Wind turbine fault detection based on the transformer model using SCADA data
Source abstract
Abstract
The growth of installed wind power worldwide and its significant contribution to the energy market is mainly due to the evolution of wind turbines (WTs) and their ability to withstand a wide range of dynamic loads. WT failures can be costly and lead to extended downtime. Early detection of such failures is critical in reducing the costs associated with operation and maintenance (O&M) tasks and unscheduled shutdowns of WTs. This paper applies two Deep Learning (DL) models based on the Transformer model to predict failures in the IGBT module of WTs at an onshore wind farm in Ecuador. To this end, SCADA (Supervisory Control and Data Acquisition) operational and alarm data are used, together with the maintenance record (MR). These data are analyzed and processed, applying different feature selection methods. The results show that the two proposed models perform well, with high accuracy and an approximate prediction of 4.25 months before failure occurrence. The results are promising due to the possibility of using SCADA data for early and accurate identification of faults in different components of WTs. • SCADA data from a wind farm located in complex terrain is analyzed. • The Transformer model can effectively predict failures in the IGBT of wind turbines. • Enhanced feature selection methods to obtain key information related to IGBT faults. • Predictive models generate a warning before a failure occurs.
Source: openalex_inverted_index · Quality: baseline
Research context
OpenAlex topics: Machine Fault Diagnosis Techniques · Silicon Carbide Semiconductor Technologies · Power System Reliability and Maintenance
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
References (79)
Show source-provided references
1. Wind energy conversions, controls, and applications: A review for sustainable technologies and directions10.3390/su15053986 ↗
2. A review on failure modes of wind turbine components10.3390/en14175241 ↗
3. Using SCADA data for wind turbine condition monitoring: A systematic literature review10.3390/en13123132 ↗
4. Wind turbine gearbox condition monitoring and fault diagnosis based on multi-sensor information fusion of SCADA and DSER-PSO-WRVM method10.1080/02286203.2018.1476008 ↗
5. Wind turbine reliability: A comprehensive review towards effective condition monitoring development10.1016/j.apenergy.2018.07.037 ↗
6. SCADA data analysis methods for diagnosis of electrical faults to wind turbine generators10.3390/app11083307 ↗
7. Condition monitoring of wind turbines and extraction of healthy training data using an ensemble of advanced statistical anomaly detection models
8. Universal transformers
9. A text-mining approach to assess the failure condition of wind turbines using maintenance service history10.3390/en12101982 ↗
10. A generalized model for wind turbine faulty condition detection using combination prediction approach and information entropy
11. Deep anomaly detection in horizontal axis wind turbines using graph convolutional autoencoders for multivariate time series10.1016/j.egyai.2022.100145 ↗
12. Anomaly detection of wind turbine gearbox based on SCADA temperature data using machine learning
13. A data-mining approach for wind turbine fault detection based on SCADA data analysis using artificial neural networks10.3390/en14071845 ↗
14. Fault detection of offshore wind turbine drivetrains in different environmental conditions through optimal selection of vibration measurements10.1016/j.renene.2022.12.049 ↗
15. Feature selection algorithms for wind turbine failure prediction10.3390/en12030453 ↗
16. Performance monitoring of wind turbines gearbox utilising artificial neural networks—Steps toward successful implementation of predictive maintenance strategy10.3390/pr11010269 ↗
17. Wind turbine prognosis models based on SCADA data and extreme learning machines10.3390/app11020590 ↗
18. A data-centric machine learning methodology: application on predictive maintenance of wind turbines10.3390/en15030826 ↗
19. Wind turbine main bearing fault prognosis based solely on scada data10.3390/s21062228 ↗
20. Identifying health status of wind turbines by using self organizing maps and interpretation-oriented post-processing tools10.3390/en11040723 ↗
21. Optimal temperature-based condition monitoring system for wind turbines10.3390/infrastructures6040050 ↗
22. Research on fault detection for three types of wind turbine subsystems using machine learning10.3390/en13020460 ↗
23. An approach combining data mining and control charts-based model for fault detection in wind turbines10.1016/j.renene.2017.09.003 ↗
24. Reliability assessment of wind power converter considering scada multistate parameters prediction using fp-growth, wpt, k-means and lstm network10.1109/ACCESS.2020.2992089 ↗
25. Machine learning and deep learning in energy systems: A review10.3390/su14084832 ↗
26. Scientometric review of artificial intelligence for operations & maintenance of wind turbines: The past, present and future10.1016/j.rser.2021.111051 ↗
27. Fault diagnosis methods based on machine learning and its applications for wind turbines: A review10.1109/ACCESS.2021.3124025 ↗
28. Comparative study of deep learning models versus machine learning models for wind turbine intelligent health diagnosis systems10.1007/s13369-023-07810-z ↗
29. Attention is all you need
30. A review of wind turbines in complex terrain10.1016/j.ijft.2023.100289 ↗
31. Wind turbine blade icing diagnosis using hybrid features and Stacked-XGBoost algorithm10.1016/j.renene.2021.09.008 ↗
32. Imbalanced classification based on minority clustering synthetic minority oversampling technique with wind turbine fault detection application10.1109/TII.2020.3046566 ↗
33. A small-sample wind turbine fault detection method with synthetic fault data using generative adversarial nets10.1109/TII.2018.2885365 ↗
34. Wind turbine gearbox failure identification with deep neural networks10.1109/TII.2016.2607179 ↗
35. Condition monitoring of wind turbines based on spatio-temporal fusion of SCADA data by convolutional neural networks and gated recurrent units10.1016/j.renene.2019.07.033 ↗
36. Anomaly detection and critical SCADA parameters identification for wind turbines based on LSTM-AE neural network10.1016/j.renene.2021.03.078 ↗
37. Devlin · 2018
38. Transformer in action: A comparative study of transformer-based acoustic models for large scale speech recognition applications
39. Swin transformer: Hierarchical vision transformer using shifted windows
40. Variational transformer-based anomaly detection approach for multivariate time series10.1016/j.measurement.2022.110791 ↗
41. Methods for advanced wind turbine condition monitoring and early diagnosis: A literature review10.3390/en11051309 ↗
42. Toward reliable power electronics: Challenges, design tools, and opportunities10.1109/MIE.2013.2252958 ↗
43. A survey on wind turbine condition monitoring and fault diagnosis—Part I: Components and subsystems10.1109/TIE.2015.2422112 ↗
44. Condition monitoring of power electronics for offshore wind
45. A dual transformer model for intelligent decision support for maintenance of wind turbines
46. BTAD: A binary transformer deep neural network model for anomaly detection in multivariate time series data10.1016/j.aei.2023.101949 ↗
47. Multivariate time series anomaly detection with generative adversarial networks based on active distortion transformer
48. A three-dimensional ResNet and transformer-based approach to anomaly detection in multivariate temporal–spatial data10.3390/e25020180 ↗
49. A two-stage anomaly decomposition scheme based on multi-variable correlation extraction for wind turbine fault detection and identification10.1016/j.apenergy.2022.119373 ↗
50. Semi-supervised multivariate time series anomaly detection for wind turbines using generator SCADA data
51. Time-series anomaly detection with stacked transformer representations and 1D convolutional network10.1016/j.engappai.2023.105964 ↗
52. Adversarial sparse transformer for time series forecasting
53. Wu · 2020
54. TGAN-AD: Transformer-based GAN for anomaly detection of time series data10.3390/app12168085 ↗
55. Marino · 2022
56. Anomaly transformer: Time series anomaly detection with association discrepancy
57. TranAD: Deep transformer networks for anomaly detection in multivariate time series data10.14778/3514061.3514067 ↗
58. Wind power resource assessment in complex terrain: Villonaco case-study using computational fluid dynamics analysis10.1016/j.egypro.2016.12.127 ↗
59. Analysis of the annual energy production of the villonaco wind farm
60. Wind resource assessment and influence of atmospheric stability on wind farm design using computational fluid dynamics in the andes mountains, ecuador10.1016/j.enconman.2023.116972 ↗
61. Matching contrastive learning: An effective and intelligent method for wind turbine fault diagnosis with imbalanced SCADA data10.1016/j.eswa.2023.119891 ↗
62. Comparison of new anomaly detection technique for wind turbine condition monitoring using gearbox SCADA data10.3390/en13195152 ↗
63. Fault detection of wind turbine based on SCADA data analysis using CNN and LSTM with attention mechanism10.1016/j.measurement.2021.109094 ↗
64. A two-stage method based on extreme learning machine for predicting the remaining useful life of rolling-element bearings10.1016/j.ymssp.2020.106899 ↗
65. Correlation analysis based relevant variable selection for wind turbine condition monitoring and fault diagnosis
66. Multivariate SCADA data analysis methods for real-world wind turbine power curve monitoring10.3390/en14041105 ↗
67. An improved feature selection method based on random forest algorithm for wind turbine condition monitoring10.3390/s21165654 ↗
68. Fault detection of wind turbines using SCADA data and genetic algorithm-based ensemble learning10.1016/j.engfailanal.2023.107209 ↗
69. Ozdemir · 2018
70. Variable selection strategies and its importance in clinical prediction modelling10.1136/fmch-2019-000262 ↗
71. Covert · 2023
72. An embedded feature selection method for imbalanced data classification10.1109/JAS.2019.1911447 ↗
73. Spatio-temporal fusion neural network for multi-class fault diagnosis of wind turbines based on SCADA data10.1016/j.renene.2020.06.154 ↗
74. Wind turbine anomaly detection based on SCADA data mining10.3390/electronics9050751 ↗
75. A robust prescriptive framework and performance metric for diagnosing and predicting wind turbine faults based on SCADA and alarms data with case study10.3390/en11071738 ↗
76. Probabilistic extension of precision, recall, and F1 score for more thorough evaluation of classification models
77. Classification assessment methods
78. DeepWind: an accurate wind turbine condition monitoring framework via deep learning on embedded platforms
79. Towards a rigorous evaluation of time-series anomaly detection