Application and Comparative Analysis of Fuzzy Inference System for Transformer Fault Diagnosis with Dissolved Gases in Oil

Abstract

Oils are used for cooling and insulation in transformers. The fault currents occurred on transformers can damage windings and insulation. As a result of this incoming fault current, gas emission occurs in materials such as oil and pressboard. The type of these gases released gives information about the fault of the transformer. In this study, it has been studied on the determination of the fault by analyzing the gases released as a result of the failure in 63 transformers using FIS. As a result of the analysis, the results obtained with FIS were compared with analyzed results with IEC 60599 and fault reports from the corporation, which these values obtained. When 63.5% true results were obtained compared to IEC 60599, 97.6% correct results were obtained compared to fault reports of the corporation because 22 of 63 transformer gas values were not actually suitable for applying the IEC Ratio method.

Fuzzy Inference System Development for Turbogenerator Failure Diagnosis on Floating Production Offloading and Storage Platform

“…Two analysis methods are commonly used in fuzzy inference systems: Mamdani and Sugeno, with the Mamdani method usually preferred [28]. This method consists of four steps as described below:…”