Research on Non-Intrusive Load Monitoring Based on Wavelet Transform and Neural Network
Smart home appliances are becoming more and more intelligent,and the working mode is also becoming more and more vari-ous.At present,the non-invasive load monitoring of household appliances only adopts the single load curve of traditional household ap-pliances,which greatly limits the application and popularization of non-invasive load monitoring technology.Therefore,different working modes of smart washing machine are taken as an example to study the power load characteristics of the same household appliance in dif-ferent working modes,the power load data of the washing machine in smart household appliance in different working modes are collect-ed,the load curve is smoothed,the feature information is extracted through the method of wavelet transform,and the feature vector repre-senting the feature information is evaluated based on the idea of statistics.A neural network model is established to identify the load curves of different working modes.Through MATLAB platform simulation,it is proved that the method based on wavelet analysis feature extraction and neural network feature recognition is feasible in non-invasive intelligent home appliance load monitoring,the proposed method is with high accuracy and good application value.
non intrusiveload monitoringwavelet analysisneural network
万方数据
维普
