Energy Usage Modelling for Residences of a South African Academic Institution

Abstract

In residential buildings, various factors often have a significant impact on a building’s energy consumption. This paper aims to utilize various existing models to evaluate the sensitivity and influence of each of these factors on the building’s energy usage. The factors considered include: average temperature, heating degree days (HDD), cooling degree days (CDD), number of workdays, number of nonworkdays and building occupancy. The models considered were: linear two variable and multivariable regression, exponential regression and polynomial regression. The data used for the modeling was that of the energy usage of all the residences of a South African University, during the 2017 academic year. The results of this study revealed that the models developed using polynomial regression produced coefficient of determination (R2 ) values ranging from 0.7 to 0.89 in the case of temperature and occupancy, and 0.39-0.69 in the case of workdays and non-workdays, which were the highest model accuracies when compared to those of other models. Analysis of the results also revealed that certain factors such as building occupancy had a greater correlation to the building energy usage. The final model developed (A linear multivariable regression model) achieved an R2 value of 0.95 indicating the model’s high accuracy in predicting the dependent variable (energy consumption) using the factors indicated as independent variables to the model.