Abstract:Because most coupling modeling methods of power information physical system rely on network topology to divide node types, the attributes of node information are not considered, and the influence of different node types on risk propagation is also ignored, which weakens the ability of risk prediction. Therefore, a risk propagation model of power information system considering node heterogeneity and fault function is proposed. Firstly, the characteristics and coupling characteristics of actual power network nodes are analyzed, and the characterization method of dual network nodes is proposed based on node heterogeneity. Based on the physical and logical dependence of dual network, the power CPS characterization model is designed. On this basis, considering the directivity and node dependence of dual-network state interaction, the factors affecting risk propagation were analyzed. By defining the fault probability function based on node degree, the power CPS risk propagation model was constructed. Experimental analysis was carried out with the coupling network of synthetic network and real Italian data to prove the effectiveness of the model.