Machine Learning Model Optimization for Energy Efficiency Prediction in Buildings Using XGBoost

Machine Learning is a field of Artificial Intelligence that has recently become very important when building intelligent systems. The goal is always to build a machine learning model with high accuracy, especially important when used for energy optimization applications such as energy performance of buildings (EPB). Due to growing concerns about energy waste and its impact on the environment, reports suggest that building energy consumption has increased over the past decades worldwide. Our goal is to create a state-of-the-art model based on Extreme Gradient Boosting (XGBoost) capable of predicting the required heating load (HL) and cooling load (CL) of a building in order to determine the specification of the heating and cooling equipment needed to maintain comfortable indoor air conditions in order to create a building designed optimized for a more sustainable energy consumption. An alternative way of achieving this would be through the use of a building energy simulation software, which is very time-consuming, using instead a machine learning solution offers the distinct advantage of an extremely fast prediction once a model is adequately trained. We were able to create an XGBoost regressor with a R² score of 0.99.