Simultaneous Load Identification Method Based on Hybrid Features and Genetic Algorithm for Nonintrusive Load Monitoring

Nonintrusive load monitoring (NILM) is a widely accepted technology to conduct monitoring. Many effective methods have been established make NILM more practical. However, the focus of current mainly on identification accuracy and efficiency single under individual appliance operated independently, which limited support for problem multiple appliances simultaneously. Therefore, simultaneous method proposed efficiently identify total simultaneously in this paper. The consists three parts: hybrid features extraction, optimization model construction, frequency-weighting-factor-based genetic algorithm (FWF-GA). Firstly, feature model, integrates active power, reactive harmonic magnitude, constructed by extraction. Secondly, employing power. Thirdly, developed FWF-GA used solve problem. In FWF-GA, relative errors frequency-weighting factor magnitude are evaluate fitness an individual. Finally, practice household demonstrate validity method.