Non-intrusive Electrical Appliance Identification System Based on Dynamic Characteristic Parameters

Non-intrusive electrical appliance identification is the key technology to realize customer-side intelligent sensing in the ubiquitous power Internet of Things. Aiming at the problem that the existing non-intrusive electrical appliance identification system has too high computational complexity, which is not conducive to low power consumption and industrialization, a low-complexity and high-precision non-intrusive electrical appliance identification method based on dynamic characteristic parameters and cumulative sum (CUSUM) is proposed. Firstly, the current is sampled synchronously. Secondly, a proposed multi-parameter detection method is used to determine the state change of the electric appliance. Thirdly the dynamic characteristic parameters are extracted according to the changed information. Fourthly, the Euclidean distance is used to identify the electric appliance which has changed. Finally, the type and quantity of current electrical appliances are identified by using CUSUM. The experimental results show that only a small amount of data training for a single electrical appliance is needed to achieve accurate identification of combined electrical appliances, which has high practical value.