The prediction and application of Polyurethane Foam in developing roofing sheets cannot be over-emphasized when considering the environmental changes coursed by thermal radiation. This paper presents an artificial neural network application to model and predict the indoor temperature resistance of polyurethane (PU) roofing in residential buildings. The study employed a data logger to measure the indoor and outdoor temperatures for three simulation environments (i.e., morning, afternoon, and evening) for two hours each. Furthermore, the authors employed the Levenberg-Marquardt algorithm to transform and predict the indoor temperature obtained in the residential building's polyurethane roofing house. The result shows that the PU roofing system could absorb the heat and reduce the house model's temperature with 6.9% in the morning, afternoon 15.8%, and 6.8% in the evening when compared with the temperature outdoor environment. The ANN was also able to train, test, and validate the experimental temperature results with 92.86%, 93.92%, and 95%, respectively. The mean square error and a testing error occurs at 0.1707 and 0.1689. Therefore, this study concluded that ANN's application in predicting the thermal insulation material such as the PU roofing system is highly efficient and will increase the manufacturer's performance evaluation. It has also created significant awareness of the community in employing the PU roofing system for residential buildings, which will reduce the rate of energy consumption in buildings.
Modeling the Bending Strength of MDF Faced, Polyurethane Foam-Cored Sandwich Panels Using Response Surface Methodology (RSM) and Artificial Neural Network (ANN)
