Research On Transient Two-Dimensional Fluid-Thermal Coupling Hybrid Numerical Method For Oil-Immersed Transformer

Oil-immersed power transformer is one of the important equipment in power system,whose safe and operation are very important to the power system.If the local temperature of the transformer winding is too high,it will accelerate the insulation aging of the transformer,affect its stable operation and significantly shorten the service life.Therefore,it is very important to analyze the transient temperature rising of the winding and predict the hot spot temperature accurately during the operation of the transformer.The analysis of the transient fluid-thermal coupling problem in the operation of oil-immersed power transformer is important to analyze the winding temperature rising and predict the hot spot temperature.In order to study the temperature variation characteristics of oil-immersed power transformer windings during operation,a fluid-thermal coupling method is proposed to calculate the transient temperature rise of transformer windings.The dimensionless least square finite element method(DLSFEM)is used to calculate the velocity distribution of the flow field of the transformer windings at different times,and the upwind finite element method(UFEM)is used to calculate the temperature distribution of the windings at different times.Finally,the sequential iterative method is used to carry out the transient fluid-thermal coupling calculation.The physical parameters of transformer oil and the temperature effect of winding loss are considered in the calculation.Compared with the traditional least square finite element method,the stiffness matrix formed by DLSFEM has smaller condition number,and the discrete equation has better convergence.Taking a local winding model of an oil-immersed power transformer as an example,the hybrid algorithm and the commercial software Fluent were used to calculate the transient temperature distribution of the windings,and the effectiveness and efficiency of the hybrid algorithm were verified.In order to solve the problems of too many degrees of freedom,too many computational resources and too long computational time when solving the transient fluid equation by numerical method,an proper orthogonal decomposition(POD)order reduction method based on singular value decomposition(SVD)is adopted in this paper.First,DLSFEM is used to calculate the velocity solution of the flow field in an initial period of time.Then,singular value decomposition is performed on the matrix formed by the velocity solution,and a set of orthogonal basis reflecting the main characteristics of the known data is extracted.Finally,Galerkin orthogonal projection method is used to construct a reduced order computational model for solving the transient flow field equation.The order reduction method is applied to the calculation of local winding model,and the calculation results show that the difference between the calculation results of the reduced order model and that of the full-order model is very small,and the calculation speed of the reduced order model is faster than that of the full-order model,which verifies the effectiveness of the order reduction method.In order to verify the effectiveness of the combination of the mixed finite element method and the order reduction technology proposed in this paper in practical engineering,this paper aims at an experimental platform of a large oil-immersed transformer,and collects temperature data varying with time through an experimental measuring device.The winding two-dimensional simulation model was established according to the experimental platform,and the experimental parameters were set into the simulation model.The hybrid algorithm proposed in this paper and the Fluent software were respectively used for numerical simulation calculation.By comparing the simulation data with the experimental measurement results,the accuracy and high efficiency of the proposed hybrid algorithm in practical engineering are verified.