Predictive Energy Management Strategies for Hybrid Electric Vehicles: Fuel Economy Improvement and Battery Capacity Sensitivity Analysis

This paper shows the influence of different battery charge management strategies on the fuel economy of a hybrid parallel axle-split vehicle in a real driving scenario, for a vehicle control system that has the additional possibility to split the torque between front and rear axles. The first section regards the validation of a self-developed Model in the Loop (MiL) environment of a P1-P4 plug-in hybrid electric car, using experimental data of a New European Driving Cycle test. In its original version, which is implemented on-board the vehicle, the energy management supervisor implements a heuristic, or rule-based, Energy Management Strategy (EMS). During this project, a different EMS has been developed, consisting of a sub-optimal control scheme called Equivalent Consumption Minimization Strategy (ECMS), explained in detail in the second section. After that, the focus is on the evaluation of the benefits coming from different battery charge management strategies, which can be charge-sustaining, charge-depleting/charge-sustaining or charge blended, since the vehicle is a PHEV. The fuel economy improvements, using each strategy, are compared and one of them is then combined with the knowledge of future driving conditions (the so-called electronic horizon), mainly speed and altitude profiles. Therefore, the proposed controller would be ready for on-board implementation. In the last section, a sensitivity analysis that relates the results obtained with the battery capacity is carried out, to evaluate the influence of this strategic parameter on the battery charge management strategy choice. The paper shows the fuel economy potential of a physics-based approach like ECMS for a plug-in HEV, and how it can directly benefit from the prediction of future driving conditions, especially if the battery capacity is limited.