In order to improve the vehicle-to-grid (V2G) stability and enhance the friendly interaction between electric vehicle and power grid, a new type of V2G charging and discharging model for electric vehicle is designed. An electric vehicle V2G charging and discharging control strategy based on virtual machine is proposed. The DC interface of electric vehicle is controlled by the virtual DC machine. The AC interface of grid connection is controlled by the virtual synchronous machine. Compared with the traditional V2G structure, an isolated bidirectional DC/DC converter is used to connect the two side interfaces. On this basis, for the setting of virtual DC machine parameters, the damping coefficient is determined according to the initial state of charge (SOC) of each electric vehicle to determine the power distribution of electric vehicle group. The virtual inertia is adjusted in real time according to the change of SOC to improve the ability of electric vehicle to participate in grid frequency regulation. The simulation test of electric vehicles with different initial SOCs is carried out under different grid conditions. The simulation results show that the strategy can effectively coordinate the two-way power flow between electric vehicle and power grid while taking into account the needs of electric vehicle users, and reduce the harmonic current of the grid-connected point, thus maintaining stable operation of power grid.