Knowledge Extraction on Energy Consumption in an Educational Institution Using Smart Energy Meter Data Analytics

The epitome need of tracing energy demand patterns for higher educational institutions (HEI) is to infer the sustainable functioning of an institute and to make consumers aware of the consumption range for incorporating appropriate energy strategies. Smart energy meters play a crucial role in enhancing building energy performance promisingly. The proposed work focuses on (1) grabbing the exact demand pattern by retrieving and processing continuous logs on energy consumption from smart meters and (2) interpreting data by k -means clustering based on underlying similarities. A generic method of consumption pattern extraction by temporal analysis was adopted in Thiagarajar College of Engineering (TCE), Madurai, for real-time imputation and analysis. The main aspect behind the proposed approach is to avoid penalty on exceeding day-wise maximum average power demand reach. With analysis, the role of the smart meter and the need for data pre-processing are emphasized. Overall energy consumption range and time duration are visualized using various kinds of plots and following that knee point of cluster behind maximum demand reach is found. The work further paves way for active participation of customers on demand-side management on accounting energy tariff.