Selection of Variables by the F-Score Algorithm for Radiated Magnetic Field Signals Discrimination of Electrical Discharges

EasyChair Preprint 359, version 2

Abstract

This paper proposes the use of a hybrid variables selection method called F-Score, which reduces the duration of the learning and test phases while improving the accuracy of recognition for radiated magnetic field signals discrimination of electrical discharges generally taking place in insulation systems and in particular in insulators of high-voltage lines and power transformers. The classifier used is based on support vector machines (SVMs). The experimental analysis was carried out on a basis of data comprising respectively 161 signals of which 106 learning and 55 for the test. The obtained results show that the proposed algorithm combined with the SVMs, allows a substantial reduction in the number of variables and a high improvement of the recognition rate compared to the pre-selection rate.

Keyphrases: Electrical discharges, F-score, Insulation Systems, Support Vector Machines (SVMs), Variables Selection

@booklet{EasyChair:359,
  author    = {Mohamed Gueraichi and Hocine Moulai and Azzedine Nacer},
  title     = {Selection of Variables by the F-Score Algorithm for Radiated Magnetic Field Signals Discrimination of Electrical Discharges},
  howpublished = {EasyChair Preprint 359},
  year      = {EasyChair, 2018}}