House and socio-demographic features vs. electricity consumption time series in main heating mode classification

Abstract

Demand-side flexibility is crucial for integrating variable renewable energy sources cost-effectively. Home heating systems determine the potential for flexibility in individual households. We examine different approaches to classify heating systems in Finland and find that using hourly electricity consumption time series is more effective than house and socio-demographic features. Classification based on electricity consumption data achieves higher precision (0.62) and recall (0.64) than house and socio-demographic features (0.41 and 0.43, respectively). Therefore, the availability of electricity consumption time series data should be considered from a competition policy perspective due to its value in estimating flexibility potential.