Neural Network to predict Energy Efficiency for Space Heating in Residential Buildings

The objective of this work is to develop a novel methodology for predicting space heating energy consumption in residential buildings using artificial neural networks in order to evaluate the energy efficiency of the building. Leveraging a dataset of 2268 simulations from "DesignBuilder" software, the ANN model was trained and optimized, achieving 99% accuracy on both training and testing. The proposed approach offers an efficient alternative for evaluating energy requirements, paving the way for sustainable energy management and informed decision-making. Future work aims to extend the model to predict other energy needs and develop user-friendly application on smartphones, and to optimize energy efficiency in buildings.