With the wide application of electric vehicles and the rapid development of smart Grid, Vehicle to Grid interaction (V2G) technology has become a research hotspot. In order to improve the charge and discharge quality of electric vehicle and suppress the current fluctuation, a finite set model predictive control strategy for electric vehicle charge and discharge system was designed. Firstly, the discrete mathematical model of the converter in synchronous rotating coordinate system is established. Then, the current value of the AC side at the next time is predicted according to the current sampling value, and the current is predicted in two steps based on the conventional one-step prediction. Finally, the value function is constructed to select the optimal voltage vector and applied to the PWM rectifier to achieve the optimal control effect. Simulink was used to build the whole simulation model of V2G system for analysis. The results showed that both PI and FCS-MPC control realized reversible charge and discharge function of electric vehicles, but the model predicted current control more effectively improved the current tracking accuracy and reduced the harmonic distortion rate of grid-connected current.