An ADP-Based Optimal Control Strategy for Virtual Synchronous Generators towards "Dual Carbon" Goals

In the context of achieving "dual carbon" goals, enhancing grid frequency stability is crucial to ensuring the efficient utilization of clean energy. This paper proposes an event-triggered optimal control strategy for Virtual Synchronous Generators (VSG) based on adaptive dynamic programming (ADP), aimed at improving frequency stability in AC microgrids to facilitate the integration and optimal use of renewable energy. Unlike traditional model-dependent algorithms, this approach utilizes a model-free algorithm that achieves optimal frequency control through event triggering, thereby reducing communication and computational burdens. The frequency dynamics are modeled as nonlinear state-space equations, with the inverse of inertia as the control input. An optimized cost function that considers angle deviation, frequency deviation, Rate of Change of Frequency (RoCoF), and control cost is used to effectively balance energy control and frequency deviation. Simulation results demonstrate that the proposed optimal control strategy significantly improves frequency response under load disturbances while maintaining effective DC-side energy management, providing technical support for green, low-carbon grid operation.