Online Neuro-Adaptive Learning For Power System Dynamic State Estimation

With the increased penetration of Distributed Generators (DGs) in the contemporary Power System, having knowledge of rapid Real-Time electro-mechanical Dynamic States has become crucial to ensure the safety and reliability of the grid. In the conventional SCADA based Dynamic State Estimation (DSE) speed was limited by the slow sampling rates (2-4 Hz) so State Estimation was limited to static states such as Voltage and Angle at the buses. Fortunately, with the advent of PMU based synchro-phasor technology in tandem with WAMS, it has become possible to avail rapid real time measurements at the network nodes. In this paper, we have proposed a novel Machine Learning (Artificial Intelligence) based Real-Time Neuroadaptive Algorithm for rotor angle and angular frequency estimation of synchronous generators, here proposed algorithm is based on reinforcement learning and adapts in real-time to achieve accurate state estimates. Applicability and accuracy of the proposed method is demonstrated under the influence of noise and faulty conditions. Simulation is carried out on Multi-Machine scenario (68 bus 16 generator NETS-NYPS model).