The invention provides a full-automatic wind generating set blade intelligent inspection and diagnosis analysis method. The method comprises the following steps that an unmanned aerial vehicle calculates and reconstructs a fan model by utilizing machine vision and laser radar fusion, and an inspection path is automatically generated; the unmanned aerial vehicle takes off and confirms an operationstarting point through a control terminal, the unmanned aerial vehicle flies along an inspection path, and an optimal image semantic segmentation model is used for identifying blades for automatic inspection; the unmanned aerial vehicle sequentially inspects three blades of the fan according to the inspection path to completely acquire images of all the blades; the unmanned aerial vehicle splicesthe acquired blade foreground images under the same path to verify the integrity of the acquired blade foreground images; and fault identification is performed on the images based on a deep learning algorithm, category and severity evaluation are performed on the fault image, and the images are uploaded to a cloud platform, classified and stored, and subjected to digital processing. According to the full-automatic wind generating set blade intelligent inspection and diagnosis analysis method, machine vision, deep learning, image semantic segmentation technology and the like are combined, and full-automatic inspection and fault diagnosis of the unmanned aerial vehicle on the blades are realized.
本发明提供了一种全自动风力发电机组叶片智能巡视及诊断分析方法,包括如下步骤:无人机利用机器视觉和激光雷达融合计算重建风机模型,自动生成巡检路径;无人机起飞经控制端确认作业起始点,无人机沿巡检路径飞行,并利用最优图像语义分割模型识别叶片进行自动巡检;无人机根据巡检路径依次巡检风机三个叶片完整采集全部叶片的图像;无人机将采集的同一条路径下叶片前景图像进行拼接验证其采集的完整性;基于深度学习算法对图像进行故障识别,并对故障图像进行类别和严重程度的评定,图像上传至云端平台,进行分类存储,对图像进行数字化处理。本发明结合机器视觉、深度学习和图像语义分割技术等,实现无人机对叶片的全自动巡检以及故障诊断。
Full-automatic wind generating set blade intelligent inspection and diagnosis analysis method
一种全自动风力发电机组叶片智能巡视及诊断分析方法
2021-02-12
Patent
Electronic Resource
Chinese
European Patent Office | 2021
|Full-automatic tunnel safety intelligent inspection equipment
European Patent Office | 2022
|