Robust Random Cut Forest Based Event Detection Method for Non-Intrusive Load Monitoring

Abstract

Non-Intrusive Load Monitoring (NILM) can provide the appliance-level power consumption information without deploying submeters in each load, in which load event detection is one of the crucial steps. However, the existing event detection methods are not efficient to detect both starting time of event (STE) and ending time of event (ETE), and the scalability is limited. To tackle these problems, this paper has proposed an event detection method based on robust random cut forest (RRCF) that is an unsupervised learning algorithm for detecting anomalous data points within a dataset. It has been validated on REDD dataset. The experimental results show that the proposed event detection method performs better than state-of-the-art.