Hourly Short Term Load Forecasting for Residential Buildings and Energy Communities
Aleksei Kychkin, Georgios C. Chasparis
TL;DR
The paper tackles the challenge of hourly short-term load forecasting for residential buildings and energy communities, comparing a wide range of forecasting models from physics-based to AI-based approaches. It introduces hourly adaptations of persistence-based models (PARH, PAReH, SPRH) and demonstrates substantial accuracy gains (15-30%) over standard methods through extensive evaluation on real-world data from a small energy community. Key contributions include a comprehensive hourly forecasting framework, detailed notations and model descriptions, and a rigorous comparison showing that parsimonious, causality-focused methods can outperform heavier models in practical, hourly decision contexts. The findings have practical implications for demand response and energy management, enabling more reliable, low-complexity hourly forecasts in residential microgrid settings.
Abstract
Electricity load consumption may be extremely complex in terms of profile patterns, as it depends on a wide range of human factors, and it is often correlated with several exogenous factors, such as the availability of renewable energy and the weather conditions. The first goal of this paper is to investigate the performance of a large selection of different types of forecasting models in predicting the electricity load consumption within the short time horizon of a day or few hours ahead. Such forecasts may be rather useful for the energy management of individual residential buildings or small energy communities. In particular, we introduce persistence models, standard auto-regressive-based machine learning models, and more advanced deep learning models. The second goal of this paper is to introduce two alternative modeling approaches that are simpler in structure while they take into account domain specific knowledge, as compared to the previously mentioned black-box modeling techniques. In particular, we consider the persistence-based auto-regressive model (PAR) and the seasonal persistence-based regressive model (SPR), priorly introduced by the authors. In this paper, we specifically tailor these models to accommodate the generation of hourly forecasts. The introduced models and the induced comparative analysis extend prior work of the authors which was restricted to day-ahead forecasts. We observed a 15-30% increase in the prediction accuracy of the newly introduced hourly-based forecasting models over existing approaches.