基于可见光和热成像的风机叶片全周期无损检测综述

风机叶片在制造、服役和维修阶段的无损检测非常重要。叶片长期在高强度的风力载荷下工作，制造过程产生的任何微小缺陷将在服役中扩大，进一步威胁到风机的正常运行。因而，风机叶片的无损检测一直是工业界与学术界探索的难题。根据叶片视觉检测方法结合无人机技术应用、相关数据包括图像处理方法以及缺陷评判方法的智能程度等方面对前人以及作者所在课题组的前期工作进行综述、总结、分析与对比。目前，可见光视觉检测与红外热成像检测等以视觉为基础的检测手段满足了风机叶片在役运维时非接触、高效、低成本、安全等需求。视觉检测与无人机巡检技术相结合能最大程度的保证人员安全，同时克服了望远镜检测视野受限的难题。然而该检测手段在风机叶片巡检中目前尚存在缺陷定量难、内部缺陷识别率低等方面的不足。通过分析对比可见光检测与热成像检测技术，认为结合智能算法的无人机搭载双光融合检测手段未来有望于解决风机叶片检测中存在的不足。

A Review: Full-cycle Nondestructive Testing Based on Visible Light and Thermography of Wind Turbine Blade

Non-destructive testing of wind turbine blades is very important during manufacturing, service, and maintenance. The blade works under the high-strength wind load for a long time, thus any tiny defects produced in the manufacturing process will expand in service, which further threatens the normal operation of the fan. Therefore, the non-destructive testing of blades has always been a difficult problem in industry and academia. According to the visual inspection method, combined with the application of unmanned arrial vehicle(UAV) technology, related data including image processing method and the intelligence of defect evaluation method, the previous work of predecessors will be summarized, analyzed and compared. At present, vision-based detection methods such as visible light visual detection and infrared thermal imaging detection have met the requirements on non-contact, high efficiency, low cost, safety, and so on. The combination of visual inspection and UAV inspection technology could ensure the safety of personnel, and overcome the problem of limited field of vision of telescope inspection. However, this detection method still has some shortcomings, such as difficulty quantifying the defects and low recognition rate of internal defects. Through the analysis and comparison of visible light detection and thermal imaging detection technology, a conclusion can be drawn that the UAV equipped with a dual light fusion detection method combined with an intelligent algorithm is expected to solve the shortcomings of wind turbine blade detection in the future.