Defect identification of wind turbine blade based on multi-feature fusion residual network and transfer learning

As a key part of the wind turbines (WTs), the blade has a direct influence on the efficiency of WT. Because the defect detection technology of WT blade is not widely used, and the robustness of traditional detection methods is poor, this paper proposes a multi-feature fusion residual network combined with transfer learning. In this paper, the WT blade image dataset is enhanced and constructed to train the convolutional network. Two residual structures of multi-feature fusion (two feature fusion and three feature fusion) are proposed and compared. At the same time, transfer learning is used to improve training process and accelerate convergence. Compared with several convolutional neural networks based on indices include training loss and testing accuracy, f1-score and confusion matrix, the method proposed greatly reduces the time while achieving accurate detection.

Automated summary

This paper proposes a method of multi-feature fusion residual network combined with transfer learning, enhancing and constructing a dataset of WT blade images to achieve accurate detection. The method greatly reduces the time while achieving accurate detection, as compared to several convolutional neural networks based on indices.