The Monitoring Method for Transmission Lines Based on Deep Learning

The surrounding environment of the transmission line is complicated, and large-scale construction machinery pose threats to the circuit facilities and wires. When anomaly occurs, too much time is required for the State Grid Corporation to fix it manually. To reduce the inspection burden, many methods have been presented in the past to diagnose and locate anomaly. In this paper, we propose a detection method for cranes, tower cranes, construction machinery based on the YOLOv3 model. In the aspect of feature extraction, Mobilenetv2 is used as the backbone network for feature extraction of abnormalities in transmission lines, and in the aspect of the recognition algorithm, the improved YOLOv3 algorithm is used as the head network for target detection based on the features extracted from the backbone network. Compared with the existing algorithms, this algorithm achieves a significant reduction in the amounts of parameters and calculations while ensuring the accuracy of the model. The results show that the proposed method can effectively and accurately realize the real-time detection of cranes, tower cranes and construction machinery in transmission lines.