Assessment of the lifespan status of power transformers based on oil analysis and frequency response using machine learning

The power transformer is a key component in any electrical transmission or distribution network. Therefore, greater attention to transformer maintenance based on the condition of the transformer is needed to increase the lifespan of the transformer and ensure maximum reliable operation. Transformer failure statistics show that most failures occur before power transformers reach their expected operational life. Transformer condition assessment methods are generally classified into three categories: chemical, electrical, and visual. In the chemical method, the condition of transformers can be determined through various types of transformer oil analysis. In the electrical method, various signals are input to the transformer at a specific time, and by analyzing the output signals, the condition of the transformer is determined. In this thesis, by using gas chromatography of oil and frequency response spectrum analysis, transformers can be classified based on the type of fault with the help of radial basis function algorithms, and equipment can be monitored based on maintenance guidelines

کلیدواژه ها:

Frequency Response Spectrum ، Gas Chromatography ، Radial Basis Function Algorithm ، Transformer Monitoring