Using Artificial Intelligence to Analyze the Thermal Behavior of Building Roofs

AbstractThis paper presents the application of an artificial neural network to model the thermal behavior of some roof coatings used in buildings. A set of test cells was built to evaluate these roof coatings. The cells were placed outdoors and several parameters were measured and collected for seve...

Full description

Saved in:
Bibliographic Details
Published in:Journal of energy engineering 2020-08, Vol.146 (4)
Main Authors: Ledesma, Sergio, Hernández-Pérez, I, Belman-Flores, J. M, Alfaro-Ayala, J. A, Xamán, J, Fallavollita, Pascal
Format: Article
Language:English
Subjects:
Artificial intelligence
Artificial neural networks
Buildings
Coatings
Computer simulation
Emissivity
Evaluation
Fluctuations
Heat
Heat flux
Heat transfer
Histograms
Irradiance
Mathematical models
Neural networks
Parameters
Reflectance
Roofing
Roofs
Statistical analysis
Technical Papers
Thermodynamic properties
Transducers
Wind measurement
Wind speed
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:AbstractThis paper presents the application of an artificial neural network to model the thermal behavior of some roof coatings used in buildings. A set of test cells was built to evaluate these roof coatings. The cells were placed outdoors and several parameters were measured and collected for several weeks. The measured parameters included the temperature in different parts of the test cells. Additionally, the solar irradiance, the humidity, and the wind speed were measured and stored. We designed, built, and calibrated several heat flux transducers to measure the heat flux in each cell. Further, the reflectance and emissivity of the roof coatings were measured and used to create the model. The main contribution of this work is the modeling of an experimental system to evaluate the variability of the heat flux in building roofs using histograms. A statistical analysis based on computer simulations employing neural networks was performed to analyze those parameters that affect the heat flux in the roofs the most and the least. Finally, it was found that under specific conditions small increments in the reflectance of the coating can produce significant changes in the heat flux in the roof.
ISSN:0733-9402
1943-7897
DOI:10.1061/(ASCE)EY.1943-7897.0000677