Adaptive Inertia Control Strategy for Virtual Synchronous Generators Based on RBF Neural Networks

An adaptive virtual synchronous generator control (VSG) algorithm is proposed in this paper which can adaptively adjust the virtual parameters of the system, thereby exhibiting good stability under different interference conditions. Compared to ordinary synchronous generators, VSGs offer more flexible parameter adjustability, with virtual inertia and virtual damping having significant impacts on system stability. Leveraging the RBF neural network's excellent approximation capabilities, simple algorithm, and fast learning characteristics, an improved RBF neural network was designed to meet real-time control requirements and was used for online adjustment of the VSG's virtual inertia. The virtual damping was adaptively regulated based on a fixed damping ratio. A photovoltaic hybrid energy storage system was constructed in Slmulink, integrating the neural network algorithm to establish an adaptive simulation model, thereby verifying the effectiveness of this control strategy. As the results shows that adaptive control strategy can effectively enhances the frequency stability of the VSG, providing new insights into frequency control for power system.