Author(s)

Moyun Wu, Lin Yuan

Corresponding Author

Moyun Wu

Abstract

Power transformers are important electrical equipment in power systems. Research on power transformer fault diagnosis methods is of great significance for early detection of potential transformer faults and improving the safety of power systems. In view of the complicated mechanism of transformer faults, this paper focuses on extracting state features with the help of transformer vibration signals, partial discharge signals and dissolved gas content data in oil, and combining machine learning theory to study transformer fault diagnosis methods. The paper proposes a method of using local waves to divide the vibration signal of the transformer body. In view of the fact that the vibration signal of the transformer body can effectively reflect the condition of the internal windings and iron core of the transformer, the local wave method is applied, and the vibration mode in the signal can be used to better understand the characteristics of the vibration signal when a fault occurs. Determine whether the transformer is faulty.

Keywords

Power transformer, Fault technology, Electricity level