Kriging and local climate zones – key to higher accuracy in range prediction?

Abstract

Range prediction of electric vehicles requires knowledge of many parameters, including information about the vehicle environment. As an alternative to onboard measurement, external data sources (secondary data) can be used. The presented methods are suitable for estimating climatic data for vehicle application including urban areas and areas of complex topography. Compared to an inverse distance weighting approach for ambient temperature values, Kriging yields better overall accuracy especially for areas of complex topography, improving accuracy in urban regions, however, remains a challenge. An approach to improve accuracy in urban areas by considering local climate zones in the selection of secondary data sources produces mixed results.