Fault Set Modeling on the Cyber-Side of the Automatic Generation Control Process

In response to China's accelerating "dual carbon" objectives and the rapid construction of new power system, the power grid's structural complexity is increasing, thereby imposing more stringent robustness requirements on Automatic Generation Control (AGC) software. The AGC process exemplifies a cyber-physical system (CPS) within the power domain. Currently, the testing and validation of the cyber-side of AGC software are often limited by incomplete scenario design and insufficient test case coverage, which hinders the comprehensive identification of latent fault risks within the information system. To enhance the systematic rigor and efficacy of AGC software validation, this paper applies cyber-physical system modeling and simulation theory. We propose novel quantitative metrics for the criticality of AGC information components, defined by generalized in-degree and generalized out-degree, and introduce a method to quantify inter-component correlation using path collinearity. Based on these metrics, we develop an N-k fault component combination strategy for the information system. A case study is presented to validate the feasibility and effectiveness of the proposed methodology, offering both theoretical support and technical assurance for the secure operation of the power system and the stable regulation of the new power system.