In order to improve the support ability of the virtual synchronous generator (VSG) control method for the system when the new energy system is connected to the grid, an adaptive control strategy based on radial basis function (RBF) neural network is designed on the basis of the traditional VSG. Because the RBF neural network has a good approximation effect for dynamic nonlinear functions, based on the characteristics of the controlled object, the angular velocity deviation and the rate of change of the angular velocity deviation of the system are taken as inputs, The designed RBF controller achieves adaptive control of rotational inertia. Establish a mathematical model in MATLAB/Simulink and conduct comparative experiments with traditional VSG control. The results show that the RBF adaptive control strategy can effectively support the frequency of the system and suppress power fluctuations when countering interference, improving the inertia and stability of the new energy grid connected system.