Abstract:To improve the accuracy and efficiency of power system network security risk assessment, this paper proposes a new method for power system network security risk assessment based on RBF neural network. Firstly, we need to establish a safety risk assessment index system and combine it with the risk assessment matrix. Next, by calculating the information entropy of the indicators, the risk level of the evaluation indicators is determined and divided into five risk levels. Finally, by calculating the fused feature values of network security risk distribution, we will input the fused feature values of security risk distribution and the risk values of security risk evaluation indicators. Based on the results of power system network security risk assessment, we will use RBF neural network to construct a power system network security risk assessment model. Thus obtaining the results of network security assessment. Experimental results have shown that this method has an accuracy of 96% in the highest security risk assessment, while achieving an evaluation efficiency of 98%. Therefore, this method has achieved the expected goal of researching power system network security risk assessment methods.