The Method Research On Hot-spot Temperature Prediction And Location For Oil-immersed Transformer

With the vigorous construction of power gird and high-speed railway network in China,large-scale and batch oil-immersed transformers have been put into use.The hot-spot temperature(HST)is the main factor that causes the thermal aging and decommissioning operation of the transformer,which is closely related to the insulation life,overload capacity and thermal design of the transformer.The method research on the HST prediction and positioning for the oil-immersed transformer is of great significance to ensure and enhance the safe,reliable,efficient,stable and energy-saving operation of the large-scale power grid and high-speed railway network in China.For this,the following research is carried out in the paper:Firstly,in order to explore the internal thermal characteristics of disc winding in oilimmersed transformer,the CFD numerical simulation modeling of temperature calculation for winding area in an oil-immersed transformer with a rated capacity of 66 MVA is perfomerd,and the validity and accuracy of the numerical modeling method are verified compare to the existing literature study and experimental measurement.Based on the numerical simulation model,the thermal characteristics analysis of the winding area are further implemented.And the following characteristics or relationship have been proved and interpreted: ○1 The coupling relationship between the liquid velocity distribution in oil duct and the temperature distribution in the winding,the HST magnitude and location;○2 The local flow/temperature distribution in duct and the formation reason of thermal characteristics such as “ hot streak ”as well as their important influence on the disc temperature distribution;○3 The change characteristics of the “ hot area ” in the winding conditions and the distribution law of oil temperature in duct under different inlet velocity.Secondly,for the oil-immersed transformer in an oil forced and directed(OD)cooling mode,the heat transfer analysis of single-turn conductor in disc is carried out and a thermal network heat transfer model,composed of heat flux and thermal resistance,is constructed in this paper based on the basic principles of heat transfer.The conductor temperature distrbution in disc can be calculated and the hot-spot locating is achievable by the model.Compared with CFD simulation results,the accuracy and validity of the model are verified.The model can provide quantitative analysis and guidance for the thermal design of the transformer winding area.Finally,in order to realize the batch prediction of HST for traction transformer group in the high-speed railway traction power supply system.From the perspective of intelligent algorithm modeling by data driven,the HTS,load factor and ambient temperature measured by a traction transformer A equipped with fiber-optic probes are divided into training set and prediction set.Improved genetic programming intelligent algorithm is used to drive training set modeling,and an explicit expression prediction model which could directly evaluate the dynamic HST is established.For the same traction transformer and other transformers in different power supply sections and operating conditions,the longitudinal and horizontal prediction of HST is carried out,which verifies the correctness of the model and the feasibility of batch prediction.