Condition Based Monitoring of Superconducting Fault Current Limiter Using Fuzzy Support Vector Regression

The superconductor-triggered type fault current limiter (STFCL), which was developed by KEPCO and LS Industrial Systems, is under operation for a verification test at KEPCO's power testing center. The STFCL is composed of a superconductor, a fast switch and a current limiting resistor. In this paper, we investigated the empirical modeling of the STFCL using principal component based and fuzzy support vector regression (PCFSVR) for the prediction and detection of faults in the STFCL. Signals for the model are the currents and voltages acquired from the high-temperature superconductor (HTS), driving coil (DC) and current limiting resistor (CLR). After developing an empirical model, we analyzed the accuracy of the model. The results were compared with those of principal component based support vector regression (PCSVR) as presented in MT21. PCFSVR showed better performance in terms of the average level of accuracy. This model can be used for the condition-based monitoring of STFCL systems to predict any fault symptoms of the system through the advantage of the auto-correction function of the model.