Research Of Identication And Inhibit Strategy Of Transformer Excitation Inrush Current

The power transformer plays an important role in the whole power system,which is the key point of voltage rise and fall.It is the key interconnection point and power transformation hub in the power system.Whether the power transformer can operate stably and correctly is directly related to the power supply reliability of the whole power system.Once a fault occurs,it will cause huge interference to the whole power system and bring inestimable loss.For the transformer differential protection can be rapid and accurate identification and suppression of inrush current has been a key reason for the misoperation of transformer protection,therefore further research on power transformer protection,on the basis of this,the more scientific method of recognition and suppression of inrush current,to improve the correct rate of transformer protection has important significance.On the basis of consulting the recognition and suppression methods of magnetizing inrush current at home and abroad,the advantages and disadvantages of various methods are compared and classified reasonably.MATLAB/SIMULINK software is used to build a simulation model of three phase transformer under the condition of unload closing and internal short circuit.The identification of inrush current based on support vector machine(SVM)is proposed,based on the difference of the waveform characteristics between the transformer and the internal fault.First,by collecting the waveform of magnetizing inrush current and internal fault current,we use phase space reconstruction to get the phase space trajectory map of current waveform,and combine fractal geometry to quantify the phase space characteristics of current waveform,and get the eigenvalues of current waveform phase space.Two dimensional eigenvectors of current waveform phase space and current time series eigenvalues are used to identify the characteristic parameters of transformer inrush current,and a two-dimensional feature vector database is built to identify the inrush current of transformers.Finally,the nonlinear classifier system is established by using the pattern recognition function of support vector machine,and the sample data is trained and tested by the established nonlinear classification machine model,and the correct recognition rate is achieved by adjusting the parameters of the model.The simulation experiment shows that the method can distinguish the internal fault current and inrush current completely.Inrush identification of transformer is a problem to solve the problem that the transformer can not be closed because the current is too large in the angle of "hiding".However,due to the various hazards of the inrush current,the inrush current must be solved in essence.Based on this,on the basis of the specific principle of the transformer,the remanence measurement combined with the selection of correlation technology to suppress the inrush current.First,according to the transient process of transformer closing and the flux characteristic of iron core when closing,the optimal closing time of transformer inrush current suppression is analyzed and expounded.Then deduce the specific remanence calculation formula,build the simulation model of transformer residual magnetism measurement by using MATLAB software,measure the residual magnetism of transformer core after breaking the circuit breaker,and calculate the best closing time of phase selection based on the result.On the basis of this,the effectiveness of the method is verified by simulation.