2016 3rd International Conference on Information Science and Control Engineering (ICISCE 2016)

70 / 2016-03-18 10:57:51

Neural Network Pattern Recognition Based Non-intrusive Load Monitoring for a Residential Energy Management System

demand side management (DSM); energy conservation; non-intrusive load monitoring (NILM); residential energy management system (REMS); neural network pattern recognition (NNPR); demand respond (DR)

Detailed energy consumption information of household appliance is meaningful for the demand side management (DSM) and home energy conservation. In this paper, a novel non-intrusive load monitoring (NILM) method is proposed for residential energy management system (REMS). Unlike existing NILM techniques, this method works effectively with very few smart meter measurement parameters obtained at a low sampling rate. A neural network pattern recognition (NNPR) model is utilized in the NILM system for the first time. The proposed method can detect finite-state appliances by changing the number of output neurons. Experimental results indicate that the proposed method provides a very high identification accuracy. Moreover, this method can estimate each appliance detail energy consumption effectively, which is ideal for scheduling the household appliances and participation in the demand respond (DR).