Defect Detection System Based on UAV Images for Wind Turbine Blades
-
摘要: 本文针对目前风机叶片人工检测工作量大、效率低、缺陷检测准确率不高的问题,提出并设计了一种基于无人机图像的缺陷自动化检测系统。本文介绍了系统的图像采集系统、采集方法、缺陷检测原理及检测效果:系统以无人机为飞行载体实现了(风机叶片的)自动巡检,从而提高了巡检效率,降低了人工的工作量;通过图像分割及缺陷检测算法设计实现了缺陷可疑区域的自动检测;可见光加红外光双光融合提高了叶片缺陷自动识别的准确性。经过多次现场测试验证,本系统可以精确、快速地实现鼓包,裂纹和褶皱等缺陷的自动识别与检测。Abstract: This study proposes and designs an automatic defect detection system based on UAV images for wind turbine blades, aimed at alleviating problems with manual detection methods, such as low efficiency and inaccurate defect detection. This paper introduces the system's image acquisition system, acquisition method, defect detection principle, and detection result. This system uses a UAV as the flying carrier to realize automatic inspection of wind turbine blades, thereby improving the inspection efficiency and reducing the manual workload. Through image segmentation and defect detection algorithm design, automatic detection of suspicious defect areas is achieved. Double light fusion of visible and infrared light improves the accuracy of automatic blade defect recognition. After multiple field tests and verification, the system is shown to accurately and quickly realize the automatic identification and detection of defects, such as bulges, cracks, and wrinkles.
-
Key words:
- UAV /
- wind turbine blade /
- defect defection /
- image processing
-
表 1 可见光机芯参数表
Table 1. Parameters of visible light module
Parameters Value Pixels 5280×3956 CMOS 4"/3" Frame ratio 16:9 表 2 红外光机芯参数表
Table 2. Parameters of infrared light module
Parameters Value Wavelength range 7.5-14 μm Pixels 640×480 Temperature range 0℃-150℃ Measurement accuracy 2℃ -
[1] 贺斌, 贾建华, 赵峰, 等.无人机在风机叶片检测中的应用[J]. 电工技术, 2019(13): 64-65. doi: 10.3969/j.issn.1002-1388.2019.13.023HE Bin, JIA Jianhua, ZHAO Feng, et al. Application of UAV in detection of wind turbine blade[J]. Electric Engineering, 2019(13): 64-65. doi: 10.3969/j.issn.1002-1388.2019.13.023 [2] Menon S, Schoess J N, Hamza R, et al. Wavelet-based acoustic emission detection method with adaptive thresholding[C]//Proc. of SPIE, 2000, 3986: 71-78. [3] CUI Hongyu, DING Ning, HONG Ming. Crack diagnosis of wind turbine blades based on MD method[J]. IOP Conference Series: Materials Science and Engineering, 2016, 157(1): 012003. http://adsabs.harvard.edu/abs/2016MS&E..157a2003H [4] WANG Long, ZHANG Zijun. Automatic detection of wind turbine blade surface cracks based on UAV-taken images[J]. IEEE Transactions on Industrial Electronics, 2017, 64(9): 7293-7303. doi: 10.1109/TIE.2017.2682037 [5] 王凯, 韦宏利, 陈超波, 等.基于红外和可见光融合的目标跟踪[J]. 计算机系统应用, 2018, 27(1): 149-153. http://d.old.wanfangdata.com.cn/Periodical/gxsfdxxb201804007WANG Kai, WEI Hongli, CHEN Chaobo, et al. Target tracking based on infrared and visible light fusion[J]. Computer Systems & Applications, 2018, 27(1): 149-153. http://d.old.wanfangdata.com.cn/Periodical/gxsfdxxb201804007 [6] 翟永杰, 刘金龙, 程海燕.基于机器视觉的汽车工件抓取定位系统设计与开发[J]. 现代电子技术, 2018, 41(20): 6-9. http://www.cqvip.com/main/zcps.aspx?c=1&id=7000859221ZHAI Yongjie, LIU Jinlong, CHENG Haiyan. Design and development of automobile workpiece grasping and positioning system based on machine vision[J]. Modern Electronics Technique, 2018, 41(20): 6-9. http://www.cqvip.com/main/zcps.aspx?c=1&id=7000859221