Security Situation Prediction Method of Industrial Control System Based on Self-Attention and GRU Neural Network

Given the current industrial control system security situation prediction method accuracy is insufficient, the model is challenging to build. Aiming at the above problems, this paper proposes a security situation prediction method for industrial control systems based on the selfattention mechanism and GRU neural network. Firstly, the self-attention mechanism generates attention weight combined with security situation data. Secondly, The data with attention weight is input to the gated cyclic unit to mine the correlation between the safety data of the industrial control system. Finally, the trained model is used to predict the security situation of the industrial control system, and the final predicted security situation value is output. Experimental results show that the proposed method has faster convergence speed and higher accuracy than existing network security situation prediction methods. © (2023). All Rights Reserved.