A Novel NILM Event Detection Algorithm Based on Different Frequency Scales

Non-intrusive Load Monitoring is a technology that can identify the users’ internal energy consumption by using the data measured at a single point on the bus and event detection is a key technical problem that need to be solved. An algorithm combining probability and expert heuristic models is proposed for event detection in this study, including an event pre-detection sub-algorithm called Voting Improved Isolated Forest for high-sensitivity events pre-detection and an events verification sub-algorithm called Time Shift Downsampling Matching for high-accuracy events verification. Voting Improved Isolated Forest is used to detect suspicious events quickly from the low-frequency characteristics of the signal; Time Shift Downsampling Matching identify real events from suspicious events through analyzing high-frequency characteristics of the signal. To evaluate the proposed algorithm, three datasets are used. Comparing with the state-of-art algorithms, the proposed algorithm has great adaptability and accuracy to long-transient events and small-signal events.