Energy consumption characteristics based on monitored data

a school case study

Abstract

This study is an initial step for a creation of smart platform for schools in Estonia which will model, analyze, and evaluate the real energy performance of school buildings. Energy meters provide electricity consumption data which can be used to understand energy usage patterns and finally improve building energy management. First, data preparation is made. On the following step hierarchical clustering is applied to identify the outliers of weekly electricity load profiles. Finally, daily electrical load patters are clustered and similar profiles are grouped using K-means centroids.