Power Quality Events’ Classification employing Discrete Wavelet Transform and Machine Learning

The Power Quality (PQ) studies have gained huge attention from the academics and the industry over the past three decades. The excessive use of power semiconductor devices in a grid utility increases the malfunction of the control system, produces Power Quality Disturbances (PQDs) and reduces the electric component life. The accurate detection and classification of PQDs in power system is key step to determine the causes of these events before any proper counter measure could be taken. Most of disturbances are non-stationary and transitory in nature; hence it requires advanced tools and techniques for the analysis of PQDs. In this paper, mathematical parametric models are used to acquire the PQ signals as per the definition provided in IEEE 1159-2009 Standards. The acquired signals are then to extract the distinctive features using Discrete Wavelet Transform (DWT) signal processing technique. Finally, the signals are classified using two machine learning algorithm (MLA) based classifiers; K-Nearest Neighbors (KNN) and Naive Bayes (NB). The proposed method is further validated in presence of noise accompanied PQ signals too.