Diagnostic expert system of transformer insulation systems using the acoustic emission method

Abstract

The subject matter of this paper refers to the improvement of the acoustic emission (AE) method when used for detection, measurement and location of partial discharges (PDs) in oil insulation systems of power appliances. In particular, presents the basic assumptions and describe the individual elements of an expert diagnostic system, which uses an acoustic method to asset the state of the measured power transformer insulation, performed during their normal work in industrial conditions. The system will consist of four basic modules, i.e. measuring system, processing-analyzing system, knowledge base and classifier. These modules are characterized successively in Point 2 of the paper. Special attention is given to the description of the group of multiparametric descriptors characterizing the AE signals in the time, frequency and time-frequency domains, to descriptive statistics indexes and correlative parameters. Accordant descriptors make it possible, at strictly defined metrological conditions, to recognize basic PD forms that may occur in paper-oil insulation. In this way a catalogued knowledge base containing the socalled 'fingerprints' was created for basic types of high-voltage defects of insulation system.