Abstract: Taking advantage of the characteristics of online rolling optimization to obtain the optimal control sequence, an engine shutdown optimization control strategy was proposed based on model predictive control (MPC) to reduce the vehicle longitudinal impact in the process of engine shutdown when switching the series-parallel hybrid electric vehicle (HEV) from the hybrid drive mode to the electric drive mode. Firstly, combining theory with experiment, an engine resistance torque model was established. Secondly, according to the different working states of the power components, the shutdown process was divided into two stages, the engine working point adjustment stage and the motor reverse dragging engine stage. And a segmented engine shutdown control strategy was designed. Finally, the designed control strategy was verified in simulation, and compared with the traditional PID control method. The simulation results show that the proposed control strategy can effectively suppress the output torque fluctuation during engine shutdown, reduce the vehicle impact and effectively improve the ride comfort of the vehicle.