Algorithm for Estimating the Degree of Polymerization of Paper Insulation Impregnated with Inhibited Insulating Oil

authored by: Bayu Hadi Nugraha, Tobias Kinkeldey, Tobias Münster, Peter Werle, Suwarno
Abstract: Paper insulation in power transformers may deteriorate due to ageing processes during the operation. The widely used parameter for determining paper insulation condition is the degree of polymerization (DP). Nowadays, DP measurement techniques without damaging the components of the transformer insulation system or interfere with the transformer service time are needed. The most common method is using a furan analysis to estimate DP value. However, there are different factors and uncertainties such as the occurrence of local overheating at one spot, different periods of measurements after an oil processing, etc. thus, they have an impact on the accuracy of the furan measurement. Therefore, several estimation algorithms had been developed by considering the chemical and dielectric parameters of un-inhibited insulating oil.This contribution aims to develop an algorithm for estimating the DP value of paper insulation on transformers immersed with inhibited insulating oil. An algorithm was implemented to field operating transformer measurement data. The dielectric characteristic parameters (acidity, interfacial tension), as well as dissolved gases (carbon monoxide, carbon dioxide), were processed and evaluated by a specialized fuzzy inference system (FIS).
Organisation(s): High Voltage Engineering and Asset Management Section (Schering Institute)
External Organisation(s): PT PLN (Persero) Kantor Pusat
Institut Teknologi Bandung (ITB)
Type: Conference contribution
No. of pages: 6
Publication date: 10.2019
Publication status: Published
Peer reviewed: Yes
ASJC Scopus subject areas: Energy Engineering and Power Technology, Control and Systems Engineering, Electrical and Electronic Engineering, Safety, Risk, Reliability and Quality, Control and Optimization
Electronic version(s): https://doi.org/10.1109/ICHVEPS47643.2019.9011070 (Access: Closed)