Approximate Power Flow Solutions-Based Forecasting-Aided State Estimation for Power Distribution Networks

Emerging forecasting-aided state estimation (FASE) frequently encounters complicated parameter analysis and observation calculation tasks, especially when confronted with intricate and uncertain scenarios. To this end, a concise FASE estimator is developed by combining the precise depiction of dynamic state change and linear power flow approximation. Designing the dynamic system state as a voltage perturbation vector around the nominal value, the forecasted state is firstly derived from the linear approximation of power injection equation solutions. The state forecasting model relies solely on nodal impedance information as the state transition matrix, eliminating the onerous parameter tuning effort. After that, the optimal filtered state is efficiently obtained utilizing line power flow measurements, with branch admittance information to construct the approximate observation matrix. Numerical simulation comparisons on a symmetric balanced 56-node distribution system verify the performance of the proposed estimator in terms of accuracy and robustness.