Application of machine learning to estimate building energy use intensities

Information on building energy consumption and its characteristics is essential for carrying out benchmarking processes. However, currently a lack of data acts as a major barrier in this regard. To address this issue, the purpose of this study is to demonstrate the application of machine learning to estimate building energy use intensities of bank branches buildings located in Brazil. The methodology proposed in this study completed a data collection regarding the bank branch typology. Then, the archetype model and its fixed and variables inputs variables were defined to generate 48,000 samples that were simulated using EnergyPlus program. A Sobol sensitivity analysis was performed, showing that the lighting power density followed by the weather variable were found to be the most influential variables when estimating the energy consumption of the bank branches. Finally, a comparison between machine learning techniques were applied to train the predictive model. The Support Vector Machine achieved MAE and RMSE values of 3.16 and 4.45 kWh/m².year, respectively, representing the most accurate model for benchmarking purposes. Due to the non-linearity among the variable parameters, optimizing sophisticated machine learning techniques significantly improved the model accuracy. The results are of great value, since the model developed can be used in future benchmarking throughout the country. The methodology showed high accuracy and could be extended to other typologies.