AIoT for Smart Homes: Real-Time Non-Intrusive Load Monitoring
H Jai Venkat Krishna, S. M. Meena · 2024
Facing rising energy demand, we need smarter solutions for conservation. Non-Intrusive Load Monitoring (NILM) offers a cost-effective way to track energy usage, but current methods struggle with real-world implementation. This presentation introduces a novel AI-powered approach to NILM using Deep Neural Networks (DNN). We convert power signatures from appliances into image representations, allowing our DNN to accurately classify which appliances are active. This breakthrough addresses key challenges, including the need for manual training and limited data availability. Our research paves the way for practical NILM solutions, enabling homeowners to gain real-time insights into their energy consumption and make informed decisions for conservation.