Abnormal vibration detection of wind turbine based on temporal convolution network and multivariate coefficient of variation

Mechanical Systems and Signal Processing, Journal Year: 2022, Volume and Issue: 174, P. 109082 - 109082

Published: April 2, 2022

Language: Английский

---

A novel fault diagnosis method for wind turbine based on adaptive multivariate time-series convolutional network using SCADA data

Advanced Engineering Informatics, Journal Year: 2023, Volume and Issue: 57, P. 102031 - 102031

Published: June 13, 2023

Language: Английский

---

Prediction of Wind Turbine Blades Icing Based on CJBM With Imbalanced Data

IEEE Sensors Journal, Journal Year: 2023, Volume and Issue: 23(17), P. 19726 - 19736

Published: July 21, 2023

Supervisory control and data acquisition (SCADA) is widely used in wind farms as an effective system for turbines (WTs). However, practical engineering applications, it difficult us to have adequate conditions collect enough WT blade icing data, which leads imbalance uneven distribution the feature space. Using classical synthetic minority oversampling technique (SMOTE) balance may increase overlap of positive negative samples, or produce some redundant samples without useful information. A center jumping boosting machine (CJBM) method proposed that combines improved clustering-based (γ mini density peaks clustering SMOTE, γMiniDPC-SMOTE) light gradient (LightGBM) prediction. First, solve problem imbalanced a ${\gamma }$ MiniDPC-SMOTE proposed, divides into multiple clusters, then increases alleviates Second, calculating intercept distance notation="LaTeX">${d}_{c}$ based on binary search adaptive selection DPC parameters step phenomenon notation="LaTeX">$\gamma $ verified by -step two are proposed. Then, low operating efficiency model under large amount LightGBM training Finally, validation was performed SCADA datasets. The results showed accuracy, precision, recall, F1-measure, running times increased significantly, proving superiority CJBM.

Language: Английский

---

ResDenIncepNet-CBAM with principal component analysis for wind turbine blade cracking fault prediction with only short time scale SCADA data

Measurement, Journal Year: 2023, Volume and Issue: 212, P. 112696 - 112696

Published: March 10, 2023

Language: Английский

---

A Review of machine learning techniques for wind turbine’s fault detection, diagnosis, and prognosis

International Journal of Green Energy, Journal Year: 2023, Volume and Issue: 21(4), P. 771 - 786

Published: May 29, 2023

Wind turbines are becoming increasingly important in the generation of clean, renewable energy worldwide. To ensure their dependable and accessible operation, advanced real-time condition monitoring technology must be implemented to guarantee efficient wind power financial viability. Machine learning (ML) has emerged as a crucial technique for systems recent years. This is especially relevant because dedicated systems, primarily focused on vibration measurements, prohibitively expensive. Preventive maintenance most effective way detect address issues before they impact performance. article provides comprehensive up-to-date review latest technologies fault detection, diagnosis, prognosis turbines, with particular focus ML algorithms critical faults failure modes, preprocessing methods, evaluation metrics. Numerous references have been analyzed evaluate past, present, potential future research development trends this field. Most these based journal articles, theses, reports found open literature.

Language: Английский

---