Multi-variable approach for evaluating transformer paper insulation

In this paper, a statistical approach for evaluating the transformer's paper insulation is proposed. The proposed approach adopts multi-variable analysis to investigate the correlations between the degree of polymerization (DP) of paper insulation and increase rates of different dissolved gases in oil. Moisture in paper, oxygen and temperature are important factors influencing paper insulation degradation and included in the investigation. The reliabilities of utilizing different dissolved gases in the evaluation of DP value of an oil-pressboard insulation system are examined. The significance test of dissolved gases is also performed to find the most influential gases in the estimation of DP value. Moreover, the mutual correlation between different dissolved gases is taken into account to improve the reliability of DP estimation. Extensive accelerated ageing experiments on model oil-pressboard insulation systems are conducted and the measurement results are used to validate the proposed approach.