|
| 基于图像识别的风力机全转速工况叶片表面故障识别研究 |
| Research on blade surface fault identification of wind turbines under full-speed operating conditions based on image recognition |
| 投稿时间：2024-10-11 |
| DOI：10.13259/j.cnki.eri.2026.01.004 |
| 中文关键词: 风力机 风电机组 叶片 故障检测 图像识别 停机检测 |
| 英文关键词:wind turbine blade fault identification image recognition shutdown inspection non-shutdown state inspection |
| 基金项目: |
|
| 摘要点击次数: 802 |
| 全文下载次数: 414 |
| 中文摘要: |
| 叶片是风力发电机组的关键部件之一，其性能对于风电机组安全稳定运行至关重要。为了保障风力机叶片的安全运行，搭建了一套无人机智能巡检系统，并结合YOLOv9图像识别算法实现风力机叶片智能巡检和故障智能识别。该系统支持风力机停机巡检和不停机在运状态巡检，可自主规划航线，能实现风力机全转速工况下叶片表面状态的监测与诊断。基于YOLOv9图像识别算法，该系统可对叶片风损、裂纹、蒙皮脱落和雷击等常见故障进行实时在线检测，效率高，成本低，可有效保障风力机的长周期安全、稳定运行。 |
| 英文摘要: |
| Blades are critical components of wind turbines, and their performance is essential for the safe and stable operation of the entire unit. To ensure safe operation, this study develops an intelligent inspection system using an unmanned aerial vehicle (UAV) equipped with the YOLOv9 image recognition algorithm for intelligent inspection and fault identification of wind turbine blades. The system supports both shutdown and in-operation inspections, autonomously planning flight routes to monitor and diagnose blade surface conditions under full-speed operating conditions. Based on the YOLOv9 algorithm, the system performs real-time detection of common faults, including wind damage, cracks, surface peeling, and lightning strikes. With high efficiency and low cost, this approach effectively ensures the long-term safe and stable operation of wind turbines. |
| HTML 查看全文 查看/发表评论 下载PDF阅读器 |
| 关闭 |
