An Approach for Rapid Online Robustness Enhancement of Distributed Power Systems using Load Classifier Forecasting

In this paper, we propose an online approach to rapidly and efficiently improve the robustness of distributed power systems (DPSs). Based on real-time monitoring of sudden voltage sag the main work includes the following two steps: offline classifier identification and online instantaneous disturbance suppression. In order to accurately forecast load classifiers, we adopt an automatic analytical solution, Wu's Elimination Method (WEM), to derive the expressions of differential equations that are used to describe the phenomena in DPS. Then the corresponding load classifiers have been obtained by BCU method (Boundary of stability region based Controlling Unstable equilibrium point method). The optimum switching time of the proposed Active Damping Generator can be chosen based on combining the obtained classifiers with digital signal processing. Simulation and experimental results show that through our work the suddenly changed voltage and distorted current of the DPS can be quickly recovered within 15ms which is faster than the standard IEC TS62749-2015.