Receding horizon control for mode selection and powertrain control of a multi-mode hybrid electric vehicle

Multi-mode Hybrid Electric Vehicles (HEVs) have shown the advantages over traditional single-mode HEVs. For multi-mode HEVs, selecting an appropriate vehicle mode and determining power- split are keys for the reduction of energy consumption. Scholars have used Dynamic Programming (DP) to determine mode selection and power-split offline due to the long computation time. This paper presents a DP-based Receding Horizon Control (RHC) method for mode selection and power- split. The DP-based RHC utilizes the Dynamic Programming algorithm to create an optimal solution within a moving prediction horizon, which has the potential for real-time application. The advantage of this DP-based receding horizon control is that it is able to handle discrete control problems. To optimize the mode selection and power-split, the experimentally determined mode shift fuel and electricity costs are incorporated in the algorithm. The simulations have been performed for DP and RHC with different sizes of prediction horizon. The control performance, fuel economy and computation time for different simulation scenarios are discussed and compared.