Fuel economy of Hybrid Electric Vehicles (HEVs) may be further improved by exploiting the increased connectivity level of next -generation vehicles. Minimization of HEVs fuel consumption is a global problem and its optimal solution inevitably entails the complete knowledge of the driving conditions. Hence, optimality can only be reached on a limited number of a priori known mission profiles, and never on real driving test cases . Thus, the capabilities of conventional Energy Management Systems (EMS) can be strongly enhanced by integrating the prediction of future vehicle speed into the powertrain control strategy. Vehicle-to-Everything (V2X) technology adoption paves the way for reliable future driving conditions forecasting. As a result, in this paper information derived from V2X connectivity was used to develop an innovative adaptation algorithm for an Equivalent Consumption Minimization Strategy (ECMS). Traffic information and driving style identification were employed to predict future driving conditions and, in turn, to adapt the equivalence factor. Hence, some innovative correction parameters were introduced in the equivalence factor formulation, in order to periodically adapt it according to the predicted vehicle speed. The continuous equivalence factor optimization was aimed at ensuring enhanced fuel economy and at guaranteeing charge sustainability. The potential of this innovative Adaptive ECMS (A-ECMS) was assessed on a P2 architecture test case by means of numerical simulation. The reliability of the simulation platform had been preliminarily validated by comparing simulation results with experimental data. The experimental measurements were obtained by testing a Mercedes-Benz E 300 de on real-world driving scenarios. The simulation results proved that the proposed approach is able to significantly improve the strategy adaptability and its fuel economy potential if compared with the conventional EMS taken as reference. Fuel consumption reductions up to 10 % were demonstrated, depending on the vehicle mission profile. Finally, a sensitivity analysis was performed in order to assess how different prediction horizons affect the adaptive algorithm.

Prof. Federico Millo, Politecnico di Torino, ITALY Dr. Luciano Rolando, Politecnico di Torino, ITALY Dr.-Ing. Luca Pulvirenti, Politecnico di Torino, ITALY