The aim is to evaluate the effectiveness of relative methods used to locate faults of 6–10 kV power cables operated at an industrial enterprise. To achieve this goal, the reflectograms obtained on damaged power lines were compared with the reference ones. The research object is represented by three 6–10 kV power cables with paper-oil insulation, 900–3000 m long, and with a service life of more than 30 years. The type of damage to all power lines was a single-phase creeping discharge. The study employed various methods (TDR, ARC, ICE/Decay) to locate faults in the lines and the test rigs. The equipment used to detect cable fault locations included a REIS-305 reflectometer, a GVI 24.3000IDM high-voltage pulse generator, and a SDC50 current sensor. Multiple soundings of the power lines under study were performed to obtain reflectograms containing information about the distance to the fault location. As a result of the conducted research, varying in informativeness reflectograms of power cables were taken. We managed to select the true ones from them. Serious discrepancies in the shape of the displayed sounding pulse in the reflectograms were identified in comparison with the reference ones given in relevant technical literature. The article highlights issues concerning the application of the ARC method. The challenges consist in the breakdown voltage limitation at the cable faults. Each cable fault is shown to have its own characteristics. However, their causes are difficult to identify since it is impossible to establish the process of and the prerequisites for electric current passage when a short circuit arises. Drawing on the results of the undertaken study, a recommended algorithm was developed to determine the distance to the location of a power cable fault. However, due to the small amount of scientific research on this subject matter, no absolutely universal approach to this problem exists at present; its creation requires further investigation.
Zhilin et al. (Sun,) studied this question.