首页|基于数字图像处理的风电机组叶片裂纹损伤识别方法研究

基于数字图像处理的风电机组叶片裂纹损伤识别方法研究

扫码查看
为实现风电机组叶片损伤检测的高效化、智能化、便捷化,研究一种基于数字图像处理技术的风电机组叶片裂纹损伤识别以及裂纹类型判断和特征参数提取的方法.以无人机采集的风电机组叶片图像为研究对象,通过对比灰度化、滤波、阈值分割等图像处理步骤的多种算法,对形态学处理方法进行改进,首先选用平均值法对叶片图像进行灰度处理,其次使用中值滤波对图像进行去噪处理,再次使用Otsu阈值分割以实现裂纹区域的分割,然后基于改进的形态学方法提取出完善的叶片裂纹损伤区域,最后基于连通域原理完成裂纹区域的框取.基于上述算法设计风电机组叶片裂纹损伤识别系统以实现叶片裂纹图像检测的可视化处理、裂纹类型判断及裂纹特征参数提取等功能.结果表明,该系统对于风电机组叶片裂纹损伤检测具有可靠的识别精度,识别准确率为85%,实现了风电机组叶片裂纹损伤的自动识别与特征参数提取,提高了叶片裂纹损伤的检测效率.
RESEARCH ON CRACK DAMAGE IDENTIFICATION METHOD OF WIND TURBINE BLADES BASED ON DIGITAL IMAGE PROCESSING
The method for crack damage identification,crack type judgment and feature parameter extraction of wind turbine blades based on digital image processing technology is studied in order to realize the high efficiency,intelligence and convenience of wind turbine blade damage detection.Different algorithms of the graying,filtering and threshold segmentation image processing steps are compared using blade images collected by the drone.The morphological processing method is improved.Firstly,the average value method is selected to conduct grayscale processing on the blade image.Secondly,the median filter is used to denoise the image.Thirdly,the Otsu threshold segmentation method is used to achieve the segmentation of the crack area.Then the perfect blade crack damage area is extracted based on the optimized morphological method,and finally the crack area is framed based on the connected domain principle.A wind turbine blade crack damage identification system is designed based on the above selected and improved algorithms to realize the visual processing of blade crack image detection,crack type judgment and crack feature parameter extraction.The results show that the system has reliable identification accuracy for the crack damage detection of wind turbine blades,and the identification accuracy is 85%.The designed system realizes the automatic identification and feature parameter extraction of wind turbine blade crack damage,and improves the detection efficiency of blade crack damage.

wind turbineblades damagedigital image processingcrack damage identificationcharacteristic extractionidentification system

石腾、许波峰、李振、陈鹏

展开 >

河海大学江苏省输配电装备技术重点实验室,常州 213000

河海大学可再生能源发电技术教育部工程研究中心,南京 211100

上海交通大学海洋工程国家重点实验室,上海 200240

风电机组 叶片损伤 数字图像处理 裂纹损伤识别 特征提取 识别系统

江苏省输配电装备技术重点实验室自主科研课题中央高校基本科研业务费专项资金

2022JSSPD07B210202063

2024

太阳能学报
中国可再生能源学会

太阳能学报

CSTPCD北大核心
影响因子:0.392
ISSN:0254-0096
年,卷(期):2024.45(2)
  • 23