摘要
为解决恶劣的矿山环境下,人工检测风机桨叶表面的裂纹、砂眼等损伤中难度大等问题,基于亚像素图像匹配技术,对图像进行分割、匹配,采用亚像素梯度匹配算法,将获取的待检测图像与原始图像进行匹配,找出最小的匹配值,当计算结果大于指定的阈值时,则认为桨叶出现损伤,从而完成对桨叶表面的损伤检测.研究结果表明:该方法有效提高风机桨叶表面损伤检测的准确性,计算精度达到了亚像素级.研究结论可为风机桨叶的表面损伤检测提供理论依据.
The wind turbine works in a harsh mining environment,so the blade surface of the wind turbine is susceptible to granular dust pollution,wind,and sand erosion,which will cause cracks,sand holes,and other damage to the blade,also the manual damage detection is inefficient and difficult to operate.Based on the sub-pixel map matching technology,the image is segmented and matched.The sub-pixel gradient matching algorithm is used to match the image to be detected with the original image to find the minimum matching value,and when the calculation result is greater than the specified threshold,the paddle is considered to be damaged,so as to complete the surface damage detection of the paddle.The research results show that the method can effectively improve the accuracy of surface damage detection of wind turbine blades,and the computational accuracy reaches sub-pixel level.The findings can provide a theoretical basis for the surface damage detection of wind turbine blades.
作者
殷孝雎
卢世昱
都治良
任娜
王黎明
赵婷婷
YIN Xiaoju;LU Shiyu;DU Zhiliang;REN Na;WANG Liming;ZHAO Tingting(School of Renewable Energy,Shenyang Institute of Engineering,Shenyang 110136,China)
出处
《辽宁工程技术大学学报（自然科学版）》
CAS
北大核心
2022年第5期439-445,共7页
Journal of Liaoning Technical University (Natural Science)
基金
国家自然科学基金(51777127)
关键词
风机桨叶
损伤
亚像素
图像识别
梯度匹配
fan blade
damage
sub-pixel
image recognition
gradient matching
作者简介
殷孝雎(1974-),男,辽宁阜新人,博士,教授,主要从事新能源发电和智能制造方面的研究.
