Inferring Energy Consumption Patterns in Public Buildings

The advent of smart meters have radically changed the mechanisms traditionally used for energy consumption monitoring. The possibility of having (i) highly frequent readings (even every minute); (ii) accurate energy data gathering; and (iii) real time data exchange capabilities to share the energy consumption data with other elements in the Smart Grid, open the door to further deeper analysis within this context. In this paper, we focus on obtaining energy consumption patterns. Our approach combines clustering and predictive techniques in order to infer these patterns using data gathered from public buildings in a university campus. Our analysis allows us to infer that clustering is not an appropriate mechanism, since the use of buildings is mixed (administrative, labs, classrooms, etc.). However, predictive approaches give promising results, specially LSTM and XGBoost.