Power distribution equipment and defect identification technology based on deep learning

Abstract

In order to realize the collection of coordinate position of line pole and tower, and realize the field data collection of distribution equipment UAV and real-time recognition of front-end image, a calculation method of front-end edge of distribution equipment based on front-end RFID and depth identification was proposed. Firstly, the RFID identification instruction is sent to the front-end equipment for scanning through the interface provided by the RFID equipment, so as to obtain the physical ID number of the tower. Then, the edge computing equipment is used to carry out the equipment identification on the photos, so as to obtain the key equipment on the column. Fine photos were taken of the distribution network poles and towers to obtain the fine photo result data. Through the coordinate of the poles and towers, the photo coordinate associated with the equipment on the column and the collection coordinate of the physical ID, the trinity correlation was carried out to complete the AUTOMATIC GIS modelling of the distribution line and realize the geographical map display of the distribution line. This method can effectively identify the coordinate of power distribution line tower and equipment, and provide necessary data support for the modelling of power distribution line.