Abstract

In this study, multiple linear and non-linear models are trained on real data to mimic the relation between electricity power consumption and indoor temperature, in the rooms of an apartment in downtown Stockholm. Most models proved to perform well when tested on validation data for almost all rooms, only showing performance dips for the middle room. A correlation between electricity consumption and indoor temperature is likely present for all rooms but all throughout small. The electricity power consumption increases the model prediction performance in comparison to only using indoor temperature.