Synchrosqueezed Wavelet transform Based Power Quality Disturbance Detection and Monitoring of Solar Integrated Micro-Grid

Abstract

Microgrid monitoring has gained popularity in recent years, owing to their popularity and the power quality disturbances (PQD) caused by renewable energies. Many artificial intelligence-based PQD monitoring strategies have been proposed by researchers. This paper presents a novel method for detecting, classifying and monitoring the power quality disturbances in solar integrated microgrids. Power quality events are detected by analysing the frequency signal with the wavelet synchro-squeezing transform (WSST). The features of power quality disturbances are used to classify them. For PQD classification, the retrieved features are fed into a Convolutional neural network (CNN) classifier, using GoogLeNet and SqueezeNetfier. Thus the accuracy was found to be 99.90% and 93.76% respectively.