Cellular Computational Networks Based Hierarchical Data-Driven Dynamic State Estimation Method Considering Uncertainties

Accurate generator information is crucial for the efficient control and operation of a power system. This study proposes a hierarchical data-driven approach for dynamic state estimation (DSE) of generators using cellular computational networks (CCNs) structure. The proposed method initially divides the problem of dynamic state estimation into multiple layers through hierarchical architecture. In the prediction layer, CCNs are employed to reduce the system scale by considering only relevant generators. In the correction layer, a novel adaptive filter is utilized to increase data abundance. Simulation results demonstrate that the proposed hierarchical data-driven method can accurately estimate states using PMU data alone while maintaining high computational efficiency. Additionally, it offers easy scalability and strong robustness against uncertainties. The proposed method has potential applications in online dynamic state estimation and real-time security monitoring.