Contactless Monitoring Technique for Live Shielded Cable via Stepped-Frequency Waveform Reflectometry and Inductive Coupler

Low-voltage shielded cables serve critical roles in controlling and measuring advanced power systems to ensure their reliability and safety. As a consequence, it is important to monitor the cables. However, conventional monitoring techniques for nonoperating power systems are limited regarding safety issues that eventually affect the reliability of a power system and result in economic losses. In this article, a contactless signal communication approach via an inductive coupler is applied to stepped-frequency waveform reflectometry to monitor operating power systems. Additionally, a long short-term memory (LSTM)-based neural network model is adapted to compensate for the nonlinear distortion in signals yielded by couplers. Compared to conventional approaches, LSTM has advantages, as it can restore the nonlinearities of couplers because it can respond to various nonlinear factors, such as the dimension of the coupler, the length of the cable, or the position of the signal application. The proposed reflectometry system for monitoring live shielded cables is tested on 75-m 2-core shielded cables with two cases of fault simulations, and the efficacy of the proposed technique in real fields is also verified.