qt8br6b7rh.pdf (418.53 kB)
Download file

A low-complexity non-intrusive approach to predict the energy demand of buildings over short-term horizons

Download (418.53 kB)
journal contribution
posted on 04.11.2021, 16:56 by Athanasios Aris Panagopoulos, Filippos Christianos, Michail Katsigiannis, Konstantinos Mykoniatis, Marco Pritoni, Orestis P. Panagopoulos, Therese Peffer, Georgios Chalkiadakis, David E. Culler, Nick JenningsNick Jennings, Timothy Lipman
Reliable, non-intrusive, short-term (of up to 12 h ahead) prediction of a building's energy demand is a critical component of intelligent energy management applications. A number of such approaches have been proposed over time, utilizing various statistical and, more recently, machine learning techniques, such as decision trees, neural networks and support vector machines. Importantly, all of these works barely outperform simple seasonal auto-regressive integrated moving average models, while their complexity is significantly higher. In this work, we propose a novel low-complexity non-intrusive approach that improves the predictive accuracy of the state-of-the-art by up to (Formula presented.). The backbone of our approach is a K-nearest neighbours search method, that exploits the demand pattern of the most similar historical days, and incorporates appropriate time-series pre-processing and easing. In the context of this work, we evaluate our approach against state-of-the-art methods and provide insights on their performance.


Published in

Advances in Building Energy Research


Taylor & Francis


AM (Accepted Manuscript)

Rights holder

© Taylor & Francis

Publisher statement

This is an Accepted Manuscript of an article published by Taylor & Francis in Advances in Building Energy Research on 23 Oct 2020, available online: http://www.tandfonline.com/10.1080/17512549.2020.1835712.

Acceptance date


Publication date


Copyright date