Scholarlink

<p>Accurate damage detection on wind turbine blades is critical for the expanding wind power industry. This study proposes WTB-YOLO, a lightweight, high-precision object detection model specifically designed for blade damage detection. The model integrates FasterNet for efficient compression, incorporates a Bidirectional Feature Pyramid Network to enhance small damage detection, introduces a C3k2_RCM module for optimized feature utilization, and employs Wise-IoU for improved localization accuracy. Experimental results demonstrate WTB-YOLO’s superior performance with mAP@0.5 and mAP@0.5:0.95 scores of 98.0% and 76.3%, significantly outperforming YOLO11s (96.8% and 69.2%) and other state-of-the-art methods. This approach provides an effective solution for ensuring safe wind turbine operation and maintenance.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Surface Damage Detection of Wind Turbine Blades Based on WTB-YOLO

  • 期刊论文

  • 发表日期：

    2026年5月15日

查看全文
  • Rui Zhao,
  • Minghan Zhang,
  • Wei Sun

摘要

Accurate damage detection on wind turbine blades is critical for the expanding wind power industry. This study proposes WTB-YOLO, a lightweight, high-precision object detection model specifically designed for blade damage detection. The model integrates FasterNet for efficient compression, incorporates a Bidirectional Feature Pyramid Network to enhance small damage detection, introduces a C3k2_RCM module for optimized feature utilization, and employs Wise-IoU for improved localization accuracy. Experimental results demonstrate WTB-YOLO’s superior performance with mAP@0.5 and mAP@0.5:0.95 scores of 98.0% and 76.3%, significantly outperforming YOLO11s (96.8% and 69.2%) and other state-of-the-art methods. This approach provides an effective solution for ensuring safe wind turbine operation and maintenance.