Advanced Inner Loop Control of Virtual Synchronous Generators Using Adaptive Neurofuzzy Inference System for Robust Grid Integration

As renewable energy integration increases, maintaining grid stability becomes increasingly challenging due to the absence of inertia in inverter-based systems. The Virtual Synchronous Generator (VSG) offers a solution by mimicking the inertia and damping of conventional generators. However, its performance relies heavily on inner loop control, which must regulate current and voltage precisely. Traditional PI controllers, while simple, are limited in managing nonlinear and time-varying system dynamics. This study introduces an advanced inner loop control strategy for VSG using the Adaptive Neuro-Fuzzy Inference System (ANFIS). By combining neural network learning with fuzzy logic reasoning, ANFIS enables real-time adjustment of control parameters, improving system response and regulation. Simulation results show that the ANFIS-based controller outperforms the conventional PI controller in terms of transient response, robustness, and power-sharing accuracy. These findings suggest that ANFIS provides a more intelligent and adaptive control solution for reliable integration of renewable energy sources into the power grid.