Classification of The NTEV Signal Problem via the Incorporation of S-Transform Features and Different Types of Neural Network

Abstract

Classification of power quality (PQ) disturbance on the commercial building is one of the most important parts in monitoring, identifying and mitigating of PQ disturbance to avoid misunderstanding the behavior of events. A novel on the Neutral to Earth Voltage (NTEV) classification using Stransform (ST) and different type of neural networks are proposed. The types of a neural network composed of general regression neural network (GRNN), probabilistic neural network (PNN) and radial basis function neural network (RBFNN). NTEV signals are needed to analyse using ST to extract their features that used as an input for the neural network classification. Finally, the GRNN, PNN, and RBFNN are trained and tested using 100 and 150 samples respectively. The performance of GRNN, PNN, and RBFNN are compared in which to identify the best technique in classification the NTEV.