Identifying Fridge Consumption Non-intrusively Based On Temporal Characteristic

摘要

Non-Intrusive Load Monitoring (NILM) is a technique of inferring the power consumption of each appliance from an aggregate power data gathered from a single metering point. Fridges and freezers occupy the central part of total power consumption. In this paper, an unsupervised approach based on DBSCAN (Density-Based Spatial Clustering of Applications with Noise) is proposed for identifying the temporal characteristic of fridges and freezes,and decomposing them from the aggregate data. Validation results on the publicly available ECO dataset have shown that the proposed approach is a promising solution to extract the pattern of fridges and that the approach can effectively disaggregate the consumption with more than 80% accuracy. Furthermore, the remaining parts of aggregate data after decomposition can be used for further NILM studies.