A Cross-Country Model for End-Use Specific Aggregated Household Load Profiles

End-use specific residential electricity load profiles are of interest for energy system modelling that requires future curves or demand-side management. We present a model is applicable across countries to predict consumption on regional and national scale, using openly available data. The uses neural networks (NNs) correlate measured from one country (United Kingdom) with weather data daily mix human activity device power profiles. then use region-specific time-use surveys as input the trained NNs unscaled electric total profile consists end-use household scaled real consumption. compare model’s results independently simulated various European countries. achieve mean absolute error compared average 6.5 33% test set. For Germany, standard deviation between simulation, H0, measurements University Applied Sciences Berlin 26.5%. Our approach reduces amount required existing models considering end-uses seasonality based parameters. Hourly 29 four historical years distributed under an open license.