An efficient vehicle-following predictive energy management strategy for PHEV based on improved sequential quadratic programming algorithm

For the vehicle-following scenario, control design of plug-in hybrid electric vehicle (PHEV) needs to care about not only the efficient energy conversion, but also the driving safety by keeping an appropriate distance. Thus, how to obtain the optimal fuel economy under the premise of maintaining a safe following distance, is a challenging and hot issue for researchers, especially in the background of autonomous driving. Aiming at above problem, this paper proposes an efficient vehicle-following energy management strategy (EMS) for PHEVs based on model prediction control (MPC). In this strategy, the values of powertrain torque and vehicle speed are predicted in the given prediction horizon, and an improved sequential quadratic programming (ISQP) algorithm is proposed to solve the receding horizon optimization problem. The real-time efficiency of engine and electric motor are estimated through the calculation from last moment. The proposed EMS is verified by using the parameters of a real-world cargo truck equipped with parallel hybrid powertrain. The results show that the proposed strategy can ensure the vehicle driving safety while obtaining excellent fuel economy. Finally, the real-time capability of proposed strategy is verified in hardware-in-loop test environment.