Methanol is considered to be a new chemical marker to estimate the aging state of transformer cellulose insulation paper. The key to improve the assessment accuracy is to establish the precise relationship between the concentration of methanol in oil and degree of polymerization (DP) of insulation paper. The previous research in this area has shortcomings and needs to be further improved. In this article, first, an aging assessment basic model (temperature 20 °C, moisture content of insulation paper 0.637%) is established based on the theoretical derivation of cellulose degradation kinetics, and is calculated by 130 °C thermal aging experiment. Furthermore, the effects of different temperatures (20 °C–90 °C) and moisture contents of insulation paper (0.6%–3.4%) on methanol in oil are studied by diffusion experiment. Based on the above effects, the temperature and moisture correction factors are established by mathematical reduction. What is more, a modified model is proposed by integrating the basic model and the correction factors. Eventually, the verification experiment is randomly prepared to test the accuracy of modified model and the result shows that the relative error of DP is within 7%. The physical meaning and accuracy of aging assessment are improved by model derivation and modification of influencing factors.

The insulation paper is a crucial factor for evaluating the insulation status of transformers. The traditional evaluation methods for insulation paper are dissolved gas component content analysis (CO and CO ₂ ) and furfural (2‐FAL) content analysis in oil. The detection principle of the former leads to its low accuracy, while the field application effect of the latter is not ideal due to the easy oxidation of furfural and low content. Methanol, result of its good stability and high production compared with other marker products (CO, CO ₂ , 2‐FAL) has been investigated as a novel marker for aging evaluation of transformer insulating paper in recent years. The results of the investigation indicated that there is a good correlation between the content of methanol in oil and the degree of polymerization under laboratory conditions, which reflected the insulation aging state of the transformer insulating paper more accurately. This paper summarizes the current research status of methanol formation mechanism, detection methods and aging evaluation in insulating oil, and points out the key problems to be solved and development prospects, hoping to provide relevant reference for aging evaluation of oil immersed power transformer.

The accuracy of predicting the degree of polymerization (DP) of insulating paper using Raman and FTIR fusion spectroscopy is affected by the initial moisture content. To address this issue, this paper constructs a robust spectral analysis framework. This method integrates Raman and infrared spectra, adopts an improved mRMR algorithm with an implicit moisture penalty term to achieve anti-interference feature extraction, and conducts efficient prediction based on the Matern 5/2 Gaussian process regression model optimized by L-BFGS. The R² of M52-GPR optimized by L-BFGS reached 99.01%. The root mean square error (RMSE) and mean absolute error (MAE) decreased to 15.14 and 11.36 respectively. This paper verified the superiority of the mRMR algorithm in spectral feature selection in water-containing environments, providing a high-precision regression analysis method for the aging assessment of insulating materials.

Replacing aged oil (RAO) can retard the aging development of transformer insulation by regularly filtering out the harmful substances in oil, however the RAO itself may affect the estimation accuracy of aging status of paper insulation based on chemical characteristic marker methods, such as carbon oxides, furfural, and methanol (MeOH). In this paper, accelerating thermal aging experiments are carried out, and the influence of RAO on MeOH concentration in insulating oil and the estimation degree of polymerization (DP) of paper insulation is investigated. The results indicate that the current estimation methods are unsuitable for RAO based on the MeOH in oil. Furthermore, the MeOH correction factor and DP correction factor are established as power-series expansion, and a two-step modification method which considered both variation of MeOH in oil and DP of paper insulation is proposed for modifying the estimation errors by RAO. Eventually, the verification experiment is carried out for verifying the estimation accuracy of the proposed method, and the results indicate that the range of average relative estimation errors is reduced to 1.8 %~3.9 %. The accuracy of aging estimation is improved and the proposed method has good applicability.

A method for predicting the failure time and failure voltage of generator stator bars under the electrical stress test was proposed. The method established a general life prediction theory model for the electrical stress test based on the negative power-exponential model, and this model could predict the insulation life of stator bars under the electrical aging test and breakdown test. Then, we utilized Weibull distribution theory to estimate the life factor in the model and predict the failure time and failure voltage of the stator bars under the step-by-step test. The test results were consistent with the estimated results, and the correctness of the proposed method was verified.