Prediction Of Transformer Full-domain Temperature Based On Fluid Network Decoupling

In recent years,the development of China’s power grid has advanced by leaps and bounds,the power generation and power supply technology has been continuously improved,and the power consumption and system capacity are also increasing.Transformer is one of the most critical links in power network.Whether it can operate safely,reliably and stably directly affects the operation of power grid.Improper design will lead to excessive loss and local temperature.The hot spot temperature of winding directly affects the insulation life of power transformer and its long-term operation reliability.Therefore,studying the temperature distribution and hot spot temperature of large power transformers is of practical significance for power grid optimization and transformer design.Analyzing the temperature rise of oil-immersed power transformer is actually studying the bidirectional coupling process between magneticthermal-current fields.Based on the fluid network method and finite volume method,this paper takes a large oil-immersed self-cooled power transformer with a capacity of 180 MVA and a voltage level of 220 k V as an example to study the temperature distribution and energy transfer form of the oil-immersed transformer.According to the actual installation layout of components in transformer box,a three-dimensional finite element electromagnetic field simulation model is established.According to the oil circuit structure of transformer and the circulation path of cooling medium,the overall network model of oil flow is established.Based on the decoupling method of fluid network,a three-dimensional finite volume calculation model of local fluid field and temperature field is established at the partitioned decoupling nodes.Through electromagnetic field analysis,the numerical value and distribution characteristics of loss are obtained.Through the analysis of fluid network,the numerical value and flow relationship of each branch are determined.Through fluid-temperature field analysis,the characteristics of flowing oil and its influence on convective heat transfer are determined,and the temperature and temperature rise distribution of each part are obtained.At the same time,the temperature rise reacts on the characteristic parameters of materials in magnetic field and flow field,and then affects the loss distribution and fluid flow characteristics.The local temperature field considers the bidirectional coupling of field circuit and multi-physical fields,and the temperature calculation results are effectively connected by the oil flow parameters of the fluid network,so as to realize the global temperature rise prediction of the transformer..At the same time,the accuracy of global fluid network decoupling method in calculating transformer global temperature is verified by experiments,which can provide reference for transformer global temperature prediction.