Two-Stage Residual Lifespan Prediction Model for Oil-Paper Insulation Based on Analysis of Furfural and Methanol in Oil

• Published in: IEEE transactions on power delivery Vol. 38; no. 1; pp. 432 - 441
• Main Authors: Liu, Jiefeng, Song, Boshu, Zhang, Enze, Zhang, Heng, Zhang, Yiyi, Goh, Hui Hwang
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.02.2023; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Aging; dielectric materials; Furfural; Insulation; Kinetic equations; Life span; lifespan estimation; Methanol; Moisture; Oil insulation; Oils; paper insulation; Power transformer insulation; Prediction models; Predictive models; Transformer
• ISSN: 0885-8977; 1937-4208
• DOI: 10.1109/TPWRD.2022.3192487

The applicability of existing single aging indicator-based lifespan prediction methods needs further improvement during different aging stages. Considering the aforementioned challenges, this work proposes a two-stage residual lifespan prediction model based on both methanol and furfural. Firstly, the changes in methanol and furfural contents in oil are examined concurrently with the aging duration. The paper insulation's life is divided into different aging stages by using identified two demarcation points according to the indicator's content. By virtue of this, the two-stage residual lifespan model of the insulating paper is obtained by coupling with the second-order aging kinetic equation. Finally, the validity of the proposed model is proved through a comparison to the conventional method of utilizing a single indicator, with a mean absolute percentage error for insulating paper's residual lifespan of 5.6%. In that regard, the research's contribution is to propose an integrated model for predicting the residual lifespan of insulating paper by using methanol and furfural indicators.