Answers are grounded in this paper and cite their sources.
This paper investigates the energy consequences of determining the energy-optimal velocity profile and torque distribution sequentially versus jointly in a battery electric vehicle (BEV) with two electric motors, one per axle. Three optimization architectures are evaluated: a centralized architecture (CA), a de-centralized architecture (DCA) and a refined de-centralized architecture (r-DCA). CA jointly optimizes the velocity trajectory and torque distribution for minimal energy consumption in a predictive framework, while DCA solves these subproblems hierarchically: velocity trajectory optimization is performed predictively, and torque distribution is computed instantaneously. The joint optimization in CA leads to a reduction in energy consumption of 3.3% at low velocities and 2.2% in an urban city cycle compared to DCA. To mitigate the energy consequences, the objective function in the predictive layer of DCA is augmented with an aggregated power loss map of the powertrain in r-DCA, which achieves energy savings close to CA. • Decoupling the optimization of velocity trajectory and torque distribution results in higher energy consumption compared to the joint optimization, where the increase is more pronounced at lower velocities. • The energy consumption of the decoupled problem is reduced by augmenting the OCP with an aggregated powertrain power loss map. • The aggregated power loss map, defined in terms of vehicle longitudinal force and forward velocity, is generated using the power-optimal torque distribution.
No takes yet. Share an insight, caveat, or question.
Torinsson et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: