融合时变滤波经验模态分解与熵峭比的行波波头标定法OA

To address the susceptibility of traveling wave wavefront calibration methods for distribution networks to noise interference and wavefront distortion,a traveling wave wavefront calibration method integrating time-varying filter⁃ing-based empirical mode decomposition(TVFEMD)and entropy-kurtosis ratio is proposed in this paper.First,the propagation process of traveling wave signals in long and short branch lines is analyzed,revealing the intrinsic link be⁃tween branch lines and traveling wave wavefront distortion.Second,the traveling wave signals are decomposed into mul⁃tiple intrinsic mode function(IMF)components via TVFEMD,which effectively suppresses mode mixing while preserv⁃ing the high-frequency wavefront features.In addition,an entropy-kurtosis ratio is introduced to select the effective IMF component.Subsequently,the Teager energy operator is applied to calibrate the wavefront for the effective IMF compo⁃nent,thus obtaining the precise arrival time of the initial wavefront at the detection terminal.Simulation results demon⁃strate that compared with the existing methods,the proposed approach not only significantly enhances the wavefront cal⁃ibration accuracy,but also excels in processing the distorted traveling wave signals.Meanwhile,its strong noise immu⁃nity is verified,indicating that it can improve the fault location precision in distribution networks.The results in this pa⁃per provide reliable technical support for the location of complex distribution network faults.