EMBED: A Dataset for Energy Monitoring through Building Electricity Disaggregation

摘要: Electricity disaggregation, the task of inferring appliances' energy consumption in a building from few sensing points, has received attention community. In this paper, we introduce EMBED, publically available dataset for Energy Monitoring through Building Disaggregation. EMBED is most comprehensive fully labeled disaggregation to date containing aggregate power measurement as well plug load data different appliances three residential units, each was collected at least two weeks. Therefore, could be used comparing performance both event-based and non-even-based methodologies. dataset, all events were uniquely by identifying separate labels transition states appliance. This feature enables evaluation classification or clustering algorithms disaggregation. The paper discusses hardware setup, collection procedure, unique characteristics compared similar works. addition, it presents preliminary analyses on benchmark encourage future research domain.