Adaptive Dynamic Meta-Heuristics for Feature Selection and Classification in Diagnostic Accuracy of Transformer Faults

Sherif S M Ghoneim,El-Sayed M El-Kenawy,Tamer Ahmed Farrag,A Ali Rashed,Abdelhameed Ibrahim

doi:10.1109/access.2021.3083593

Abstract

Detection of transformer faults avoids the transformer's undesirable loss from service and ensures utility service continuity. Diagnosis of transformer faults is determined using dissolved gas analysis (DGA). Several traditional DGA techniques, such as IEC code 60599, Rogers' ratio method, Dornenburg method, Key gas method, and Duval triangle method, but these DGA techniques suffer from poor diagnosis transformer faults. Therefore, more research was used to diagnose transformer fault and diagnostic accuracy by combining traditional DGA techniques with artificial intelligence and optimization techniques. In this paper, a proposed Adaptive Dynamic Polar Rose Guided Whale Optimization algorithm (AD-PRS-Guided WOA) improves the classification techniques' parameters that were used to enhance the transformer diagnostic accuracy. The results showed that the proposed AD-PRS-Guided WOA provides high diagnostic accuracy of transformer faults as 97.1%, which is higher than other DGA techniques in the literature. The statistical analysis based on different tests, including ANOVA and Wilcoxon's rank-sum, confirms the algorithm's accuracy.

Highlights

The power transformers are very crucial in the electrical power system, and the electricity utilities are keen to carry out inspections to monitor their status regularly
A voting classifier based on the proposed algorithm is developed to improve the tested dataset classification accuracy
The binary AD-PRSGuided WOA algorithm is evaluated in compared with the Grey Wolf Optimizer (GWO), Particle swarm optimizer (PSO) [43], Bat Algorithm (BA) [45], [46], WOA [47], Bowerbird Optimizer (SBO) [49], Multiverse Optimization (MVO) [50], Biogeography-Based Optimizer (BBO) [51], Firefly Algorithm (FA) [52], and Genetic Algorithm (GA) [53]

Summary

Introduction

The power transformers are very crucial in the electrical power system, and the electricity utilities are keen to carry out inspections to monitor their status regularly. The malfunction in their operation will lead to disconnection of the system and to revenue loss [1], [2]. Dissolved gas analysis (DGA) is a method that interprets the cause of transformer faults and identifies the fault types [4]. Some graphical representations are designed to identify the transformer faults such as Duval Triangle [7], [8], pentagon [10], [11], and heptagon [12]. The poor accuracy of the traditional DGA methods is observed, and decreasing the errors between the estimated and actual diagnostic faults requires other tools to solve this shortcoming; the intelligent techniques are utilized

Methods

Results

Conclusion