Combined Optimization of Energy and Battery Thermal Management Control for a Plug-in HEV 2019-24-0249

This paper presents an optimization algorithm, based on discrete dynamic programming, that aims to find the optimal control inputs both for energy and thermal management control strategies of a Plug-in Hybrid Electric Vehicle, in order to minimize the energy consumption over a given driving mission. The chosen vehicle has a complex P1-P4 architecture, with two electrical machines on the front axle and an additional one directly coupled with the engine, on the rear axle. In the first section, the algorithm structure is presented, including the cost-function definition, the disturbances, the state variables and the control variables chosen for the optimal control problem formulation. The second section reports the simplified quasi-static analytical model of the powertrain, which has been used for backward optimization. For this purpose, only the vehicle longitudinal dynamics have been considered. The third section describes the Model-in-the-Loop environment of the vehicle, implemented in Simulink. In particular, the validation of the fuel consumption and the battery temperature models against experimental data is shown, and the original control strategies for the energy and thermal management are described, as well. This powertrain model is used to evaluate vehicle performance. As the powertrain architecture offers different torque split possibilities, different approaches to the powertrain control are considered, starting from the baseline rule-based controllers for both the thermal and energy management, to the combined-optimization based controllers. This paper shows a consistent fuel economy improvement due to energy management optimization, which becomes even larger if thermal management is included in the optimization algorithm.