Prediction of transformer full-domain temperature based on fluid network decoupling

Abstract

In this paper, a large oil-immersed self-cooling power transformer with a capacity of 180 MVA and a voltage grade of 220 kV is taken as the research object, and the global fluid network model of the transformer is established according to the oil circuit structure of the transformer and the circulation path of the cooling medium. Based on the decoupling method of fluid network, a three-dimensional finite element calculation model of local fluid field and temperature field is established at the decoupled location. Finally, the hot spot temperature of the winding appears at the 5th layer of the high-voltage winding, which is 89.2 °C, and the lowest temperature appears at the bottom of the low-voltage winding, which is 55.1 °C. The top oil temperature is 69.2 °C, and the bottom oil temperature is 47.6 °C. The temperature experiment of the prototype shows that the relative error of the calculation result of the hot spot temperature of the high voltage winding is 4.45%, the relative error of the calculation result of the hot spot temperature of the medium voltage winding is 4.43%, and the relative error of the calculation result of the hot spot temperature of the low voltage winding is 3.58%. It proves the accuracy of the global fluid network decoupling method to calculate the global temperature of the transformer, which provides a reference for the prediction of the global temperature of the transformer.