Predictive Energy Management Strategy for Fully Electric Vehicles Based on Preceding Vehicle Movement

Shuwei Zhang,Yugong Luo,Junmin Wang,Xiao Wang,Keqiang Li

DOI: https://doi.org/10.1109/tits.2017.2672542

IF: 8.5

2017-01-01

IEEE Transactions on Intelligent Transportation Systems

Abstract:This paper presents an energy-efficient and terrain-information-and-preceding-vehicle-information-incorporated energy management strategy for fully electric vehicles (FEVs) equipped with in-wheel motors. Saving driving energy with terrain preview and preceding vehicle movement prediction are crucial to prolong the driving distance for an FEV. Unlike conducting energy optimization under the assumption that the preceding vehicle movements are already known in most studies, the front vehicle movements are predicted during each control cycle based on the vehicle-to-vehicle communication, and the FEV vehicle velocity and motor torque distribution are optimized by a nonlinear model predictive controller to reduce energy consumption. The energy-saving objective is achieved by including, in the cost function, the motor energy consumption in each control cycle, while the safety objective is accomplished by keeping a suitable relative distance from the preceding vehicle. Since the nonlinear vehicle longitudinal model is applied, the gridding initial torque plane is utilized in each time step to search for the global minimum. Simulation results show that this method has a better energy-saving performance than the control method without using the preceding vehicle movement information, and the algorithm proposed here has a wide applicability under various driving conditions.