Abstract
Wind turbine anomaly detection is an important but challenging work. Though deep learning is promising in this filed, it may fail for large error distributions and often lacks interpretability due to its complex structure. To this end, we develop a composite quantile regression long short-term memory network with group lasso (CQR-LSTM-GL). The CQR-LSTM-GL model extends the LSTM network to the quantile regression framework, which makes LSTM always valid regardless of the error distribution. Additionally, the CQR-LSTM-GL model adopts the group lasso method to compress the network structure and select key features. To illustrate its efficacy, we conduct a numerical experiment on the public server machine datasets. The experimental results show that the CQR-LSTM-GL model is not only slightly better than several competitive models in anomaly detection performance, but also far ahead in sparsity performance. The sparsity of CQR-LSTM-GL is 1.63 times that of CQR-LSTM-L1, and much more than that of CQR-LSTM. We then apply the CQR-LSTM-GL model to the anomaly detection of a wind turbine using the real Supervisory Control and Data Acquisition data, where all three anomalies are identified.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Data availibility statement
The datasets analysed during the current study are not publicly available due to confidentiality requirements of the provider but are available from the corresponding author on reasonable request.
References
Cambron P, Masson C, Tahan A, Pelletier F (2018) Control chart monitoring of wind turbine generators using the statistical inertia of a wind farm average. Renew Energy 116:88–98
Cannon AJ (2011) Quantile regression neural networks: implementation in R and application to precipitation downscaling. Comput Geosci 37(9):1277–1284
Chan CW, Hua S, Hong-Yue Z (2006) Application of fully decoupled parity equation in fault detection and identification of DC motors. IEEE Trans Indust Electron 53(4):1277–1284
Chen H, Hsu JY, Hsieh JY, Hsu HY, Chang CH, Lin YJ (2021) Predictive maintenance of abnormal wind turbine events by using machine learning based on condition monitoring for anomaly detection. J Mech Sci Technol 35(12):5323–5333
Devarajan G, Chinnusamy M, Kaliappan L (2021) Detection and classification of mechanical faults of three phase induction motor via pixels analysis of thermal image and adaptive neuro-fuzzy inference system. J Ambient Intell Humaniz Comput 12(5):4619–4630
Genyuan D, Qiang X, Yang X (2020) Fault diagnosis of rotating machinery components using a deep kernel extreme learning machine under different working conditions. Meas Sci Technol 31(11):115901
Hector S, Teresa E, Vicenc P, Fogh OP (2015) Fault diagnosis of an advanced wind turbine benchmark using interval-based ARRs and observers. IEEE Trans Indust Electron 62(6):3783–3793
Hundman K, Constantinou V, Laporte C, Colwell I, Soderstrom T (2018) Detecting spacecraft anomalies using lstms and nonparametric dynamic thresholding. In: Proceedings of the 24th ACM SIGKDD international conference on knowledge discovery and data mining, pages 387–395
Kim K, Parthasarathy G, Uluyol O, Foslien W, Sheng S, Fleming P (2011) Use of SCADA data for failure detection in wind turbines. Energy Sustain 54686:2071–2079
Koenker R, Bassett Jr G (1978) Regression quantiles. Econometrica 46(1):33–50
Kraus M, Feuerriegel S (2019) Forecasting remaining useful life: interpretable deep learning approach via variational bayesian inferences. Decis Support Syst 125:113100
Kusiak A, Zhang Z (2010) Analysis of wind turbine vibrations based on SCADA data. J Sol Energy Eng 132(3):1–12
Li S, Liu G, Tang X, Jianguang L, Jianjun H (2017) An ensemble deep convolutional neural network model with improved DS evidence fusion for bearing fault diagnosis. Sensors 17(8):1729
Lu S, Xu Q, Jiang C, Liu Y, Kusiak A (2022a) Probabilistic load forecasting with a non-crossing sparse-group lasso-quantile regression deep neural network. Energy 242:122955
Lu SX, Gao ZW, Xu QF, Jiang CX, Zhang AH, Wang XX (2022b) Class-imbalance privacy-preserving federated learning for decentralized fault diagnosis with biometric authentication. IEEE Trans Industr Inf 18(12):9101–9111
Meier L, van de Geer SA, Buhlmann P (2008) The group lasso for logistic regression. J Roy Stat Soc Ser B 70:53–71
Pedro S, Villa Luisa F, Aníbal R, Andres B, Jesús M (2015) An SVM-based solution for fault detection in wind turbines. Sensors 15(3):5627–5648
Saha S, Bovolo F, Bruzzone L (2022) Change detection in image time-series using unsupervised LSTM. IEEE Geosci Remote Sens Lett 19:8005205
Scardapane S, Comminiello D, Hussain A, Uncini A (2017) Group sparse regularization for deep neural networks. Neurocomputing 241:81–89
Schlechtingen M, Santos IF, Achiche S (2013) Wind turbine condition monitoring based on SCADA data using normal behavior models. Part 1: System description. Appl Soft Comput 13(1):259–270
Shang ZW, Liao XX, Geng R, Gao MS, Liu X (2018) Fault diagnosis method of rolling bearing based on deep belief network. J Mech Sci Technol 32(11):5139–5145
Shazib UM, Kumar S (2014) Energy, emissions and environmental impact analysis of wind turbine using life cycle assessment technique. J Clean Prod 69:153–164
Sudha MS, Valarmathi K (2021) An optimized deep belief network to detect anomalous behavior in social media. J Ambient Intell Human Comput. https://doi.org/10.1007/s12652-020-02708-2
Sun P, Li J, Wang CS, Lei X (2016) A generalized model for wind turbine anomaly identification based on SCADA data. Appl Energy 168:550–567
Tibshirani R (1996) Regression shrinkage and selection via the lasso. J Roy Stat Soc B 58(1):267–288
Wang Z, Ma H, Chen H, Yan B, Chu X (2020) Performance degradation assessment of rolling bearing based on convolutional neural network and deep long-short term memory network. Int J Prod Res 58(13):3931–3943
Wang J, Wang M, Liu Q, Yin G, Zhang Y (2022) Deep anomaly detection in expressway based on edge computing and deep learning. J Ambient Intell Human Comput 13:1293–1305. https://doi.org/10.1007/s12652-020-02574-y
Wei X, Verhaegen M, van Engelen T (2010) Sensor fault detection and isolation for wind turbines based on subspace identification and Kalman filter techniques. Int J Adapt Control Signal Process 24(8):687–707
Xu Q, Deng K, Jiang C, Sun F, Huang X (2017) Composite quantile regression neural network with applications. Expert Syst Appl 76:129–139
Xu Q, Fan Z, Jia W, Jiang C (2019) Quantile regression neural network-based fault detection scheme for wind turbines with application to monitoring a bearing. Wind Energy 22(10):1390–1401
Xu Q, Lu S, Zhai Z, Jiang C (2020) Adaptive fault detection in wind turbine via RF and CUSUM. IET Renew Power Gener 14(10):1789–1796
Xie ZX, Wen H (2019) Composite quantile regression long short-term memory network. In: 28th International Conference on Artificial Neural Networks (ICANN), volume 11730 of Lecture Notes in Computer Science, pages 513–524
Xing SB, Lei YG, Wang SH, Jia F (2021) Distribution-invariant deep belief network for intelligent fault diagnosis of machines under new working conditions. IEEE Trans Industr Electron 68(3):2617–2625
Yuan M, Lin Y (2006) Model selection and estimation in regression with grouped variables. J Roy Stat Soc 68(1):49–67
Zhao H, Li L (2016) Fault diagnosis of wind turbine bearing based on variational mode decomposition and teager energy operator. IET Renew Power Gener 11(4):453–460
Zhao H, Liu H, Wenjing H, Yan X (2018) Anomaly detection and fault analysis of wind turbine components based on deep learning network. Renew Energy 127:825–834
Zhao WL, Wang ZJ, Cai WA, Zhang QQ, Wang JY, Du WH, Yang NN, He XX (2022) Multiscale inverted residual convolutional neural network for intelligent diagnosis of bearings under variable load condition. Measurement 188:110511
Zhou XK, Hu YY, Liang W, Ma JH, Jin Q (2021) Variational LSTM enhanced anomaly detection for industrial big data. IEEE Trans Industr Inf 17(5):3469–3477
Zou H, Yuan M (2008) Composite quantile regression and the oracle model selection theory. Ann Stat 36(3):1108–1126
Acknowledgements
The authors are grateful to the Co-Editor and two anonymous referees for their helpful comments and constructive guidance. The authors gratefully acknowledge financial support from the Key Research and Development Program of Anhui Province (202004a05020020) and the National Natural Science Foundation of PR China (72171070). We also thank the partner, Anhui Ronds Science & Technology Incorporated Company, for providing the SCADA data of wind turbines.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The authors declare no conflict of interest regarding the publication of this paper.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Xu, Q., Wu, D., Jiang, C. et al. A composite quantile regression long short-term memory network with group lasso for wind turbine anomaly detection. J Ambient Intell Human Comput 14, 2261–2274 (2023). https://doi.org/10.1007/s12652-022-04484-7
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s12652-022-04484-7