Grid information tampering cyber attack location method based on improved SA-RNN model

With the rapid increase in the number of intelligent terminal access in the current new power system, the possibility that th e po wer grid may be attacked by information tampering is also increasing. In order to solve the problem that the power grid is su bj ect to information tampering cyber attack, and the associated trip fault cannot be captured by the power grid condition monitor ing system, this paper first establishes the attack model of power grid information tampering attack, and then analyzes the influ ence of information tampering attack on the power grid state quantity. Then, based on the improved recurrent neural network model integrating self-attention mechanism, feature extraction is carried out on the power grid load information after informati on tampering, so as to locate the information tampering attack. In this paper, simulations are carried out on the IEEE-9 node sy stem to compare and verify the effectiveness of the proposed model for locating information tampering attacks.