Rule-based Self-supervised learning Non-Intrusive Load Monitoring system for residential houses

Real-time non-intrusive load monitoring (NILM) systems have gained significant importance in various applications, including demand-side management, appliance anomaly detection, and ambient assisted living. While numerous research-based solutions have been developed over the past two decades, still it remains a need for an affordable application-level NILM solution. In this paper, we present a novel rule-based self-supervised approach that operates within a low-price range. Our proposed methodology offers automatic identification of unknown appliances within the system, achieving real-time appliance identification with a satisfactory level of accuracy. We contribute to the development of a cost-effective and efficient solution for non-intrusive load monitoring, enabling applications across diverse domains.