Time–Frequency Fusion Algorithms for Cable Fault Diagnosis Based on Stepped-Frequency Waveform Reflectometry

The traveling-wave reflection method helps detect soft faults in cables to ensure safe and stable operation of modern industrial systems. However, the inherent drawbacks of fault location algorithms (e.g., time-domain broadening and spectral leakage) and multiple reflection (MR) seriously affect the reliability of the traveling-wave reflection method. Thus, the time–frequency fusion algorithms based on stepped-frequency waveform reflectometry (SFWR) are proposed to overcome these issues, mainly including the pulse segmentation technique and MR suppression technique. The pulse segmentation technique aims to segment SFWR pulse signals in the time domain, prior to the fault location, which can avoid the inherent drawbacks of the traditional fault location algorithms. The MR suppression technique can effectively suppress the interference of MR on fault location through the recursive relationship between single reflection (SR) and MRs. Simulations and experiments show that the SFWR time–frequency fusion algorithms exhibit good performance in different cases.