The Multisource Information Fusion and Intelligent Diagnosis Algorithm in an Integrated Air-Ground-Space Power Transmission Line Monitoring System

Aiming at the shortcomings of traditional monitoring methods in monitoring scope, real-time and accuracy, this paper proposes an Air-Ground-Space monitoring system which integrates satellite remote sensing, unmanned aerial vehicle inspection and ground sensors. The system realizes all-round and multi-level monitoring of transmission lines through Multisource Information Fusion (MIF) technology, which significantly improves the comprehensiveness and accuracy of monitoring data. In this paper, a multi-modal fusion algorithm based on deep learning is designed, which can integrate data from different dimensions and modes, and extract more comprehensive and accurate transmission line state information through feature extraction and fusion module. The algorithm designs feature extraction networks on satellite remote sensing data, UAV inspection images and ground sensor data respectively, and introduces attention mechanism for adaptive fusion of cross-modal information. The experimental results show that the multi-modal fusion algorithm is significantly superior to a single data source in monitoring accuracy, with an average monitoring accuracy of 94.74%. Support vector machine (SVM) is used as the classification algorithm to classify the fused multi-source information, and the accurate identification of transmission line faults is realized. The experimental data show that the algorithm has high accuracy and recall in identifying broken wires, damaged insulators and icing faults, which is significantly superior to the traditional machine learning algorithm. In addition, the hardware and software platform of the Air-Ground-Space transmission line monitoring system is built, and the system integration and performance evaluation are carried out. The experimental results show that the system can receive and process information from different data sources in real time, and give diagnosis results quickly, which meets the requirements of real-time monitoring. The system has obvious advantages in real-time, stability and intelligence, which provides a strong guarantee for the safe operation of the power grid.