A Non-Intrusive Load Monitoring System Based on Blacklist Detection

Abstract

Electric utilisation safety is widely recognised as an important issue. Most electrical accidents can be concluded to the unallowed access of certain kinds of electrical appliances, hence can be effectively prevented by a blacklist-based detection. However, this ability is lacked in many of the existing Non-Intrusive Load Monitoring (NILM) system. This paper proposes a novel blacklist-based NILM system capable of extracting valuable information from the main wire when unallowed risky appliances are switched on in a household. Comprehensive methods are proposed to optimise the accuracy of detection and the self-learning ability on blacklist extension. Experimental results show that the proposed methods yield outstanding performance. The proposed system is applicable for unallowed appliances management, from the inhabitants or from the grid operator's point of view.