Design Optimization and Optimal Control for Hybrid Vehicles

International audienceGrowing environmental and global crude oil supplies concerns are stimulating research on new vehicle technologies. Hybrid-electric vehicles appear to be one of the most promising technologies for reducing fuel consumption and pollutant emissions. Different types of hybrid-electric powertrains exist: from the mild-hybrid vehicle, equipped with a small electric motor, to the combined hybrid like the Toyota prius. This paper presents a parametric study focused on variations of the size of the powertrain components , and optimization of the power split between the engine and electric motor with respect to fuel consumption. To perform this optimization on a prescribed driving cycle (for instance, the New European Driving Cycle), a dynamic programming algorithm based on a reduced model is implemented. This simplified model allows a fast optimization with a fine parameterization of the controller: it furnishes the optimal power repartition at each time step regarding fuel consumption under constraints on the battery state of charge. The obtained results may be used to determine the best components of a given powertrain, for a prescribed vehicle cycle. The optimal split obtained thanks to dynamic programming algorithm can not be used directly on a vehicle as a real time control law, as the future can not be known in advance in normal driving conditions. To overcome this difficulty, we implement, as a real-time strategy, the Equivalent Consumption Minimization Strategy (ECMS): the battery being considered as an auxiliary reversible fuel reservoir, an instantaneous minimization of ECMS is performed. This control law is inferred from Pontryagin's Minimum Principle, where the Lagrange multiplier can be deduced from previous optimization results on given driving cycles. Offline optimization results and real-time control laws are compared for a realistic hybrid vehicle application. 2