MPC-based energy management with short-term driving condition prediction for a plug-in hybrid electric truck

This study presents a short-term predicted energy management strategy (EMS) for a plug-in hybrid electric truck (PHET). Within the framework of model predictive control (MPC), an adaptive equivalent consumption minimization strategy (ECMS) is proposed according to future driving conditions, including velocity and road grade information. Firstly, using minimum historical driving data, vehicle velocity, and road grade are predicted simultaneously through a discrete grey forecasting model (DGM) without the intelligent transportation system and extra devices, and the accuracy of the prediction models is verified. Unlike ordinary MPC-based EMS, the effect of the road grade variation on future power demand is taken into consideration, especially during downhill roads. Then, for rational and efficient utilization of the recuperated energy, the control parameter of ECMS, i.e. equivalent factor (EF), is adjusted adaptively according to not only the SOC reference but also the termination state constraints in each preview horizon. The proposed strategy (ECMS-MPC) aims at improving the energy utilization efficiency of driving on varying gradients roads, resulting in a better performance in fuel economy and calculation efficiency. A comparison of results shows that the fuel economy of the ECMS-MPC is better than that of DP-based MPC and close to offline global optimization strategies, and it costs half the time of DP-based MPC. Furthermore, the influence of EF adaptation and preview horizon on control performance are analyzed. This article also contributes to the development of sustainable energy technologies.