A fast and reliable estimation of building energy need is essential in agricultural building design, nonetheless, a large number of simulations is required to obtain better energy saving solutions. The aim of this work is to understand if machine learning can substitute numerical simulations and speed up the building design process and assess the incidence of specific architectural elements. Supervised regression models has been trained and tested in a data-set of thousands simulations performed on a case-study agricultural building. Among the algorithms, the tree-based Extreme Gradient Boosting showed the best performance. A study on model explainability has been carried out using SHAP and features importance, which is fundamental to help academics and professionals devise better design strategies for both new constructions and retrofitting interventions.

Simulations in agricultural buildings: a machine learning approach to forecast seasonal energy need / Ceccarelli, Mattia; Barbaresi, Alberto; Menichetti, Giulia; Santolini, Enrica; Bovo, Marco; Tassinari, Patrizia; Barreca, Francesco; Torreggiani, Daniele. - (2022), pp. 116-120. [10.1109/MetroAgriFor55389.2022.9965083]

Simulations in agricultural buildings: a machine learning approach to forecast seasonal energy need

Barreca, Francesco
Methodology
;
2022-01-01

Abstract

A fast and reliable estimation of building energy need is essential in agricultural building design, nonetheless, a large number of simulations is required to obtain better energy saving solutions. The aim of this work is to understand if machine learning can substitute numerical simulations and speed up the building design process and assess the incidence of specific architectural elements. Supervised regression models has been trained and tested in a data-set of thousands simulations performed on a case-study agricultural building. Among the algorithms, the tree-based Extreme Gradient Boosting showed the best performance. A study on model explainability has been carried out using SHAP and features importance, which is fundamental to help academics and professionals devise better design strategies for both new constructions and retrofitting interventions.
2022
978-1-6654-6998-2
Radio frequency;Computational modeling;Buildings;Predictive models;Reliability engineering;Prediction algorithms;Numerical simulation;machine learning;building energy simulation;energy saving;ML explainability;food storage buildings
File in questo prodotto:
File Dimensione Formato  
Ceccarelli_2022_Simulations_in_agricultural_buildings.pdf

non disponibili

Tipologia: Versione Editoriale (PDF)
Licenza: Tutti i diritti riservati (All rights reserved)
Dimensione 1.24 MB
Formato Adobe PDF
1.24 MB Adobe PDF   Visualizza/Apri   Richiedi una copia

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/20.500.12318/143809
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 4
  • ???jsp.display-item.citation.isi??? 4
social impact