Real-Time Non-Intrusive Load Monitoring System for Residential Appliance Identification

Non-intrusive load monitoring (NILM) has developed as a prominent and intriguing topic in energy management recently. Even though a variety of NILM-based solutions have been proposed, a commercial-grade universal solution has not yet been established. Event detection and appliance identification algorithms have inherent critical limitations; therefore, this paper provides a novel real-time identification method aimed at being effective in NILM-based systems created for residential loads. Additionally, this work proposes an efficient event detection technique and a state identification method for multi-state appliances. A verification procedure allows for the correct identification of similar-featured appliances. The proposed algorithm identifies appliances with a high degree of accuracy by displaying their functioning condition in real-world conditions.