Virtual synchronous generator (VSG)-controlled energy storage can enhance the inertia of power systems which have a high proportion of renewable energy and power electronic devices, and suppress frequency fluctuation during disturbance. Model predictive control (MPC) can be used to adaptively control the input power of VSG-controlled energy storage to improve the frequency response further. By establishing the prediction model of VSG-controlled energy storage, the cost function of the weighted sum of frequency and input power is designed. The optimal control sequence is computed by quadratic programming to modify the VSG input power. The results show that this method can effectively increase system inertia, suppress the maximum frequency deviation and rate of change of frequency during a load disturbance. The rate of change of frequency is effectively improved compared with existing control methods. This work is supported by the National Natural Science Foundation of China (No. 51821005).