Research Article
Attention Mechanism-Based CNN-LSTM Model for Wind Turbine Fault Prediction Using SSN Ontology Annotation
Table 3
Experimental results of the model on the test datasets of No. 15 and No. 3 wind turbine.
| Algorithm | Wind turbine icing fault test dataset | Wind turbine yaw fault test dataset | No. 15 wind turbine | No. 3 wind turbine | | | | | | | | |
| CLA | 0.9646 | 0.9730 | 0.9634 | 0.9712 | 0.9821 | 0.9805 | 0.9754 | 0.9819 | LSTM | 0.9388 | 0.9227 | 0.9580 | 0.9400 | 0.9793 | 0.9803 | 0.9781 | 0.9792 | RNN | 0.7986 | 0.8083 | 0.7833 | 0.7956 | 0.9672 | 0.9609 | 0.9738 | 0.9673 | XGBoost | 0.9805 | 0.9875 | 0.9733 | 0.9804 | 0.9855 | 0.9906 | 0.9852 | 0.9873 |
|
|