Research On Fault Location Of Traveling Wave Fault Of Transmission Line Based On VMD

With the development of power system in China,high-voltage transmission lines have been an important part of China's power system.However,the failure will affect the system safety,the national economy and the quality of people's lives.Therefore,the transmission line must be monitored in real time to detect and clear faults in time.In this paper,a noise elimination and distance measurement algorithm,which combines the variational mode decomposition and the morphology,is proposed in order to collect the fault signal interfered by the surrounding environment,the high sensitivity of the collector and the pulse signal.On the basis of understanding the propagation characteristics of the traveling wave signal,the variational mode decomposition is proposed to decompose the signal.The variational mode decomposition is efficient.Based on the matching mathematical theory,the complexity of the steps is low and the robustness of the noise is very good.The inevitable noise in the signal will affect the final location of the transmission line fault.In this paper,morphological filtering is used to denoise.Morphological filtering is based on mathematical morphology.It can remove the noise and retain the characteristic information of the signal.The specific step is to build a simulation model in PSCAD/EMTDC,get the fault signal and use MATLAB to process the signal.First,the new phase mode change is carried out on the voltage traveling wave signal.Then,the signal is decomposed by the variational mode decomposition.The decomposition number and penalty factor of the signal are determined by the instantaneous frequency mean of component.The adaptive structural element morphological filter is de-noised for the dominant mode after decomposition.Then,the inner gradient transform is used to transform the denoised signal to effectively enlarge the singularity of the signal.Finally,a flexible morphological edge detection method is introduced to detect the edge of the signal to improve the accuracy of the algorithm.Comparing the simulation with the data,it is shown that the morphological Variational Mode Decomposition(VMD)can effectively denoise and retain the signal mutation point for fault location.This paper also briefly introduces Empirical Mode Decomposition(EMD),instantaneous frequency and wavelet denoising principle.A signal is decomposed with VMD and EMD and the instantaneous frequency of components are calculated.Comparing the decomposition result and the corresponding spectrum,the superiority of VMD decomposition is explained.The morphological filter and wavelet denoising results are compared and analyzed.Simulation shows that the mathematical morphology filtering has better denoising and smoothness.To sum up,the method of de-noising and distance measurement with the combination of variational mode decomposition and morphology is proposed.This method is based on a complete set of mathematical theories.The principle is simpler,the operation speed is faster,and the precision of the calculation results is higher.