A Novel Method for Cable Defect Location Based on Reflection Coefficient Spectral Variational Mode Decomposition

Frequency domain reflectometry (FDR) based on reflection coefficient spectrum (RCS) is a fast, effective, and non-destructive method for cable defect location. However, due to the impedance mismatch between the tested cable and the signal connection line, as well as multiple reflections of the testing signal within the cable, spurious peaks frequently appear in the location curve, which will affect field judgment and limit the practical applications of FDR. In order to address this issue, an algorithm based on reflection coefficient spectral variational mode decomposition (RCS-VMD) is proposed to improve the conventional cable defect location method based on FDR. The RCS-VMD algorithm determines the order of the RCS decomposition components and identifies the components to be removed based on the principles of VMD. The Pearson correlation coefficient is also introduced as a criterion for removing and reconstructing the VMD components. The simulation and experimental results demonstrate that the improved method can effectively distinguish and eliminate abnormal frequency components caused by the aforementioned factors, thereby suppressing corresponding spurious peaks in the location curve. The peak values of the spurious peaks caused by impedance mismatch at the test terminal and multiple reflections can be reduced by over 90.02%.