The impact of indoor climate and occupancy rates on electricity use in campus buildings: Analyse the impact level through the linear regression model

Energy efficiency in buildings can be achieved by many approaches including a high-performance HVAC system, passive design strategies, and accurate energy consumption prediction. Energy evaluation is crucial in existing buildings to improve system control by detecting any unusual pattern of consumption as well as identifying the most influential variables in energy usage for a better target in retrofit projects. This paper aims to identify the significant contributors to energy consumption in both heating and non-heating periods of campus buildings situated in the south of France. The linear regression model is adopted to investigate the relationship between the dependent variable (response) and independent variables (predictors). During the heating period, the analysis focuses on five potential variables, namely, time index (daytime/night-time), day index (weekday/weekend), building size, indoor and outdoor climate, and their confounding. When electric heaters (reversible air-conditioners) are likely to be involved, results have shown that the model with the interactive effect variables achieves a coefficient of determination of 68%. That model consists of CO2, time index, day index, building size, indoor humidity, and indoor temperature. During the non-heating period, the occupancy rate is the main target for its impact level on electricity use. The regression model has found that the occupancy rate and day index alone can explain 50% of the consumption variation. Together, this study points out the importance of the occupant’s precision to better the predictive model in future studies. This includes a better assumption of occupied space and schedule and a better understanding of the scale of the occupant.