Mechanical defect recognition method for transmission line structural components based on association rules

Due to the different mechanical stresses experienced by transmission line structural components, the mechanical defect data after sample cutting and expansion exhibit different characteristics, resulting in overfitting of structural component data and poor positioning performance. Therefore, a correlation rule-based method for identifying mechanical defects in transmission line structural components is studied. Adaptive weighted fusion of multi-source sensing data for transmission line structural components, extraction of mechanical defect edge features through point cloud binary fitting, and overfitting control of transmission line structural component data; Extract mechanical damage feature points, calculate the clustering distance of association rules, and after distance mapping, directly locate the mechanical defect target of the structural component. The experimental results show that this method can effectively fuse data from different sensors, the image peak signal to noise ratio after over fitting control is high, the mechanical defect target location effect of structural parts is better, and the accurate recognition of different types of mechanical defects can be completed.