Battery-powered electric mobility is currently the most promising technology for the decarbonisation of the transport sector, alongside hydrogen-powered vehicles, provided that the electricity used comes 100% from renewable energy sources. To estimate its electricity demand both nationwide and in individual smaller communities, a calculation based assessment on driving profiles that are as realistic as possible is required. The developed model based analysis presented in this paper for the creation of driving and thus electricity load profiles makes it possible to build different compositions of driving profiles. The focus of this paper lies in the analysis of motorised private transport, which makes it possible to assess future charging and load control potentials in a subsequent analysis. We outline the differences in demand and driving profiles for weekdays as well as for Saturdays, Sundays and holidays in general. Furthermore, the modelling considers the length distribution of the individual trips per trip purpose and different start times. The developed method allows to create individual driving and electric vehicle (EV) demand profiles as well as averaged driving profiles, which can then be scaled up and analysed for an entire country.
Multifeature pool importance fusion based GBDT (MPIF-GBDT) for short-term electricity load prediction
