Design and analysis of wind turbine fault diagnosis system based on convolutional neural network

• Autorzy: Luo Xiaoli , Ning Jinye , Wu Min

Identyfikatory

DOI

10.29354/diag/201334

• Języki publikacji: EN

Abstrakty

EN

Wind turbines are apt to diverse faults during long-term operation in natural environments, which affect their power generation efficiency and lifespan. Therefore, based on convolutional neural networks, gradient descent method was introduced to optimize their parameter training. Meanwhile, synchronous compressed wavelet transform was utilized to enhance the fault signal's time-frequency information. The fault detection correlation operation was optimized through Pearson correlation coefficient. Finally, a new type of fan fault detection model was proposed. The average fault detecting accuracy of this model was the highest at 98.98%, the average loss value was the lowest at 0.08%, and the average time consumption was the shortest at 16.52s. The minimum mean square error for detecting inner and outer ring pitting of fan bearings was 0.016 and 0.018, respectively. As a result, the proposed new model performs excellently in terms of accuracy and reliability in fault detection, with detection accuracy generally superior to other existing models. This model can significantly improve wind turbine fault detection, reduce false alarm and false alarm rates, and provide effective guarantees for wind turbines' stable operation.

Słowa kluczowe

EN

convolutional neural network; wind turbines; fault; bearings; synchrosqueezed wavelet transform

• Wydawca: Polskie Towarzystwo Diagnostyki Technicznej
• Czasopismo: Diagnostyka
• Rocznik: 2025
• Tom: Vol. 26, No. 1
• Strony: art. no. 2025113
• Opis fizyczny: Bibliogr. 20 poz., rys., tab.

Twórcy

autor

Luo Xiaoli

• Scientific Research Department, Hunan Electrical College of Technology, Xiangtan, 411101, China

• https://orcid.org/0009-0005-1650-9164

autor

Ning Jinye

• Institute of Big Data and Artificial Intelligence Application Technology, Hunan Electrical College of Technology, Xiangtan, 411101, China
• School of Mechanical Engineering, Hunan University of Science and Technology, Xiangtan City, 411201, China

• https://orcid.org/0009-0005-4159-0283

autor

Wu Min

• Scientific Research Department, Hunan Electrical College of Technology, Xiangtan, 411101, China

• https://orcid.org/0009-0009-9857-1939

Bibliografia

• 1. Yi H, Jiang Q, Yan X. Imbalanced classification based on minority clustering synthetic minority oversampling technique with wind turbine fault detection application. IEEE Transactions on Industrial Informatics. 2021;17(9):5867-5875. https://doi.org/10.1109/TII.2020.3046566.
• 2. Liu X, Yang L, Zhang Z. Short-term multi-step ahead wind power predictions based on a novel deep convolutional recurrent network method. IEEE Transactions on Sustainable Energy. 2021;12(3):1820-1833. https://doi.org/10.1109/TSTE.2021.3067436.
• 3. Corley B, Koukoura S, Carroll J, McDonald A. Combination of thermal modelling and machine learning approaches for fault detection in wind turbine gearboxes. Energies. 2021;14(5):1375-1376. https://doi.org/10.3390/en14051375.
• 4. Badihi H, Zhang Y, Pillay P. Fault-tolerant individual pitch control for load mitigation in wind turbines with actuator faults. IEEE Transactions on Industrial Electronics. 2021;68(1):532-543. https://doi.org/10.1109/TIE.2020.2965479.
• 5. Cui Y, Bangalore P, Tjernberg LB. A fault detection framework using recurrent neural networks for condition monitoring of wind turbines. Wind Energy. 2021;24(11):1249-1250. https://doi.org/10.1002/we.2628.
• 6. Arasteh A, Zeni L, Cutululis NA. Fault ride through capability of grid forming wind turbines: A comparison of three control schemes. IET Renewable Power Generation. 2022; 16(9): 1866-1881. https://doi.org/10.1049/icp.2021.1349.
• 7. Wu Y, Ma X. A hybrid LSTM-KLD approach to condition monitoring of operational wind turbines. Renewable Energy. 2022; 181(1): 554-566. https://doi.org/10.1016/j.renene.2021.09.067.
• 8. Rahimilarki R, Gao Z, Jin N, Zhang A. Convolutional neural network fault classification based on timeseries analysis for benchmark wind turbine machine. Renewable Energy. 2022; 185(2): 916-931. https://doi.org/10.1016/j.renene.2021.12.056.
• 9. Xu Z, Mei X, Wang X, Yue M, Jin J, Yang Y, Li C. Fault diagnosis of wind turbine bearing using a multiscale convolutional neural network with bidirectional long short term memory and weighted majority voting for multi-sensors. Renewable Energy. 2022, 182(1): 615-626. https://doi.org/10.1016/j.renene.2021.10.024.
• 10. Guo S, Yang T, Hua H, Cao J. Coupling fault diagnosis of wind turbine gearbox based on multitask parallel convolutional neural networks with overall information. Renewable Energy. 2021; 178(11): 639-650. https://doi.org/10.1016/j.renene.2021.06.088.
• 11. Xing Z, Chen M, Cui J, Chen Z, Xu J. Detection of magnitude and position of rotor aerodynamic imbalance of wind turbines using convolutional neural network. Renewable Energy. 2022; 197(9): 1020-1033. https://doi.org/10.1016/j.renene.2022.07.152.
• 12. Zhang Z, Doganaksoy N. Change point detection and issue localization based on fleet-wide fault data. Journal of Quality Technology. 2022; 54(1): 453-465. https://doi.org/10.1080/00224065.2021.1937409.
• 13. Zheng X, Zeng Y, Zhao M, Vebkatesh B. Early identification and location of short-circuit fault in gridconnected AC microgrid. IEEE Transactions on Smart Grid. 2021;12(4):2869-2878. https://doi.org/10.1109/TSG.2021.3066803.
• 14. Dao PB. Condition monitoring and fault diagnosis of wind turbines based on structural break detection in SCADA data. Renewable Energy. 2022;185(3):641-654. https://doi.org/10.1016/j.renene.2021.12.051.
• 15. Peng Y, Qiao W, Qu L. Compressive sensing-based missing-data-tolerant fault detection for remote condition monitoring of wind turbines. IEEE Transactions on Industrial Electronics. 2021;69(2): 1937-1947. https://doi.org/10.1109/TIE.2021.3057039.
• 16. Wang X, Tang G, Yan X, He Y, Zhang X, Zhang C. Fault diagnosis of wind turbine bearing based on optimized adaptive chirp mode decomposition. IEEE Sensors Journal. 2021; 21(12): 13649-13666. https://doi.org/10.1109/JSEN.2021.3071164.
• 17. Jin X, Nian H. Overvoltage suppression strategy for sending AC grid with high penetration of wind power in the LCC-HVdc system under commutation failure. IEEE Transactions on Power Electronics. 2021;36(9):10265-10277. https://doi.org/10.1109/TPEL.2021.3066641.
• 18. Wang MH, Lu SD, Hsieh CC. Fault detection of wind turbine blades using multi-channel CNN. Sustainability. 2022;14(3):1781-1783. https://doi.org/10.3390/su14031781.
• 19. Vives J. Vibration analysis for fault detection in wind turbines using machine learning techniques. Advances in Computational Intelligence. 2022;2(1):15-16. https://doi.org/10.1007/s43674-021-00029-1.
• 20. Hebbi C, Mamatha H. Comprehensive dataset building and recognition of isolated handwritten Kannada characters using machine learning models. Artificial Intelligence and Applications. 2023; 1(3): 179-190. https://doi.org/10.47852/bonviewAIA3202624.