On-Line Classification of Excessive Neutral-to-Earth-Voltage (NTEV) Sources Using LabVIEW Software with Incorporating the Statistical-Based S-Transform and One-Versus-One SVM (OVO-SVM)

Abstract

The excessive Neutral-to-Earth-Voltage (NTEV) could lead to the unnecessary losses and safety hazard issue to the electrical consumer. Hence, it is a paramount need to mitigate the excessive NTEV from occurring in the system. In order to mitigate the excessive NTEV, it is very crucial to identify the source of this problem so that the proper mitigation technique can be deployed. One of the identifying approaches is by classifying the type of signal which is associated to excessive NTEV. In this paper, the propose classification technique is using Statistical-based S-transform and one-versus-one SVM (OVO-SVM) via the LabVIEW platform. In this case, the MathScript Node is utilized as an interactive interface in .m file commands with LabVIEW platform. As for SVM, different types of kernel function such as linear, gaussian, sigmoid, and polynomial have been used for performance evaluation comparison. The finding from this work indicates that OVO-SVM learning tool using Gaussian technique produced the highest classification of accuracy result.