Application of Flux-Coupling-Type SFCLs in a Power System Considering Fault Current Prediction Based on SAO-LSTM Model

Abstract

The improved flux-coupling-type superconducting fault current limiter (FC-SFCL) is a favorable option to fulfill the short-circuit current suppression in a power system, due to its fully-controlled features and two-level current-limiting modes. This paper explores the application of the FC-SFCLs in the power system, where the accurate fault current prediction is realized by the SAO-LSTM model. By integrating the snow ablation optimizer (SAO) and the long short-term memory (LSTM) network, the SAO-LSTM model is capable of extracting the fault current characteristics and reflecting the long-term historical process to forecast the fault current. Thus, the FC-SFCLs may select the appropriate current-limiting modes to handle the faults. The theoretical description of the FC-SFCL is conducted, and the fault current prediction method based on the SAO-LSTM model is expatiated. Using MATLAB, a modified IEEE 13-node system equipped with the FC-SFCLs is modeled. Different fault locations, fault types, and fault resistances are considered to build the dataset, and the prediction performance of the SAO-LSTM model for minor and severe faults is checked. The findings show that the SAO-LSTM model can effectively identify the severity of the faults, and the current-limiting efficiency of the FC-SFCLs under different fault severities and phase angles can be well exploited.