Aging Evaluation of Transformer Oil-Immersed Insulation Combining Frequency Domain Spectroscopy and Support Vector Machine

The aging-moisture synergistic effect of transformer oil-paper insulation can significantly affect its frequency-domain spectroscopy (FDS) data. It is difficult to use the traditional FDS technique to accurately assess the aging state of paper insulation under serious damp conditions. Therefore, combined with the FDS and support vector machine (SVM), a novel method of aging evaluation of transformer oil-immersed paper insulation considering the aging-moisture synergistic effect was proposed. Firstly, the oil-immersed paper insulation samples with different aging and moisture states were prepared, and the feature parameters used to characterize the deterioration state were extracted based on the FDS test. Then, a fitting database of dielectric feature parameters for training classification model of aging state is constructed based on multiple regression analysis. Finally, a classification model for aging state evaluation is constructed based on SVM and fitting database. The research findings prove that the constructed classification model can accurately evaluate the aging state of transformer oil-immersed paper insulation samples with different damp conditions. © 2021, Electrical Technology Press Co. Ltd. All right reserved.