To enhance frequency and active power control performance, this research proposes a decentralized approach for power grid frequency regulation support using grid-forming (GFM) converters. The proposed approach employs the Kharitonov method to design a robust proportional-integral (PI) controller in the outer control loop of a grid-connected GFM converter. The PI controller is made expressly to deal with the uncertainties and disruptions that are common in power grid operations. The PI controller successfully mitigates system risks by demonstrating strong stability through the application of the Kharitonov approach. Additionally, depending on the expected system performance, this work uses Bayesian optimization to choose the optimal parameters from the robust Kharitonov solutions family. In decentralized frequency regulation, the Kharitonov approach combined with Bayesian optimization presents a viable framework for robust-optimal control system parameters tuning. The efficacy of the proposed method in attaining robust optimal frequency regulation support in a power grid with a grid-connected GFM converter is demonstrated and the result is examined in the simulation environment.
