Time-Frequency Domain Reflectometry for Live HTS Cable System via Inductive Couplers and Neural Network

Abstract

High-temperature superconducting (HTS) cable has become one of the most promising technology along with an adaptation of a unique superconducting conductor that allows long-distance transmission of the large volume of electronic power with less power loss. Diagnostic and monitoring technique for the reliable and safe operation of HTS cable system has also become a crucial factor to commercialize the HTS cable system in the regarding field. HTS cable system which needs a cooling process for re-activating superconductors is more affected by the loss of cost and time. This paper is proposing the time-frequency domain reflectometry (TFDR) based diagnostic technique for a live HTS cable system. By adapting inductive couplers, the live HTS cable system and the diagnostic signal would communicate without direct contact. Signal attenuation and distortion during the signal induction are compensated by LSTM-based neural network. The proposed method is applied to an AC 154 kV 600 MVA HTS cable system in and is further verified through Advanced Design System (ADS) software.