With the rapid development of the wind power industry, the proportion of wind turbine failures resulting in downtime is also increasing, particularly yaw system failures, which account for nearly one-third (28.7%) of total downtime. To reduce downtime and operational costs, this paper proposes a deep learning model based on SCADA data, named CNN-smart_Linformer (CNN-SLinformer), for predicting the occurrence time of yaw system failures in wind turbines. This model introduces dynamic self-attention weight calculations for the linear projection matrix, allowing it to adaptively capture changes in the input sequence and significantly enhancing the model’s generalization ability in different operating environments. It combines the advantages of convolutional neural networks (CNN) in local feature extraction with the capability of SLinformer to capture long-term dependencies. Experimental results using actual SCADA data from wind farms show that the CNN-SLinformer model significantly improves prediction accuracy for yaw failure tasks, reducing the score to 144.50, while it also has a shorter runtime, providing an effective predictive tool for wind farms.
WANGY J, WANGR, TANAKAK,et al. Accelerating the energy transition towards photovoltaic and wind in China[J]. Nature,2023,619(7971):761-767.
[2]
HANJ L, WANGX, YANGX B,et al. Yaw system restart strategy optimization of wind turbines in mountain wind farms based on operational data mining and multi-objective optimization[J]. Engineering Applications of Artificial Intelligence,2023,126:107036.
[3]
MAHESHWARIZ, KENGNEK, BHATO. A comprehensive review on wind turbine emulators[J]. Renewable and Sustainable Energy Reviews, 2023, 180: 113297.
[4]
LINM, PORTÉ-AGELF. Wake meandering of wind turbines under dynamic yaw control and impacts on power and fatigue[J].Renewable Energy,2024,223:120003.
ZHAOS Z, LINT, GUOW M,et al. Overview of research on fault prediction,diagnosis and processing methods of wind turbine mandrel[J]. China Foundry Machinery & Technology,2021, 56(5):86-94.(in Chinese)
[7]
JINX H, XUZ W, QIAOW .Condition monitoring of wind turbine generators using SCADA data analysis[J]. IEEE Transactions on Sustainable Energy,2021,12(1):202-210.
[8]
WUZ Y, LIY T, WANGP .A hierarchical modeling strategy for condition monitoring and fault diagnosis of wind turbine using SCADA data[J].Measurement,2024,227:114325.
[9]
NASCIMENTOE G S, DE MELOT A C, MOREIRAD M .A transformer-based deep neural network with wavelet transform for forecasting wind speed and wind energy[J].Energy,2023,278:127678.
[10]
FONSECA SANTIAGO R ADA, BARBOSAN B, MERGULHÃOH G,et al .Data-driven models applied to predictive and prescriptive maintenance of wind turbine:a systematic review of approaches based on failure detection,diagnosis,and prognosis[J].Energies,2024,17(5):1010.
[11]
KUSIAKA, ZHANGZ J .Analysis of wind turbine vibrations based on SCADA data[J].Journal of Solar Energy Engineering,2010,132(3):031008.
[12]
ZHANGG Y, LIY T, ZHAOY .A novel fault diagnosis method for wind turbine based on adaptive multivariate time-series convolutional network using SCADA data[J]. Advanced Engineering Informatics,2023,57:102031.
[13]
BIR, ZHOUC K, HEPBURND M .Detection and classification of faults in pitch-regulated wind turbine generators using normal behaviour models based on performance curves[J]. Renewable Energy,2017,105:674-688.
DUANS M, ZHANGF, LIUJ,et al .Real-time analysis system for operation data of offshore wind turbines[J].Ship Engineering,2023,45(Sup.2):13-17.(in Chinese)
[16]
LIANGP F, DENGC, YUANX M,et al .A deep capsule neural network with data augmentation generative adversarial networks for single and simultaneous fault diagnosis of wind turbine gearbox[J].ISA Transactions,2023,135:462-475.
LIC, LIW H, WANGY H,et al .Predicting the antigenicity of H1N1 influenza by the gated twin-tower model based on transformer[J].Computer Science,2022,49(Sup.2):71-76.(in Chinese)
YANGG L, XIH, GONGJ R,et al .Short term traffic flow forecasting based on transformer[J].Computer Applications and Software,2024,41(3):169-173.(in Chinese)
[21]
LIS Y, JINX Y, XUANY,et al .Enhancing the locality and breaking the memory bottleneck of transformer on time series forecasting[EB/OL].2019:1907.00235.org/abs/1907.00235v3.
[22]
HOUY D, WANGJ J, CHENZ Q,et al. Diagnosisformer:an efficient rolling bearing fault diagnosis method based on improved Transformer[J]. Engineering Applications of Artificial Intelligence,2023,124:106507.
[23]
TURNBULLA, CARROLLJ, MCDONALDA. Combining SCADA and vibration data into a single anomaly detection model to predict wind turbine component failure[J].Wind Energy,2021,24(3):197-211.
[24]
郑小东 .设备全寿命周期管理与故障预测研究[D].绵阳:西南科技大学,2021: 28-30.
[25]
ZHENGX D .Research on equipment life cycle management and fault prediction[D].Mianyang:Southwest University of Science and Technology,2021: 28-30.(in Chinese)
[26]
WANGS N, LIB Z, KHABSAM,et al .Linformer:self-attention with linear complexity[J]. ArXiv e-Prints,2020:arXiv:2006. 04768.
[27]
LUJ, YAOJ, ZHANGJ, et al. Soft: Softmax-free transformer with linear complexity[J]. Advances in Neural Information Processing Systems, 2021, 34: 21297-21309.
[28]
QIUY N, FENGY H, INFIELDD .Fault diagnosis of wind turbine with SCADA alarms based multidimensional information processing method[J].Renewable Energy,2020,145:1923-1931.
[29]
MALDONADO-CORREAJ, TORRES-CABRERAJ, MARTÍN-MARTÍNEZS,et al .Wind turbine fault detection based on the transformer model using SCADA data[J].Engineering Failure Analysis,2024,162:108354.
[30]
KHANP W, BYUNY C. A Review of machine learning techniques for wind turbine’s fault detection,diagnosis,and prognosis[J]. International Journal of Green Energy,2024, 21(4): 771-786.
[31]
GAOZ W, ODGAARDP. Real-time monitoring,fault prediction and health management for offshore wind turbine systems[J].Renewable Energy, 2023, 218: 119258.
基金资助
国家自然科学基金资助项目(62262038)
National Nature ScienceFoundation of China(62262038)
甘肃省重点研发计划-工业项目(25YFGA045)
Key Research and Development Program of Gansu Province(25YFGA045)