Spatiotemporal Feature Extraction from Integrated Multivariate Time Series via MCNN-LSTM for Wind Turbine Fault Diagnosis

Xiangjun Zeng; Xiangqing Fang; Binqiao Zhang; Chen Feng; Shengyuan Zhou

doi:10.35833/MPCE.2025.000413

欢迎来到能源电力期刊网！

English Version 期刊登录

您当前的位置：

首页 >

文章列表页 >

Spatiotemporal Feature Extraction from Integrated Multivariate Time Series via MCNN-LSTM for Wind Turbine Fault Diagnosis

Original Paper | 更新时间：2026-09-28

- Spatiotemporal Feature Extraction from Integrated Multivariate Time Series via MCNN-LSTM for Wind Turbine Fault Diagnosis
- Spatiotemporal Feature Extraction from Integrated Multivariate Time Series via MCNN-LSTM for Wind Turbine Fault Diagnosis
- MPCE 2026年14卷第5期页码：1768-1779
- 作者机构：
  
  1.College of Electrical Engineering & New energy, China Three Gorges University, Yichang, China
  2.Hubei Provincial Key Laboratory for Operation and Control of Cascaded Hydropower Station, China Three Gorges University, Yichang, China
  3.College of Electrical Engineering and Automation, Shandong University of Science and Technology, Qingdao, China
  4.Jinan Power Supply Company of State Grid Shandong Electric Power, Jinan, China
- 作者简介：
  
  Xiangjun Zeng
  Xiangqing Fang received the B.Eng. degree from Quancheng College, University of Jinan, Jinan, China, in 2022. He is currently pursuing the M.Eng. degree at China Three Gorges University, Yichang, China. His research interests include deep learning-based fault diagnosis of electromechanical equipment. Email:15908075013@163.com
  Binqiao Zhang received the B.Sc. degree in electrical technology from Henan University, Kaifeng, China, in 2001, the M.Sc. degree in computer application from China Three Gorges University, Yichang, China, in 2007, and the Ph.D. degree in system analysis and integration from Huazhong University of Science and Technology, Wuhan, China, in 2017. Since 2001, he has been engaged in teaching and research activities at the College of Electrical Engineering and New Energy, China Three Gorges University. Currently, he serves as an Associate Professor at the same institution. His research interests include the informatization of power industry and operation simulation of hydropower station. Email:zb@ctgu.edu.cn
  Chen Feng received the B.S. degree from Shandong University of Science and Technology, Qingdao, China, in 2012, and the M.S. degree from China University of Mining and Technology, Beijing, China, in 2016. From 2017 to 2018, he was a Visiting Scholar with the Ohio State University, Columbus, USA. He received the Ph.D. degree in electrical engineering from China University of Mining and Technology, in 2020. He is now the Associate Professor in Shandong University of Science and Technology. His research interests include anomaly detection, machine learning-based maintenance, and intelligent fault diagnosis for power equipment. Email: fchen@sdust.edu.cn
  Shengyuan Zhou received the B.Sc. degree from the School of Automation and Electrical Engineering, University of Jinan, Jinan, China, in 2018, the M.Sc. degree from the School of Electrical Engineering, Shandong University, Jinan, China, in 2021. Currently, he is working at Jinan Power Supply Company of State Grid Shandong Electric Power Company, Jinan, China. His main research interests include condition assessment and operation and maintenance of transmission line. Email: 453932802@qq.com
- 基金信息：
- DOI：10.35833/MPCE.2025.000413
  中图分类号：
- 收稿：2025-05-13，
  
  修回：2025-08-04，
  
  录用：2025-12-31，
  
  网络首发：2026-02-27，
  
  纸质出版：2026-09-15
- 稿件说明：
移动端阅览
Xiangjun Zeng, Xiangqing Fang, Binqiao Zhang, 等. Spatiotemporal Feature Extraction from Integrated Multivariate Time Series via MCNN-LSTM for Wind Turbine Fault Diagnosis[J]. MPCE, 2026,14(5):1768-1779.

Zeng Xiangjun,Fang Xiangqing,Zhang Binqiao,et al.Spatiotemporal Feature Extraction from Integrated Multivariate Time Series via MCNN-LSTM for Wind Turbine Fault Diagnosis[J].Journal of Modern Power Systems and Clean Energy,2026,14(05):1768-1779.
Xiangjun Zeng, Xiangqing Fang, Binqiao Zhang, 等. Spatiotemporal Feature Extraction from Integrated Multivariate Time Series via MCNN-LSTM for Wind Turbine Fault Diagnosis[J]. MPCE, 2026,14(5):1768-1779. DOI： 10.35833/MPCE.2025.000413.

Zeng Xiangjun,Fang Xiangqing,Zhang Binqiao,et al.Spatiotemporal Feature Extraction from Integrated Multivariate Time Series via MCNN-LSTM for Wind Turbine Fault Diagnosis[J].Journal of Modern Power Systems and Clean Energy,2026,14(05):1768-1779. DOI： 10.35833/MPCE.2025.000413.

摘要

Abstract

Wind turbine (WT) fault diagnosis using supervisory control and data acquisition (SCADA) data is challenged by severe class imbalance

which deteriorates the recognition of minority class. To address this issue

a spatiotemporal feature extraction framework from integrated multivariate time series (IMTS) is proposed

with coordinated designs in data representation

model architecture

and loss regularization. First

a dual-label IMTS is constructed to fuse multi-source SCADA data via sliding windows. Second

a spatiotemporal feature learning architecture is developed by integrating a multi-scale convolutional neural network (MCNN) and a stacked long short-term memory (LSTM) network and designing the MCNN-LSTM model. Finally

an improved weighted cross-entropy loss is proposed

which incorporates kernel density estimation-based sample probabilities as regularization

thereby directing model attention toward minority and hard-to-classify samples. Experimental results on a real SCADA dataset demonstrate that the proposed MCNN-LSTM model achieves a Macro-recall score of 0.943 and a G-mean score of 0.942

outperforming comparison models in comprehensive performance.

关键词

Keywords

references

Q. Zhong, S. Liu, C. Liu et al. , “Fault diagnosis of wind turbine gearboxes based on multisource signal fusion,” IEEE Transactions on Instrumentation and Measurement , vol. 74 , no. 4 , pp. 1 - 13 , Apr. 2025 .

L. Sheng, C. Li, M. Gao et al. , “A review of SCADA-based condition monitoring for wind turbines via artificial neural networks,” Neurocomputing , vol. 633 , no. 6 , p. 129830 , Jun. 2025 .

W. Gu, Z. Chen, Q. Li et al. , “Torque limit-based inertial control method based on delayed support for primary frequency control of wind turbines,” Journal of Modern Power System and Clean Energy , vol. 12 , no. 2 , pp. 561 - 570 , Jun. 2023 .

L. Zhang, H. Zhang, and G. Cai, “The multiclass fault diagnosis of wind turbine bearing based on multisource signal fusion and deep learning generative model,” IEEE Transactions on Instrumentation and Measurement , vol. 71 , p. 3514212 , May 2022.

Y. Zhu, C. Zhu, J. Tan et al. , “Fault detection of offshore wind turbine gearboxes based on deep adaptive networks via considering Spatio-temporal fusion,” Renewable Energy , vol. 200 , no. 11 , pp. 1023 - 1036 , Nov. 2022 .

H. Lai, J. Chen, K. Xiahou et al. , “Fault diagnosis of wind turbine bearing based on parameter-optimized VMD and multi-feature fusion,” Arabian Journal for Science and Engineering , vol. 50 , no. 8 , pp. 5425 - 5438 , Apr. 2025 .

T. Roy, D. K. Ray, and S. Chattopadhyay, “Skewness driven DFIG’s DC link bus short and open circuit faults diagnosis using rotor and stator currents,” IEEE Transactions on Energy Conversion , vol. 39 , no. 1 , pp. 107 - 115 , Mar. 2024 .

X. Zeng, M. Yang, C. Feng et al. , “Data-driven anomaly detection method based on similarities of multiple wind turbines,” Journal of Modern Power System and Clean Energy , vol. 12 , no. 3 , pp. 803 - 818 , May 2024.

Y. Zhu, B. Xie, A. Wang et al. , “Wind turbine fault detection and identification via self-attention-based dynamic graph representation learning and variable-level normalizing flow,” Reliability Engineering & System Safety , vol. 253 , no. 1 , p. 110554 , Jan. 2025 .

A. Wang, Y. Pei, Y. Zhu et al. , “Wind turbine fault detection and identification through self-attention-based mechanism embedded with a multivariable query pattern,” Renewable Energy , vol. 211 , pp. 918 - 937 , Jul. 2023 .

Y. Zhu, C. Zhu, J. Tan et al. , “Anomal y detection and condition monitoring of wind turbine gearbox based on LSTM-FS and transfer learning,” Renewable Energy , vol. 189 , no. 4 , pp. 90 - 103 , Apr. 2022 .

L. Xiang, X. Yang, A. Hu et al. , “Condition monitoring and anomaly detection of wind turbine based on cascaded and bidirectional deep learning networks,” Applied Energy , vol. 305 , p. 117925 , Jan. 2022 .

J. Liu, X. Wang, F. Xie et al. , “Condition monitoring of wind turbines with the implementation of spatio-temporal graph neural network,” Engineering Applications of Artificial Intelligence , vol. 121 , p. 106000 , May 2023.

X. Wen, K. Yang, P. Xin et al. , “A novel SE-weighted multi-scale hedging CNN approach for fault diagnosis of wind turbine,” Measurement Science And Technology , vol. 34 , no. 9 , p. 95905 , Jun. 2023 .

P. Xin, X. Zhang, C. Yuan et al. , “Fault diagnosis of wind turbine based on CNN-LSTM parallel network model,” Academic Journal of Engineering and Technology Science , vol. 6 , no. 12 , pp. 115 - 121 , Dec. 2023 .

J. Zhan, C. Wu, C. Yang et al. , “Condition monitoring of wind turbines based on spatial-temporal feature aggregation networks,” Renewable Energy , vol. 200 , no. 11 , pp. 751 - 766 , Nov. 2022 .

S. Sun, W. Hu, Y. Liu et al. , “Matching contrastive learning: an effective and intelligent method for wind turbine fault diagnosis with imbalanced SCADA data,” Expert Systems With Applications , vol. 223 , p. 119891 , Aug. 2023 .

H. Yi, Q. Jiang, G. Yan et al. , “Imbalanced classification based on minority clustering synthetic minority oversampling technique with wind turbine fault detection application,” IEEE Transactions on Industrial Informatics , vol. 17 , no. 9 , pp. 5867 - 5875 , Sept. 2021.

M. Qian and Y. Li, “A weakly supervised learning-based oversampling framework for class-imbalanced fault diagnosis,” IEEE Transactions on Reliability , vol. 71 , no. 1 , pp. 429 - 442 , Mar. 2022 .

S. Yang, Y. Zhou, X. Chen et al. , “Fault diagnosis of wind turbines with generative adversarial network-based oversampling method,” Measurement Science and Technology , vol. 34 , no. 4 , p. 044004 , Jan. 2023 .

Y. Zhang, Y. Zhang Y. Zhang et al. , “Multi-class data augmentation and fault diagnosis of wind turbine blades based on ISOMAP-CGAN under high-dimensional imbalanced samples,” Renewable Energy , vol. 243 , p. 122609 , Apr. 2025 .

X. Zeng, L. Xia, C. Feng et al. , “An improved conditional generative adversarial method for wind turbine gearbox fault diagnosis with imbalance data,” IEEE Sensors Journal , vol. 24 , no. 23 , pp. 38727 - 38739 , Nov. 2024 .

N. Anantrasirichai and D. Bull, “Defectnet: multi-class fault detection on highly-imbalanced datasets,” in Proceedings of 2019 IEEE International Conference on Image Processing , Taipei, China, Sept. 2019, pp. 2481 - 2485 .

S. Sun, T. Wang, and F. Chu, “A multi-learner neural network approach to wind turbine fault diagnosis with imbalanced data,” Renewable Energy , vol. 208 , no. 5 , pp. 420 - 430 , May 2023.

J. Guo, Y. Liu, Y. Li et al. , “Imbalanced learning for wind turbine blad e icing detection via spatio-temporal attention model with a self-adaptive weight loss function,” Expert Systems with Applications , vol. 229 , p. 120428 , Nov. 2023 .

Q. He, Y. Pang, G. Jiang et al. , “A spatio-temporal multiscale neural network approach for wind turbine fault diagnosis with imbalanced SCADA data,” IEEE Transactions on Industrial Informatics , vol. 17 , no. 10 , pp. 6875 - 6884 , Oct. 2021 .

M. Modirrousta, M. Shoorehdeli, M. Yari et al ., “Imbalanced classification in faulty turbine data: new proximal policy optimization,” IET Collaborative Intelligent Manufacturing , vol. 6 , no. 3 , p. 12144 , Jan. 2023 .

J. Ma, Y. Fu, T. Cheng et al. , “SCADA data-driven spatio-temporal graph convolutional neural network for wind turbine fault diagnosis,” IEEE Transactions on Instrumentation and Measurement , vol. 74 , p. 3521910 , Mar. 2025 .

J. Guo, C. Chong, P. Abreu et al. , “Reparameterization convolutional neural networks for handling imbalanced datasets in solar panel fault classification,” Engineering Applications of Artificial Intelligence , vol. 150 , p. 110541 , Jun. 2025 .

Z. Xu, M. Bashir, Y. Yang et al. , “Multisensory collaborative damage diagnosis of a 10 MW floating offshore wind turbine tendons using multi-scale convolutional neural network with attention mechanism,” Renewable Energy , vol. 199 , no. 11 , pp. 21 - 34 , Nov. 2022 .

Z. Xu, X. Mei, X. Wang et al. , “Fault diagnosis of wind turbine bearing using a multi-scale convolutional neural network with bidirectional long short term memory and weighted majority voting for multi-sensors,” Renewable Energy , vol. 182 , pp. 615 - 626 , Jan. 2022 .

C. Sudre, W. Li, T. Vercauteren et al ., “Generalised Dice overlap as a deep learning loss function for highly unbalanced segmentations,” in Proceedings of International Workshop on Deep Learning in Medical Image Analysis , Québec City, Canada, Sept. 2017, pp. 240 - 248 .

H. Majzoub, I. Elgedawy, O. Akaydin et al. , “HCAB-SMOTE: a hybrid clustered affinitive borderline SMOTE approach for imbalanced data binary classification,” Arabian Journal for Science and Engineering , vol. 45 , no. 4 , pp. 3205 - 3222 , Apr. 2020 .

M. Kumari and N. Subbarao, “A hybrid resampling algorithms SMOTE and ENN based deep learning models for identification of marburg virus inhibitors,” Future Medicinal Chemistry , vol. 14 , no. 10 , pp. 701 - 715 , May 2022.

Y. Pang, G. Jiang, Q. He et al. , “Multi kernel fusion convolutional neural network for wind turbine fault diagnosis,” in Proceedings of 2019 Chinese Automation Congress , Hangzhou, China, Nov. 2019 , pp. 2871 - 2876 .

Z. Kong, B. Tang, L. Deng et al. , “Condition monitoring of wind turbines based on spatio-temporal fusion of SCADA data by convolutional neural networks and gated recurrent units,” Renewable Energy , vol. 146 , pp. 760 - 768 , Feb. 2020 .

T. Huang, Q. Zhang, X. Tang et al. , “A novel fault diagnosis method based on CNN and LSTM and its application in fault diagnosis for complex systems,” Artificial Intelligence Review , vol. 55 , no. 2 , pp. 1289 - 1315 , Feb. 2022 .

J. He, W. Huang, Y. Liu et al. , “Data imbalance fault diagnosis method based on an ensemble multi-scale convolutional attention network,” Mechanical Systems and Signal Processing , vol. 236 , p. 112934 , Aug. 2025 .

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

Multi-turbine Cooperative Frequency Regulation Control for Wind Farms Considering Torque-limited Inertial Control Power Characteristics

Asynchronous Hierarchical Volt/var Control for Offshore Wind Farms

Detection and Mitigation of False Data Injection Attacks Against Wind Farm Active Power Controllers in Power Grids

A Fault Diagnosis Method for Smart Meters via Two-layer Stacking Ensemble Optimization and Data Augmentation

Fault Location and Classification for Distribution Systems Based on Deep Graph Learning Methods