AI-Based Control Scheme for Resilient Grid-Forming Inverters Under DC Link Voltage Disturbances

For the integration of distributed energy resources (DERs) into the power grid, grid-forming inverters (GFMIs) play a key role due to their voltage support capability. However, the low-inertia inverter based generation can pose stability challenges to the power system. One of the major challenges is the dynamic behavior of GFMIs when subjected to disturbances and irregularities, including disturbances in DC link voltage. Most of the existing studies do not consider the DC link voltage variations for GFMIs, overlooking the impact of their sudden fluctuations on the system performance. This paper proposes a decentralized control approach that uses a neural network-based method to dynamically adjust the inertia and damping coefficient of the Virtual Synchronous Generator (VSG) in real time for improving the dynamic response of GFMIs to DC link voltage disturbances. The effectiveness of the proposed approach is validated through various case studies on a 9-bus, four-GFMI test system in MATLAB/Simulink. Furthermore, experimental hardware results also validate the effectiveness of the proposed approach.