Repairing Deteriorated Data of Wind Turbines by Multi-head Attention Bi-directional Long Short Time Memory Networks under Complex Working Conditions

doi:10.3901/JME.2023.14.001

Abstract

Abstract: To address the problem of low early warning accuracy of condition monitoring models due to a large number of missing data in wind turbine monitoring parameters under variable speed and variable load, multi-headed attention bidirectional long and short-term memory networks (MA-BiLSTM) is proposed for repair those data. The multi-headed attention machine is used to suppress the interference of variable loads on the neural network feature extraction under complex working conditions. In addition, the model feature extraction ability is increased by constructing a cross-layer of residual units, and the hidden features of existing monitoring data and the correlation between multi-source parameters are fully learned. The Bi-LSTM cells are used to simultaneously learn the law of the monitoring data of wind turbines to achieve the prediction and repair of incomplete data. The application results show that the proposed MA-BiLSTM networks can suppress the multivariate load disturbance under complex working conditions and realize the repair of incomplete data for improving fault detection accuracy of wind turbines.

Key words: wind turbine, deteriorated data repair, multi-headed attention mechanism, bi-directional long and short term memory network

CLC Number:

TH17

YU Xiaoxia, TANG Baoping, WANG Weiying, WU Xuanyong, LI Biao. Repairing Deteriorated Data of Wind Turbines by Multi-head Attention Bi-directional Long Short Time Memory Networks under Complex Working Conditions[J]. Journal of Mechanical Engineering, 2023, 59(14): 1-9.

References

[1] 张帆,刘德顺,戴巨川,等. 一种基于SCADA参数关系的风电机组运行状态识别方法[J]. 机械工程学报,2019,55(4):1-9. ZHANG Fan,LIU Deshun,DAI Juchuan,et al. An operating condition recognition method of wind turbine based on SCADA parameter relations[J]. Journal of Mechanical Engineering,2019,55(4):1-9.
[2] WANG Jinjiang,LIANG Yuanyuan,ZHENG Yinghao,et al. An integrated fault diagnosis and prognosis approach for predictive maintenance of wind turbine bearing with limited samples[J]. Renewable Energy,2020,145:642-650.
[3] TENG Wei,DING Xian,CHENG Hao,et al. Compound faults diagnosis and analysis for a wind turbine gearbox via a novel vibration model and empirical wavelet transform[J]. Renewable Energy,2019,136:393-402.
[4] 雷亚国,许学方,蔡潇,等. 面向机械装备健康监测的数据质量保障方法研究[J]. 机械工程学报,2021,57(4):1-9. LEI Yaguo,XU Xuefang,CAI Xiao,et al. Research on data quality assurance for health condition monitoring of machinery[J]. Journal of Mechanical Engineering,2021,57(4):1-9.
[5] VELASCO-GALLEGO C,LAZAKIS I. Real-time data-driven missing data imputation for short-term sensor data of marine systems. A comparative study[J]. Ocean Engineering,2020,218:108261.
[6] KAISER J. Dealing with missing values in data[J]. Journal of Systems Integration,2014,5(1):42-51.
[7] LI D,DEOGUN J,SPAULDING W,et al. Towards missing data imputation:a study of fuzzy k-means clustering method[C]//International Conference on Rough Sets and Current Trends in Computing. Springer,Berlin,Heidelberg,2004:573-579.
[8] WANG Guanjin,LU Jie,CHOI K,et al. A transfer-based additive LS-SVM classifier for handling missing data[J]. IEEE Transactions on Cybernetics,2018,50(2):739-752.
[9] HOU K,XIA D,LI Q,et al. A two-stage ensemble of diverse models for recognition of abnormal data in raw wind data[C]//2016 IEEE PES Asia-Pacific Power and Energy Engineering Conference (APPEEC). IEEE,2016:880-884.
[10] ZHANG Y,THORBURN P. A dual-head attention model for time series data imputation[J]. Computers and Electronics in Agriculture,2021,189:106377.
[11] 张晟斐,李天梅,胡昌华,等. 基于深度卷积生成对抗网络的缺失数据生成方法及其在剩余寿命预测中的应用[J].航空学报,2021,42(6):625708. ZHANG Shengfei,LI Tianmei,HU Changhua,et al. Deep convolutional generative adversarial network based missing data generation method and its application in remaining useful life prediction[J]. Acta Aeronautica et Astronautica Sinica,2021,42(6):625708.
[12] AWAN S E,BENNAMOUN M,SOHEL F,et al. Imputation of missing data with class imbalance using conditional generative adversarial networks[J]. Neurocomputing,2021,453:164-171.
[13] QU Fuming,LIU Jinhai,MA Yanjuan,et al. A novel wind turbine data imputation method with multiple optimizations based on GANs[J]. Mechanical Systems and Signal Processing,2020,139:106610.
[14] LI Dong,LI Linhao,LI Xianling,et al. A spatio-temporal deep model for multiple time-series missing imputation[J]. Neurocomputing,2020,411:351-363.
[15] FURFARI F A. The transformer[J]. IEEE Industry Applications Magazine,2002,8:8-15.
[16] TAY Y,DEHGHANI M,BAHRI D,et al. Efficient transformers:A survey[J]. ACM Computing Surveys,2020,55(6):1-28.
[17] WANG Zhiwei,MA Yao,LIU Zitao,et al. R-transformer:Recurrent neural network enhanced transformer[J]. arXiv preprint arXiv,2019:1-11.
[18] AZEVEDO H D M,ARAÚJO A M,BOUCHONNEAU N. A review of wind turbine bearing condition monitoring:State of the art and challenges[J]. Renewable and Sustainable Energy Reviews,2016,56:368-379.
[19] RENSTRÖM N,BANGALORE P,HIGHCOCK E. System-wide anomaly detection in wind turbines using deep autoencoders[J]. Renewable Energy,2020,157:647-659.
[20] WANG S,ZHANG Z,WANG P,et al. Failure warning of gearbox for wind turbine based on 3σ-median criterion and NSET[J]. Energy Reports,2021,7:1182-1197.