Journal Article

GCB‐YOLO: A Lightweight Algorithm for Wind Turbine Blade Defect Detection

Zhang, Zhiming, Chen, Xiaoyan, Wei, Ze, Dong, Chaoyi, Zan, Weidong, Xue, Yao

2025

Overview

For the current visual detection methods of wind turbine blade defects, their detection models are usually excessively large, making it difficult to achieve a balance between model accuracy and inference speed. To address this problem, this paper introduces a lightweight wind turbine blade defect detection network, GCB‐YOLO, which attempts to maintain high detection accuracy and simultaneously achieve rapid detection speed. Initially, a GhostNet network was employed to replace a portion of the YOLOv5s backbone network responsible for feature extraction. This replacement serves to reduce the network's parameter size and computational load, thereby achieving compression of the feature extraction network. A coordinate attention (CA) mechanism is subsequently incorporated into the backbone network, which enhances its ability to focus on small defects. Finally, the neck network was ultimately replaced with a bidirectional feature pyramid network (BiFPN) to optimize multiscale feature fusion, bolstering its ability to discern small defects. A series of validation experiments were conducted using an image dataset gathered from real wind farms. Compared with YOLOv5s, GCB‐YOLO resulted in a 46.2% reduction in the number of model parameters. The improved model has a 7.5 MB volume. Hence, in GPU computation mode, the image detection speed reached 115.3 frames per second. More importantly, the proposed method achieves an mAP@0.5 of 94.72%, simplifying deployment on edge computing devices and simultaneously meeting the real‐time defect detection requirement with a sustained high level of detection accuracy.

• Publisher: John Wiley & Sons, Inc,Wiley
• Subject: Accuracy / attention mechanism / Defects / Edge computing / Feature extraction / Image detection / lightweight model / Object recognition / Parameters / Turbine blades / Turbines / Wind farms / Wind power / wind turbine blade defect detection / Wind turbines / YOLOv5