Optimal energy management of hybrid electric vehicles considering pollutant emissions during transient operations

Abstract

The reduction of fossil-fuel vehicles’ pollutant emissions has been improved with exhaust aftertreatment systems and, more recently, with electric hybridization. Indeed, hybrid electric vehicles (HEVs) can shift engine operating points toward low fuel consumption and pollutant emissions regions. Also, engine transient operations have already been shown to impact pollutant emissions negatively. As a result, the electric motor can be additionally used to reduce the engine transient operations. The engine optimal control, i.e., minimizing fuel consumption and pollutant emissions, must consequently consider the engine transient dynamics. Therefore, this paper introduces a parametric approximation of the optimal engine control variable based on a weighted sum of smooth basis functions to directly consider transient engine dynamics and guarantee smooth engine operations. Two approaches are presented and compared to find the parametric approximation of the engine control, minimizing fuel consumption and pollutant emissions. The proposed direct and indirect approaches rely on polynomial approximations of the vehicle components’ models, leading to efficient quadratic programming algorithms. In this paper, a high-fidelity simulation platform of a heavy-duty HEV is employed to calibrate the controller model and then compare it to classical control methods. The results from the proposed minimization approaches are shown to be close to the dynamic programming optimality, yet being much faster alternatives.