输电线路锈蚀缺陷无人机自主飞行巡检方法OA

[Objective]Overhead high-voltage transmission lines,as an important component of the power system,are exposed to the natural environment for a long time.Their metal components are prone to rust,which seriously affects the safe and stable operation of the power grid.The traditional manual inspection method is limited by the danger and complexity of high-altitude operations,making it difficult to achieve comprehensive and accurate defect detection.Although the existing inspection methods based on unmanned aerial vehicles(UAVs)have improved in efficiency,they still have deficiencies in aspects such as the acquisition of three-dimensional spatial information,the accuracy of path planning,and the accuracy of rust identification.An autonomous flight inspection method based on UAVs is expected to be developed combining laser three-dimensional(3D)modeling and support vector machine(SVM),so as to enhance the accuracy and efficiency of rust detection for transmission lines and provide reliable technical support for the intelligent operation and maintenance.[Methods]The multi-technology integration strategy was adopted to achieve the precise detection of rust defects in transmission lines.A laser 3D scanner was used to scan the transmission lines and their surrounding environment.Based on the kernel density evaluation function,the point cloud data were processed to establish a high-precision 3D model.In terms of prone autonomous flight,the altitude ratio parameter was introduced to identify obstacles.Combined with image processing and sonar feedback,constant altitude flight was achieved.Based on bounding box analysis,the heading angle was dynamically adjusted to ensure the safety and efficiency of the flight path.In the rust defect identification,the SVM algorithm was adopted to extract features and classify the pre-processed images.By normalizing the input data and optimizing the classification hyperplane,the accuracy of rust detection was improved.The experiment employed a high-resolution camera(4 096 pixels×3 072 pixels)to collect images of transmission lines.A total of 1 201 sample images were obtained and divided into the training set and the test set at a ratio of 7∶3 to verify the effectiveness of the method.[Results]The experimental results show that the proposed method demonstrates significant advantages in path planning and defect identification.The UAVs can precisely avoid randomly distributed obstacles and plan the optimal inspection path,superior to the traditional methods in both safety and efficiency.In terms of rust defect identification,the identification rate of the SVM-based model for 360 test images stably remains above 95%,which is significantly superior to the reinforcement learning method and the deep residual network method.In the rust defect detection of different areas,this method exhibits good adaptability and stability.Through the frames per second(FPS)evaluation,the real-time detection performance of this method is excellent,which meets the needs of large-scale transmission line inspections.According to the visualization results,this method can accurately mark the rusted area and effectively avoid false detection and missed detection.[Conclusions]The proposed autonomous flight inspection method of UAVs for rust defects in transmission lines,integrated with laser 3D modeling,intelligent path planning,and the optimized SVM model,excels at the accuracy of path planning,defect identification rate,and real-time capability.The experiment verifies the engineering application value of this method in increasing efficiency,accuracy,and safety of inspection.Future research can further optimize the adaptability of the algorithm in complex environments and expand its application in defect detection of other power equipment.