Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Fifty shades of black: Uncovering physical models from symbolic regressions for scalable building heat dynamics identification:Uncovering physical models from symbolic regressions for scalable building heat dynamics identification
The rapid growth of machine learning (black-box) techniques and computing capacity has started to transform many research domains, including building performance analysis. However, physical interpretation of these models remains a challenge due to their opaque nature. This paper outlines an experiment to unveil analytical expressions from an open-source machine-learning-based algorithm, i.e., symbolic regression. From 241 residential buildings in the Netherlands, 50 unique analytical expressions were produced demonstrating overall better characterization accuracies than an XGBoost baseline, while providing a powerful mean of interpretability from model structures and coefficients. These insights present a starting point for further work towards highly scalable models yielding new characterizations of residential buildings.
Fifty shades of black: Uncovering physical models from symbolic regressions for scalable building heat dynamics identification:Uncovering physical models from symbolic regressions for scalable building heat dynamics identification
The rapid growth of machine learning (black-box) techniques and computing capacity has started to transform many research domains, including building performance analysis. However, physical interpretation of these models remains a challenge due to their opaque nature. This paper outlines an experiment to unveil analytical expressions from an open-source machine-learning-based algorithm, i.e., symbolic regression. From 241 residential buildings in the Netherlands, 50 unique analytical expressions were produced demonstrating overall better characterization accuracies than an XGBoost baseline, while providing a powerful mean of interpretability from model structures and coefficients. These insights present a starting point for further work towards highly scalable models yielding new characterizations of residential buildings.
Fifty shades of black: Uncovering physical models from symbolic regressions for scalable building heat dynamics identification:Uncovering physical models from symbolic regressions for scalable building heat dynamics identification
Leprince, Julien (Autor:in) / Miller, Clayton (Autor:in) / Frei, Mario (Autor:in) / Madsen, Henrik (Autor:in) / Zeiler, Wim (Autor:in)
01.01.2021
Leprince , J , Miller , C , Frei , M , Madsen , H & Zeiler , W 2021 , Fifty shades of black: Uncovering physical models from symbolic regressions for scalable building heat dynamics identification : Uncovering physical models from symbolic regressions for scalable building heat dynamics identification . in Proceedings of 8 th ACM International Conference on Systems for Energy-Efficient Built Environments . Association for Computing Machinery , pp. 345-348 , 8 th ACM International Conference on Systems for Energy-Efficient Built Environments , Coimbra , Portugal , 17/11/2021 . https://doi.org/10.1145/3486611.3491120
Aufsatz (Zeitschrift)
Elektronische Ressource
Englisch
Fifty shades of grey:Automated stochastic model identification of building heat dynamics
| BASE | 2022
Fifty shades of grey:Automated stochastic model identification of building heat dynamics
| BASE | 2022
| Springer Verlag | 2018
Gaussian Process regressions for seismic fragility assessment of building portfolios
| BASE | 2019
Linear regressions and structural correlations for models of stray-current circuits
| British Library Online Contents | 2006