The Accurate Detection Method For Fault Voltage Traveling Wave Based On Truncated Singular Value Decomposition Regularization Combined With Discrete Deconvolution

Since the propagation mechanism of traveling wave signals in the power grid and the interaction mechanism between traveling wave signals of different frequency bands are not clear,the existing traveling wave signal detection methods cannot accurately obtain the waveform characteristics of the primary traveling wave signal generated by the grid fault,and rely on Traveling wave positioning and protection schemes based on the characteristics of real fault waveforms cannot be practical.Therefore,it is necessary to study new methods of traveling wave signal detection to improve the accuracy of traveling wave positioning and the reliability of traveling wave protection.In order to select an appropriate method to solve the problem of inconsistent waveforms between the primary and secondary sides in the process of traveling wave detection,the paper first analyzes the detection principle of the signal from the voltage traveling wave sensor based on the preliminary research of the research group.Through experimental tests,the problem of waveform distortion generated in the process of traveling wave detection is discussed,and the physical and mathematical model of traveling wave sensor is established.Then based on the theoretical model of the traveling wave sensor and the high-frequency transmission characteristics of the voltage traveling wave sensor obtained by the analysis of laboratory equipment,the transfer function of the relatively simple concentrated model is selected as the forward model.Secondly,according to the derived forward model,the authenticity of the discrete deconvolution inversion algorithm applied to restore the waveform is studied.Aiming at the ill-conditioned problems in the process of deconvolution inversion,the truncated singular value decomposition(TSVD)regularization method is used to correct the ill-conditioned problems in waveform inversion.Finally,a voltage traveling wave inversion method combining truncated singular value decomposition regularization and discrete deconvolution is designed.This method uses the electromagnetic transient simulation software PSCAD/EMTDC to establish a transmission line simulation model,and obtains the simulated fault primary traveling wave signal by setting different fault conditions.The forward model of the special voltage traveling wave sensor built by MATLAB/Simulink can be used to obtain the simulated secondary traveling wave signal.The forward model of the special voltage traveling wave sensor built by MATLAB/Simulink can be used to obtain the simulated secondary traveling wave signal.Through the inversion algorithm that combines truncated singular value decomposition regularization and discrete deconvolution proposed in the paper,the simulated secondary traveling wave signal is inverted,and the inverted primary traveling wave signal can be accurately and effectively obtained.The simulation results show that the waveform inversion method proposed in the paper can achieve accurate inversion of traveling wave signals under different fault conditions.When using the inverted primary traveling wave signal to analyze the application of traveling wave positioning and traveling wave protection,it can effectively improve the positioning accuracy and the reliability of protection,and it is expected to promote the practical process of traveling wave technology.