Evaluation Study of a Multi-Mode HEV with a Dedicated Hybrid Transmission

Abstract

The current legislation requires the increasing levels of vehicle powertrain electrification or hybridization to fulfill the limits of green-house gas emissions. Parallel hybrid electric vehicle (HEV) powertrain topologies are among the frequently used layouts, because of their easy applicability on an existing conventional powertrain by the addition of hybrid modules with mild, full, or plug-in capability. A more “HEV-tailored” approach merges a dedicated hybrid transmission (DHT), an ICE, a gearbox, and one or more electric motors more closely together, reducing HEV powertrain’s mechanical complexity and costs. The current paper investigates a multi-mode HEV with a DHT and two electric machines in optional electric serial or parallel operating modes. Great challenge for the multi-mode HEV topology with DHT is the component sizing: ICE, electric machines, and battery with optimized capacity, and yet fulfilling several roles. On one hand, satisfactory responsiveness on a dynamic driving demands (vehicle acceleration capabilities, top speed etc.). On the other hand, energy consumption achieving better levels than standard parallel HEV topologies. The other challenge for a multi-mode HEV powertrain – similarly as for all HEV powertrains – is a development and optimization of a powertrain supervisory control, based on energy management strategy. These challenges – overall DHT layout, component sizing, and energy management – closely interact and, therefore, must be optimized together. The paper addresses all these challenges using multi-parametric optimization toolchain, that combines a parametric HEV model, energy management strategy, and optimization software. The optimization software includes various optimization strategies: genetic algorithms, gradient based methods, or their combinations. The parametric HEV model is built in GT-Suite 0D/1D/3D multi-physics CAE system simulation software, that also includes different energy management strategies, either locally or globally optimal, considering the whole driving cycle. The toolchain is then used to evaluate the optimal component sizing for two different vehicle classes and fuel economy in a homologation driving cycle and in some real-world driving scenarios. All results are compared to in-house benchmark method.