This paper proposes a weak point identification method for power communication network based on multi-modal deep learning, which aims to solve the problems of security vulnerabilities, network failures and insufficient capacity in the power communication network architecture. The traditional detection method of power communication network mainly relies on manual inspection and single data source collection, which has the problems of low efficiency, high cost and potential safety hazards. The method in this paper realizes the comprehensive identification and prediction of the weak points of the power communication network through multi-modal data acquisition, feature extraction, deep learning model training, adaptive learning and optimization, and visual interactive interface. By integrating multiple data sources such as network management system, traffic monitoring probes, current sensors and voltage sensors, this method can collect and analyze various data of power communication network in real time and accurately. Deep learning model training combined with support vector machine (SVM) and artificial neural network (ANN) algorithms can effectively identify abnormal behaviors in power communication networks, and continuously improve model performance through adaptive learning and optimization. Finally, by designing an intuitive visual interactive interface, operation and maintenance personnel can find and solve potential problems in the power communication network in time, and improve the security, stability and reliability of the system. The experimental results show that the proposed method performs well in the accuracy and real-time performance of weak point identification, and significantly improves the operation efficiency and reliability of power communication system[1].
Research on the application of multi-modal deep learning anomaly detection in power transmission and distribution equipment
2024-10-23
627324 byte
Conference paper
Electronic Resource
English
Improving multi-modal data fusion by anomaly detection
British Library Online Contents | 2015
|SADDLE: Spacecraft Anomaly Detection using Deep Learning
IEEE | 2024
|Rail surface anomaly detection based on deep learning
British Library Conference Proceedings | 2023
|