Abstract
The fault diagnosis and prediction technology of wind turbines are of great significance for increasing the power generation and reducing the downtime of wind turbines. However, most of the current fault detection approaches are realized by setting a single alarm threshold. Considering the complicated working conditions of wind farms, such methods are prone to ignore the fault, send out a false alarm, or leave insufficient troubleshooting time. In this work, we propose a gearbox fault prediction approach of wind turbines based on the supervisory control and data acquisition (SCADA) data. A stacking model composed of Random Forest (RF), Gradient Boosting Decision Tree (GBDT), and Extreme Gradient Boosting (XGBOOST) was constructed as the normal behavior model to describe the normal conditions of the wind turbines. We used the Mahalanobis distance (MD) instead of the residual to measure the deviation of the current state from the normal conditions of the turbines. By inputting the MD series into the proposed change-point detection algorithm, we can obtain the change point at which the fault symptom begins to appear, and thus achieving the fault prediction of the gearbox. The proposed approach is validated on the historical data of 5 wind turbines in a wind farm, which proves its effectiveness to detect the fault in advance.
Highlights
In recent years, due to the increasingly serious energy crisis and environmental pollution, many countries of the world have vigorously developed new energy sources
We can observe that the three models, Random Forest (RF), Gradient Boosting Decision Tree (GBDT), and XGBOOST, achieve better performance than other single models, and the stacking model based on these three models achieves the best performance with the highest R2 scores and the lowest mean absolute error (MAE) and root mean square error (RMSE)
Change point 1point indicates by the change-point detectionwhich algorithm which wereinmarked in theThe figure
Summary
Due to the increasingly serious energy crisis and environmental pollution, many countries of the world have vigorously developed new energy sources. The approaches for wind turbine fault detection/prediction through data mining algorithms are divided into two major ways. In [16], the authors proposed a model-based approach to predict generator bearing temperature and achieving the early detection of bearing and gearbox faults. The framework of the proposed method can be divided into four parts, which are the data preprocessing, feature selection, the normal behavior modeling process, and the fault prediction process, where several algorithms or their combinations were applied. During the data preprocessing process, we select the SCADA historical data under the normal working conditions and apply the Density Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm and the quartile method to recognize and filter out the three types of outliers and abnormal data.
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call