Non-Destructive Detection of Aging State of Insulating Paper Based on Self-Similarity Feature of Terahertz Absorption Spectra

The polymerization degree of insulating paper/board is an important parameter to characterize its aging degree, and its rapid and accurate measurement can grasp the insulation aging status of oil-immersed power transformers. The traditional viscous polymerization measurement method not only needs to destroy the samples, but also leads to large dispersion of the test results due to the cumbersome pre-processing and operation procedures. Previous studies have used terahertz technology to quantitatively study the aging degree of oil-paper insulation, but the universality is poor. In the previous experiments, researchers found that the terahertz absorption spectra of the same insulation paper samples at the end of the aging period would change significantly compared with the initial period. In order to quantify this process, the characteristic parameter “self-similarity” is proposed in this paper. This characteristic coefficient shows a nonlinear decrease with aging, which is not only easy to operate, but also less dispersive and more generalizable. The establishment of the markers can realize the non-destructive assessment of the aging state of insulating paper samples in the field.