Power Intelligent Terminal Intrusion Detection Based on Deep Learning and Cloud Computing

Abstract

Numerous internal and external intrusion attacks have appeared one after another, which has become a major problem affecting the normal operation of the power system. The power system is the infrastructure of the national economy, ensuring that the information security of its network not only is an aspect of computer information security but also must consider high-standard security requirements. This paper analyzes the intrusion threat brought by the power information network and conducts in-depth research and investigation combined with the intrusion detection technology of the power information network. It analyzes the structure of the power knowledge network and cloud computing through deep learning-based methods and provides a network interference detection model. The model combines the methods of abuse detection and anomaly detection, which solves the problem that the abuse analysis model does not detect new attack variants. At the same time, for big data network data retrieval, it retrieves and analyzes data flow quickly and accurately with the help of deep learning of data components. It uses a fuzzy integral method to optimize the accuracy of power information network intrusion prediction, and the accuracy reaches 98.11%, with an increase of 0.6%.