Multi-fault diagnosis method for wind power generation system based on recurrent neural network

Abstract

With the continuous expansion of the scale of wind turbine system, wind power production, operation and equipment control of wind turbine have become more and more significant. To improve the reliability of wind turbine systems fault diagnosis, combining with data-driven technology, this paper proposes a multi-fault diagnosis method for wind power system based on recurrent neural network. According to the actual wind speed data, the normal operation and fault data of the wind turbine system are obtained by system modeling, and the classification and prediction model based on the recurrent neural network algorithm is established, which takes 30 characteristic parameters such as wind speed, rotor speed, generator speed and power generation as input, and 10 different types faults labels of the wind turbine as output. Specific rules formed inside the sample data of the wind turbine system are learned intelligently by the model which is continuously trained, optimized and tested to verify the feasibility of the algorithm. The results of evaluation standards such as accuracy rate, missed detection rate and F1-measure that compared with other related algorithms such as deep belief network show that the proposed algorithm can solve the problem of multi-classification fault diagnosis for wind power generation system efficiently.

Keywords

Wind turbine fault diagnosis data driven deep learning classification prediction

Get full access to this article

View all access options for this article.

References

Simani

Farsoni

Castaldi

. Fault diagnosis of a wind turbine benchmark via identified fuzzy models. IEEE Transac Ind Electron 2015; 62: 3775–3782.

. Fault diagnosis of variable pitch for wind turbines based on the multi-innovation forgetting gradient identification algorithm. Nonlinear Dyn 2015; 79: 2069–2077.

Jlassi

, et al. Multiple open-circuit fsaults diagnosis in back-to-back converters of PMSG drives for wind turbine systems. IEEE Transac Power Electron 2015; 30: 2689–2702.

Liu

, et al. A new wind turbine fault diagnosis method based on the local mean decomposition. Renew Energy 2012; 48: 411–415.

Zhang

. Rotor fault slot positioning of doubly fed induction generators based on wavelet decomposition. Lecture Notes Electric Eng 2015; 334: 47–54.

, et al. Sparse feature identification based on union of redundant dictionary for wind turbine gearbox fault diagnosis. IEEE Transac Ind Electron 2015; 62: 6594–6605.

Gerber

Martin

Mailhes

. Time-frequency tracking of spectral structures estimated by a data-driven method. IEEE Transac Ind Electron 2015; 62: 6616–6626.

Joshuva

Sugumaran

. A data driven approach for condition monitoring of wind turbine blade using vibration signals through best-first tree algorithm and functional trees algorithm: a comparative study. ISA Transac 2017; 67: 160–172.

Zhang

, et al. A data-driven design for fault detection of wind turbines using random forests and XGBoost. IEEE Access 2018; 6: 21020–21031.

10.

Liu

, et al. The application of intelligent fuzzy inference to the fault diagnosis in pitch-controlled system. Energy Proc 2012; 16: 1839–1844.

11.

Bakri

Boumhidi

. Fuzzy model-based faults diagnosis of the wind turbine benchmark. Proc Comput Sci 2018; 127: 464–470.

12.

Feng

Liang

. Complex signal analysis for wind turbine planetary gearbox fault diagnosis via iterative atomic decomposition thresholding. J Sound Vib 2014; 333: 5196–5211.

13.

Yan

Xiang

. A new wind turbine fault diagnosis method based on ensemble intrinsic time-scale decomposition and WPT-fractal dimension. Renew Energy 2015; 83: 767–778.

14.

Tang

, et al. Fault diagnosis for a wind turbine transmission system based on manifold learning and Shannon wavelet support vector machine. Renew Energy 2014; 62: 1–9.

15.

Yang

Liu

Chen

. Fault diagnosis for a wind turbine generator bearing via sparse representation and shift-invariant K-SVD. IEEE Transac Ind Inform 2017; 13: 1321–1331.

16.

, et al. A data-driven residual-based method for fault diagnosis and isolation in wind turbines. IEEE Transac Sustain Energy 2018; 10: 1109–1109.

17.

Bessa

IVD

Palhares

Filho

JEC

. Data-driven fault detection and isolation scheme for a wind turbine benchmark. Renew Energy 2016; 87: 634–645.

18.

Pashazadeh

Salmasi

Araabi

. Data driven sensor and actuator fault detection and isolation in wind turbine using classifier fusion. Renew Energy 2017; 116: 99–106.

19.

Joshuva

Sugumaran

20.

Jiang

Yan

, et al. Multiscale convolutional neural networks for fault diagnosis of wind turbine gearbox. IEEE Transac Ind Electron 2018; 66: 3196–3207.

21.

Qian

Cross

. Integrated data-driven model-based approach to condition monitoring of the wind turbine gearbox. IET Renew Power Generat 2017; 11: 1177–1185.

22.

Odgaard

Stoustrup

. Fault tolerant control of wind turbines using unknown input observers. IFAC Proc Vol 2012; 45: 313–318.

23.

Odgaard

Stoustrup

. A benchmark evaluation of fault tolerant wind turbine control concepts. IEEE Transac Control Syst Technol 2015; 23: 1221–1228.

24.

Mesnil

, et al. Using recurrent neural networks for slot filling in spoken language understanding. IEEE/ACM Transac Audio Speech and Lang Process 2015; 23: 530–539.

25.

Baolin

Kaisheng

. Recurrent neural networks with external memory for spoken language understanding. Interspeech 2015; 10: 25–35.

26.

Jozefowicz

Zaremba

Sutskever

. An empirical exploration of recurrent network architectures. Int Conf Int Conf Mach Learn 2015; 37: 2342–2350.

27.

Chen

Xiahou

, et al. A radically data-driven method for fault detection and diagnosis in wind turbines. Electric Power Energy Syst 2018; 99: 577–584.