FMCW Reflectometry Method for Diagnosis and Prognosis of Cable Insulation Degradation

This paper proposes a high-precision FMCW (frequency-modulated continuous wave)-based reflectometry method with a high-fidelity electromagnetic (EM) cable model for cable health monitoring, which aims to estimate cable insulation degradation and predict the remaining useful life (RUL). Since natural cable aging often takes many years, a cable EM model is constructed whose parameter can be changed to simulate the natural aging process of cable insulation. The FMCW incident testing signal is injected into this model with changing parameters, features that indicate the degradation are then extracted from the reflected FMCW signal. The dynamics of the degradation are then developed, based on which a probabilistic particle filtering-based state estimation method is proposed to diagnose cable insulation degradation and predict the RUL. To further verify the proposed method, accelerated thermal aging experiments of cable are conducted, and experimental results on diagnosis and RUL prognosis are presented. The experimental results show that the proposed method can achieve good performance with the help of the high-fidelity EM model.