Modeling the effect of roof coatings materials on the building thermal temperature variations based on an artificial intelligence

Abstract This paper presents the benefits of applying the artificial neural networks technique to model the building indoor temperature variations caused by implementing roof coatings. Four of the most common roof coating used in Mexican constructions were considered. The study evaluated test cells at outdoors and instrumented them to measure the indoor temperature variation. The environmental parameters solar radiation, wind speed, ambient temperature, and relative humidity were also monitored. The results showed the viability of modeling the thermal behavior generated by the roof coatings with an accuracy greater than 90%. It implies the ability to predict internal temperatures based on the physical properties of the material and environmental variables, being a feasible tool in the search for thermal comfort for buildings in the region. In addition, the results of the sensitivity analysis identified to what extent to improve the optical properties of the material impact on the heat transferred to the interior.