ANN-based Dynamic Frequency Regulation of PV-based Hybrid Microgrid system

The high penetration of inverter-based renewable energy resources, such as Photovoltaic (PV) systems in modern power systems, often leads to reduced system inertia which makes efficient frequency control highly imperative. This research develops a novel method of frequency regulation using PV system. The proposed method which uses the conventional swing equation to generate the needed virtual inertia (VI) based on the frequency dynamics of the system and then feeds an already trained Artificial Neural Network (ANN) to determine the corresponding operating point for the PV. A Proportional-Integral (PI) controller is then used to drive the operating point of PV towards the ANN-generated voltage reference. The shifting of the operating point to the desired voltage value will adjust the active power injection from PV in order to provide the desired frequency regulation. Simulation results show that the proposed control mechanism provides high accuracy, a fast-tracking speed, and has the potential to significantly improve stability during frequency disturbance in PV-based power systems.