Real-driving-based comparison of the eco-impact of powertrain concepts using a data-driven optimization environment

Abstract

In order to limit the effects of man-made climate change, the assessment of the ecological impact of different powertrain concepts is of increasing relevance and intensely studied. In this contribution we present a data-driven optimization environment that enables to identify the ecological potential of different concepts for different scenarios. The parametrization of each powertrain concept is dedicatedly optimized to minimize the ecological impact, which allows for an unbiased and reliable comparison on an uniform evaluation basis. To exploit the potential of each single powertrain parametrization, the operating strategy of the powertrain is adapted. Naturalistic driving profiles, including the speed, acceleration and road-slope information are depicted by multidimensional and representative driving cycles, allowing for an efficient search of the real-driving-optimal powertrain parametrizations within the optimization. In this study, we investigate long-range capable vehicles for a scenario in the reference year 2030 in Germany. Conventional vehicles, battery electric vehicles, fuel cell electric vehicles and plug-in hybrid electric vehicles are examined. Finally, the results are compared to an evaluation of the CO2 emissions according to the Worldwide harmonized Light vehicles Test Procedure (WLTP).