Gradient descent based dynamic optimization for VSG dominated microgrid

Abstract As the integration of distributed generation units into the power grid continues to grow, the system becomes increasingly vulnerable to power fluctuations and potential failures, which significantly challenge grid stability. Virtual synchronous generators (VSGs) have emerged as a key technology to enhance grid stability by emulating the behavior of traditional synchronous generators. However, with multiple VSGs operating within independent microgrids, coordinating their control parameters becomes critical to ensuring stable and reliable operation. This article proposes an optimization method based on the gradient descent algorithm to fine‐tune the control parameters of multiple VSGs in independent microgrids. A small‐signal model for microgrids with multiple VSGs is developed to analyze the system's dynamic behavior under small disturbances. The proposed method optimizes VSG control parameters to improve the frequency and voltage dynamic response and enhance the overall system stability. The simulation results demonstrate that the proposed multi‐VSG control parameter optimization method significantly enhances the frequency and voltage dynamic response, as well as the stability of the independent microgrid under small disturbances, offering a practical solution for improving microgrid performance in real‐world applications.