Abstract: The virtual synchronous generator（VSG） control strategy enables the power electronic converter to have the moment of rotational inertia and damping coefficient of the synchronous generator, but the relationship between the two parameters and the frequency in the regulation process is a nonlinear function, which is regarded as a linear relationship in the traditional methods. The control strategy can only roughly adjust two parameters and the adjustment frequency is too high. Therefore, this paper proposes an adaptive control strategy of inertia and damping for VSG, which is based on Radial Basis Function（RBF）, using the artificial intelligence algorithm to improve the control strategy. Firstly, this paper analyzes the control methods of moment of inertia and damping coefficient from the perspectives of VSG mathematical model, output characteristics and small signal model, and gives the value ranges of corresponding parameters. Secondly, according to the unique nonlinear relationship of VSG, a double-input and double-output RBF neural network control strategy is established. Finally, the transient response of the traditional control strategy and the control strategy proposed in this paper are compared by Matlab/Simulink simulation to verify the effectiveness of the proposed control strategy.