The Application of Machine Learning in Parameter Adaptive Control of Virtual Synchronous Generators

Virtual Synchronous Generator (VSG) enhances system stability by simulating the inertia and damping characteristics of synchronous generators.However, due to the fixed parameters, it is prone to cause low-frequency oscillations.To address this issue, scholars at home and abroad have proposed stability analysis methods based on small-signal models and introduced recurrent neural networks (RNN) and other neural network technologies into VSG control to achieve online adaptive parameter adjustment.This paper reviews the application of machine learning methods in the adaptive control of VSG parameters.Firstly, the classification and principles of machine learning were introduced.Then, the applications of small-signal modeling methods, RNN and other neural network algorithms in the VSG system were respectively presented, and specific cases and comparative analyses were given.Finally, the main challenges and research trends currently faced are discussed to provide reference opinions for the further development of related fields.Therefore, in-depth research on the application of machine learning in the adaptive control of VSG is of great significance for enhancing the stability and operational efficiency of power systems with a high proportion of new energy.