Exploring CNN Model with Inrush Current Pattern for Non-Intrusive Load Monitoring

Non-Intrusive Load Monitoring (NILM) has gradually become a research focus in recent years to measure the power consumption households for energy conservation. Most of existing algorithms on NILM models independently when total current load appliances occurs, and usually undergoes problem signatures appliance. This paper presents distingue design classify by investigating inrush pattern alliances begin. The proposed method is implemented while five operate simultaneously. high sampling rate field-programmable gate array (FPGA) used sample current, then converted be image patterns using kurtogram technique. These images are arranged four groups data set depending number operating Furthermore, modifications convolutional neural networks (CNN), which based very deep (VGGNet), designed adjusting size decrease training time increase faster operation. CNNs implement as classification model compare with previous models. F1 score Recall accuracy classification. results showed that system could achieved at 99.06