Network Security Protection Based on Deep Learning in Power Grid Information Construction

Abstract

Aiming at the problems that traditional network security protection methods ignore the timeliness of intrusion and information leakage, a network security protection method based on deep learning in power grid information construction is proposed. Firstly, combined with the development needs of modern power grid, the overall architecture of information power grid is constructed to achieve multi service integration. Then, it quantifies the network information risk based on the attack graph and sends it into the Transformer model for analysis to detect the type of network attack and the location of attack nodes. Finally, the terminal active immune structure of trusted computing is designed to encrypt the information and complete the optimization of power grid information leakage prevention technology. Based on the KDD '99 data set, the experimental demonstration of the proposed method is carried out. The results show that the precision, recall and F1 value of the proposed method have reached 98.031%, 96.574% and 97.293% respectively, and the number of information leakage has been significantly reduced, effectively improving the security protection capability of the power grid.