Simulation And Experimental Study On Deformation Detection Of Power Transformer Winding Based On SFRA Method

With the rapid development of the global economy,there is an increasing demand for the construction of a strong power grid.How to ensure the safe and stable operation of power grid equipments has become a research hotspot in electric power science and technology field.Power transformer is one of the most important power transmission equipments in power system.The winding deformation caused by the excessive short circuit electromagnetic force and collision of transportation is one of the main causes of transformer failure.Therefore,it is of great significance to detect the state of transformer winding and to study the relationship between the frequency response curve characteristics and fault conditions.In this paper,three kinds of faults such as : the short circuit fault between the disks,the space changing between the disks and the radial deformation are analyzed based on the special model transformer.Characteristics of frequency response curves under different fault conditions are studied.The statistical parameters are used to analyze the experimental data of fault winding.At the same time,in order to simulate the transformer winding fault type which is difficult to realize in the experiment in the further study,a simulation method of transformer winding based on finite element method is proposed.The main achievements of this paper are as follows:(1)The modeling method of lumped parameter circuit of power transformer winding is presented.3D / 2D geometric models of transformer are set up in ANSYS Maxwell software.A method for calculating the equivalent longitudinal capacitance of continuous and interleaved transformer winding based on the single pancake is deduced.Combining the model and the algorithm,the capacitance,inductance and resistance parameters of the equivalent circuit are obtained.The equivalent lumped parameter circuit of the high voltage winding is built and simulated to obtain the sweep frequency response signature.In order to verify the correctness of the circuit model and the calculation method,off-line tests are carried out on the healthy and faulty windings respectively.(2)The changing trend of frequency response curves under different fault conditions of power transformer winding is revealed.For different fault locations and degrees,the changing characteristics of frequency response curves are obtained based on the results of experiments.(3)Analysis of test data based on statistical parameters.Based on a large number of transformer winding sweep frequency response test data under the healthy and faulty states,the parameters such as frequency offset,amplitude variation,mean deviation,mean standard error,correlation coefficient,covariance and so on are extracted to analysis the trend of frequency response curves.It is found that the identification of the mean square error parameter is better for the experimental data.This can provide more reference for the fault diagnosis and pattern recognition of transformer winding.