基于人工和传统自动化检测算法在风电叶片表面裂纹和剥落腐蚀缺陷检测任务中存在精度低和效率差等问题,提出了基于优化YOLOv5s的风电叶片表面缺陷检测轻量化算法.首先,构建不同自然环境条件下叶片表面缺陷数据集;其次,在YOLOv5s中引入GhostNet的Ghost模块和Ghost BotleNeck结构;最后,针对真实框与预测框不一致的问题,引入损失函数SIoU,通过聚焦普通质量锚框的预测回归提升叶片表面缺陷识别准确率和定位精度.优化后模型的参数量、计算量和权重文件分别为YOLOv5s的52.6%、51.6%和55.2%,叶片表面缺陷识别平均精度均值为88.3%.与Faster-RCNN、SSD、YOLOv4、YOLOv5s和YOLOv8s模型相比,平均精度均值分别提高了34.2%、36.5%、24.5%、1.9%和1.8%.结果表明,YOLOv5s优化后成功防止漏检和误检,提升了目标定位精度,更适用于检测复杂工况下叶片表面裂纹和剥落腐蚀缺陷.
Given the low accuracy and poor efficiency of manual and traditional automated detection algorithms for identifying surface cracks and exfoliation corrosion defects in wind turbine blades, a lightweight algorithm based on optimized YOLOv5s for surface defect detection in wind turbine blades has been proposed. This algorithm involves the creation of a dataset encompassing blade surface defects observed under various natural conditions. Ghost modules and Ghost Bottleneck structures from the GhostNet are integrated into YOLOv5s to enhance its capabilities. Soft Intersection over Union (SIoU) is introduced to improve prediction regression, with a focus on standard quality anchor boxes, bridging the gap between actual and predicted bounding boxes. This enhancement significantly boosts the accuracy of blade surface defect recognition and improves localization precision. Furthermore, the optimized model reduces parameters, computational load, and the weight file size of YOLOv5s by 52.6%, 51.6%, and 55.2%, respectively. The algorithm achieves an average precision of 88.3% in recognizing blade surface defects. Comparative analysis against Faster-RCNN, SSD, YOLOv4, YOLOv5s, and YOLOv8s algorithms reveals improvements in average precision by 34.2%, 36.5%, 24.5%, 1.9%, and 1.8%, respectively. The experimental results indicate that the algorithm successfully prevents both missed detections and false positives after optimization, while also improving the accuracy of target localization, making it more suitable for detecting surface cracks and exfoliation corrosion defects on blades under complex working conditions.
[1] 张柏林, 张生杨, 邬博宇,等.废旧风力发电机叶片资源化利用研究进展[J].工程科学学报,2023,45(12):2150-2161.
[2] 尹玉, 张永, 王健,等.基于热红外图像的风力机叶片损伤识别方法研究[J].太阳能学报,2022,43(2):492-497.
[3] 王道累, 肖佳威, 刘易腾,等.风力机组叶片损伤检测技术研究与进展[J].中国电机工程学报,2023,43(12):4614-4631.
[4] Freitas P, Vieira G, Canario J, et al.A trained Mask R-CNN model over PlanetScope imagery for very-high resolution surface water mapping in boreal forest-tundra[J].Remote Sensing of Environment,2024,304:114047.
[5] Zhang Y, Ni Y Q, Jia X Y, et al.Identification of concrete surface damage based on probabilistic deep learning of images[J].Automation in Construction,2023,156:105141.
[6] 李凯, 林宇舜, 吴晓琳, 等.基于多尺度融合与注意力机制的小目标车辆检测[J].浙江大学学报(工学版),2022,56(11):2241-2250.
[7] 陈海永, 袁乐, 王世杰,等.基于多尺度编码互补注意力网络的光伏缺陷检测[J].太阳能学报,2023,44(10):191-197.
[8] Yu Y J, Cao H, Yan X Y, et al. Defect identification of wind turbine blades based on defect semantic features with transfer feature extractor[J].Neurocomputing,2020,376:1-9.
[9] Guo J H, Liu C, Cao J F, et al. Damage identification of wind turbine blades with deep convolutional neural networks[J].Renewable Energy,2021,174:122-133.
[10] Dwivedi D, Babu K V S M, Yemulapk, et al.Identification of surface defects on solar PV panels and wind turbine blades using attention based deep learning model[J].Engineering Applications of Artificial Intelligence,2024,131:107836.
[11] 蒋兴群, 刘波, 宋力,等.基于优化YOLO-v3的风力机叶片表面损伤检测识别[J].太阳能学报,2023,44(3):212-217.
[12] Zhang Y S, Tang Y L, Sun J Q, et al.Surface defect detection of wind turbine based on lightweight YOLOv5s model[J].Measurement,2023,220:113222;
[13] Jiang Y, Gong T Y, He L F, et al.Fall detection on embedded platform using infrared array sensor for healthcare applications[J].Neural Computing and Applications,2023,36(9):5093-5108.
[14] 龙燕, 杨智优, 何梦菲.基于优化YOLOv7的疏果期苹果目标检测方法[J].农业工程学报,2023,39(14):191-199.
[15] 何宇豪, 曹学国, 刘信良, 等.基于SW-YOLO模型的航空发动机叶片损伤实时检测[J].推进技术,2024,45(2):197-203.
[16] 邓天民, 程鑫鑫, 刘金凤,等.基于特征复用机制的航拍图像小目标检测算法[J].浙江大学学报(工学版),2024,58(3):437-448.
[17] Wang Z H, Luo K W, Yu H S, et al.NOx Emission prediction of heavy-duty diesel vehicles based on Bayesian optimization:gated recurrent unit algorithm[J].Energies,2024,292:130559.
[18] 陈辉, 田博, 赵永红, 等.基于改进YOLOv7的遥感图像目标检测方法[J].兰州理工大学学报,2026,52(1):93-100.
[19] Chen X, Zhang Y H, Li D L, et al.Chinese mitten crab detection and gender classification method based on GMNet-YOLOv4[J].Computers and Electronics in Agriculture,2023,214:108318.
[20] Qiu S, Cai B X, Wang W D, et al.Automated detection of railway defective fasteners based on YOLOv8-FAM and synthetic data using style transfer[J].Automation in Construction,2024,162:105363.
