)数据,设计了基于深度自编码(DAE)网络和XGBoost的故障诊断算法。该算法包含两部分:第一部分是DAE故障检测算法,通过DAE获取SCADA数据的重构值,分析重构误差的变化趋势与其超越阈值的情况以预测风机故障和提取故障样本;第二部分是XGBoost故障识别算法,用贝叶斯优化搜索XGBoost的最优超参数,建立XGBoost多分类故障识别模型。算例结果表明,DAE算法能够捕获风电机组发电机早期故障,XGBoost比其他算法更精确地识别不同故障类型。
Aiming at the problem that wind turbine field fault samples are difficult to obtain and to realize the fault diagnosis of generator components of wind turbine generators,through the analysis of supervisory control and data acquisition(SCADA)data,a fault diagnosis algorithm based on deep autoencoder(DAE)network and XGBoost is designed.The algorithm consists of two parts.The first part is the DAE fault detection algorithm,which obtains the reconstructed values of the SCADA data through DAE and analyzes the trend of the reconstruction error and its situation beyond the threshold to predict a fault of wind turbine and to extract the fault samples.The second part is the XGBoost fault identification algorithm.By using Bayesian optimization to search the optimal hyper-parameters of XGBoost,an XGBoost multi-class fault identification model is established.The results of the example show that the DAE algorithm can capture the early fault of wind turbine generators,and XGBoost can identify different fault types more accurately than other algorithms.
Aiming at the problem that wind turbine field fault samples are difficult to obtain and to realize the fault diagnosis of generator components of wind turbine generators,through the analysis of supervisory control and data acquisition(SCADA)data,a fault diagnosis algorithm based on deep autoencoder(DAE)network and XGBoost is designed.The algorithm consists of two parts.The first part is the DAE fault detection algorithm,which obtains the reconstructed values of the SCADA data through DAE and analyzes the trend of the reconstruction error and its situation beyond the threshold to predict a fault of wind turbine and to extract the fault samples.The second part is the XGBoost fault identification algorithm.By using Bayesian optimization to search the optimal hyper-parameters of XGBoost,an XGBoost multi-class fault identification model is established.The results of the example show that the DAE algorithm can capture the early fault of wind turbine generators,and XGBoost can identify different fault types more accurately than other algorithms.
引文
[1]QIAO W,LU D.A survey on wind turbine condition monitoring and fault diagnosis:PartⅠcomponents and subsystems[J].IEEE Transactions on Industrial Electronics,2015,62(10):6536-6545.
[2]LU Bin,LI Yaoyu,WU Xin,et al.A review of recent advances in wind turbine condition monitoring and fault diagnosis[C]//Power Electronics and Machines in Wind Applications,June 24-26,2009,Lincoln,USA:1-7.
[3]PREZ J M P,MRQUEZ F P G,TOBIAS A,et al.Wind turbine reliability analysis[J].Renewable and Sustainable Energy Reviews,2013,23:463-472.
[4]ZHAO Yingying,LI Dongsheng,DONG Ao,et al.Fault prediction and diagnosis of wind turbine generators using SCADA data[J].Energies,2017,10(8):1210.
[5]SIMANI S,FARSONI S.Fault diagnosis and sustainable control of wind turbines:robust data-driven and model-based strategies[M].Oxford:Butterworth Heinemann,2018.
[6]MEHRANBOD N,SOROUSH M,PANJAPORNPON C.Amethod of sensor fault detection and identification[J].Journal of Process Control,2005,15(3):321-339.
[7]郭慧东,王玮,夏明超.基于数据挖掘的风电机组变桨系统劣化状态在线辨识方法[J].中国电机工程学报,2016,36(9):2389-2397.GUO Huidong,WANG Wei,XIA Mingchao.On-line identification for wind turbine pitch system degradation based on data mining technology[J].Proceedings of the CSEE,2016,36(9):2389-2397.
[8]郭鹏,INFIELD D,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,2011,31(32):129-136.GUO Peng,INFIELD D,YANG Xiyun.Wind turbine gearbox condition monitoring using temperature trend analysis[J].Proceedings of the CSEE,2011,31(32):129-136.
[9]梁颖,方瑞明.基于SCADA和支持向量回归的风电机组状态在线评估方法[J].电力系统自动化,2013,37(14):7-12.LIANG Ying,FANG Ruiming.An online wind turbine condition assessment method based on SCADA and support vector regression[J].Automation of Electric Power Systems,2013,37(14):7-12.
[10]BANGALORE P,TJERNBERG L B.An artificial neural network approach for early fault detection of gearbox bearings[J].IEEE Transactions on Smart Grid,2015,6(2):980-987.
[11]NITHYA M,NAGARAJAN S,NAVASEELAN P.Fault detection of wind turbine system using neural networks[C]//Technological Innovations in ICT for Agriculture and Rural Development(TIAR),April 7-8,2017,Chennai,India:103-108.
[12]曾军,陈艳峰,杨苹,等.大型风力发电机组故障诊断综述[J].电网技术,2018,42(3):849-860.ZHENG Jun,CHEN Yanfeng,YANG Ping,et al.Review of fault diagnosis methods of large-scale wind turbines[J].Power System Technology,2018,42(3):849-860.
[13]WANG L,ZHANG Z,LONG H,et al.Wind turbine gearbox failure identification with deep neural networks[J].IEEETransactions on Industrial Informatics,2017,13(3):1360-1368.
[14]JIANG G,XIE P,HE H,et al.Wind turbine fault detection using a denoising autoencoder with temporal information[J].IEEE-ASME Transactions on Mechatronics,2018,23(1):89-100.
[15]赵洪山,刘辉海,刘宏杨,等.基于堆叠自编码网络的风电机组发电机状态监测与故障诊断[J].电力系统自动化,2018,42(11):102-108.DOI:10.7500/AEPS20170730001.ZHAO Hongshan,LIU Huihai,LIU Hongyang,et al.Condition monitoring and fault diagnosis of wind turbine generator based on stacked autoencoder network[J].Automation of Electric Power Systems,2018,42(11):102-108.DOI:10.7500/AEPS20170730001.
[16]YANG Zhixin,WANG Xianbo,ZHONG Jianhua.Representational learning for fault diagnosis of wind turbine equipment:a multi-layered extreme learning machines approach[J].Energies,2016,9(6):379.
[17]张少敏,毛冬,王保义.大数据处理技术在风电机组齿轮箱故障诊断与预警中的应用[J].电力系统自动化,2016,40(14):129-134.DOI:10.7500/AEPS20160316013.ZHANG Shaomin,MAO Dong,WANG Baoyi.Application of big data processing technology in fault diagnosis and early warning of wind turbine gearbox[J].Automation of Electric Power Systems,2016,40(14):129-134.DOI:10.7500/AEPS20160316013.
[18]ZHANG Z,VERMA A,KUSIAK A.Fault analysis and condition monitoring of the wind turbine gearbox[J].IEEETransactions on Energy Conversion,2012,27(2):526-535.
[19]TIAN J,AZARIAN M H,PECHT M,et al.An ensemble learning-based fault diagnosis method for rotating machinery[C]//Prognostics and System Health Management Conference(PHM-Harbin),July 9-12,2017,Harbin,China:1-6.
[20]NIELSEN D.Tree boosting with XGBoost-why does XGBoost win“every”machine learning competition[D].Trondheim:Norwegian University of Science and Technology,2016.
[21]FISCHER A,IGEL C.Training restricted Boltzmann machines:an introduction[J].Pattern Recognition,2014,47(1):25-39.
[22]CHEN T,GUESTRIN C.XGBoost:a scalable tree boosting system[C]//Proceedings of the 22nd ACM SIGKDDInternational Conference on Knowledge Discovery and Data Mining,August 13-17,2016,San Francisco,USA:785-794.
[23]PEDREGOSA F,VAROQUAUX G,GRAMFORT A,et al.Scikit-learn:machine learning in Python[J].Journal of Machine Learning Research,2011,12(1):2825-2830.
[24]SNOEK J,LAROCHELLE H,ADAMS R P.Practical Bayesian optimization of machine learning algorithms[C]//Advances in Neural Information Processing Systems,December 2-6,2012,Lake Tahoe,United States:2951-2959.
[2]LU Bin,LI Yaoyu,WU Xin,et al.A review of recent advances in wind turbine condition monitoring and fault diagnosis[C]//Power Electronics and Machines in Wind Applications,June 24-26,2009,Lincoln,USA:1-7.
[3]PREZ J M P,MRQUEZ F P G,TOBIAS A,et al.Wind turbine reliability analysis[J].Renewable and Sustainable Energy Reviews,2013,23:463-472.
[4]ZHAO Yingying,LI Dongsheng,DONG Ao,et al.Fault prediction and diagnosis of wind turbine generators using SCADA data[J].Energies,2017,10(8):1210.
[5]SIMANI S,FARSONI S.Fault diagnosis and sustainable control of wind turbines:robust data-driven and model-based strategies[M].Oxford:Butterworth Heinemann,2018.
[6]MEHRANBOD N,SOROUSH M,PANJAPORNPON C.Amethod of sensor fault detection and identification[J].Journal of Process Control,2005,15(3):321-339.
[7]郭慧东,王玮,夏明超.基于数据挖掘的风电机组变桨系统劣化状态在线辨识方法[J].中国电机工程学报,2016,36(9):2389-2397.GUO Huidong,WANG Wei,XIA Mingchao.On-line identification for wind turbine pitch system degradation based on data mining technology[J].Proceedings of the CSEE,2016,36(9):2389-2397.
[8]郭鹏,INFIELD D,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,2011,31(32):129-136.GUO Peng,INFIELD D,YANG Xiyun.Wind turbine gearbox condition monitoring using temperature trend analysis[J].Proceedings of the CSEE,2011,31(32):129-136.
[9]梁颖,方瑞明.基于SCADA和支持向量回归的风电机组状态在线评估方法[J].电力系统自动化,2013,37(14):7-12.LIANG Ying,FANG Ruiming.An online wind turbine condition assessment method based on SCADA and support vector regression[J].Automation of Electric Power Systems,2013,37(14):7-12.
[10]BANGALORE P,TJERNBERG L B.An artificial neural network approach for early fault detection of gearbox bearings[J].IEEE Transactions on Smart Grid,2015,6(2):980-987.
[11]NITHYA M,NAGARAJAN S,NAVASEELAN P.Fault detection of wind turbine system using neural networks[C]//Technological Innovations in ICT for Agriculture and Rural Development(TIAR),April 7-8,2017,Chennai,India:103-108.
[12]曾军,陈艳峰,杨苹,等.大型风力发电机组故障诊断综述[J].电网技术,2018,42(3):849-860.ZHENG Jun,CHEN Yanfeng,YANG Ping,et al.Review of fault diagnosis methods of large-scale wind turbines[J].Power System Technology,2018,42(3):849-860.
[13]WANG L,ZHANG Z,LONG H,et al.Wind turbine gearbox failure identification with deep neural networks[J].IEEETransactions on Industrial Informatics,2017,13(3):1360-1368.
[14]JIANG G,XIE P,HE H,et al.Wind turbine fault detection using a denoising autoencoder with temporal information[J].IEEE-ASME Transactions on Mechatronics,2018,23(1):89-100.
[15]赵洪山,刘辉海,刘宏杨,等.基于堆叠自编码网络的风电机组发电机状态监测与故障诊断[J].电力系统自动化,2018,42(11):102-108.DOI:10.7500/AEPS20170730001.ZHAO Hongshan,LIU Huihai,LIU Hongyang,et al.Condition monitoring and fault diagnosis of wind turbine generator based on stacked autoencoder network[J].Automation of Electric Power Systems,2018,42(11):102-108.DOI:10.7500/AEPS20170730001.
[16]YANG Zhixin,WANG Xianbo,ZHONG Jianhua.Representational learning for fault diagnosis of wind turbine equipment:a multi-layered extreme learning machines approach[J].Energies,2016,9(6):379.
[17]张少敏,毛冬,王保义.大数据处理技术在风电机组齿轮箱故障诊断与预警中的应用[J].电力系统自动化,2016,40(14):129-134.DOI:10.7500/AEPS20160316013.ZHANG Shaomin,MAO Dong,WANG Baoyi.Application of big data processing technology in fault diagnosis and early warning of wind turbine gearbox[J].Automation of Electric Power Systems,2016,40(14):129-134.DOI:10.7500/AEPS20160316013.
[18]ZHANG Z,VERMA A,KUSIAK A.Fault analysis and condition monitoring of the wind turbine gearbox[J].IEEETransactions on Energy Conversion,2012,27(2):526-535.
[19]TIAN J,AZARIAN M H,PECHT M,et al.An ensemble learning-based fault diagnosis method for rotating machinery[C]//Prognostics and System Health Management Conference(PHM-Harbin),July 9-12,2017,Harbin,China:1-6.
[20]NIELSEN D.Tree boosting with XGBoost-why does XGBoost win“every”machine learning competition[D].Trondheim:Norwegian University of Science and Technology,2016.
[21]FISCHER A,IGEL C.Training restricted Boltzmann machines:an introduction[J].Pattern Recognition,2014,47(1):25-39.
[22]CHEN T,GUESTRIN C.XGBoost:a scalable tree boosting system[C]//Proceedings of the 22nd ACM SIGKDDInternational Conference on Knowledge Discovery and Data Mining,August 13-17,2016,San Francisco,USA:785-794.
[23]PEDREGOSA F,VAROQUAUX G,GRAMFORT A,et al.Scikit-learn:machine learning in Python[J].Journal of Machine Learning Research,2011,12(1):2825-2830.
[24]SNOEK J,LAROCHELLE H,ADAMS R P.Practical Bayesian optimization of machine learning algorithms[C]//Advances in Neural Information Processing Systems,December 2-6,2012,Lake Tahoe,United States:2951-2959.