DGA and Weibull Distribution Model-based Transformer Fault Early Warning

Abstract

As an important power transmission and transformation equipment, transformer fault has a great impact on the safe and stable operation of smart grid and probably incurs serious consequences. Therefore, how to detect and warn the fault of transformers as early as possible becomes particularly critical. In this paper, a fault early warning method based on DGA and Weibull distribution model is proposed for a large number of transformers in smart grid and the calculation formulas of the attention value and warning value of transformer fault are given. First, the defect rate and fault rate of transformers can be obtained by analysing the transformer maintenance data. Then, the attention value and warning value of transformer fault are calculated with the Weibull distribution model according to the gases volume distribution of oil chromatographic data, which provides an effective method for the transformer fault early warning. The actual case study shows that the proposed method can effectively achieve the transformer fault early warning.