Determination of RMS based Continuous S-Transform with Deep Learning for Detecting and Classifying Voltage Sag and Swell

Abstract

Voltage sag and swell can cause major issues in power quality, including as instability, low lifetime, and data mistakes.  Voltage swell normally associate with the system fault condition. The purpose of this paper is to show detection and classification of voltage sag and swell. The Root Mean Square (RMS) approach uses the S-Transform as a base to detect the triggering point of disturbances. This research also shows the different types of sags and swells by incorporating the features into a MATLAB-based Extreme Learning Machine (ELM) neural network technique. In addition, the ELM approach is compared to the Support Vector Machine (SVM) and Decision Tree methods to see which one performs the best categorization. The classification accuracy was expressed as a percentage. Because the findings clearly illustrate the advantages of RMS in detecting and ELM in categorizing power quality problems, it was proved that detection and classification using RMS and ELM are possible.