Reflectometry-Based Cable Insulation Aging Diagnosis and Prognosis
Xuan Wang, Enhui Liu, Bin Zhang
Abstract
This article proposes a new diagnostic and prognostic approach that integrates joint time-frequency domain reflectometry (JTFDR), continuous wavelet transform (CWT), and particle filter to estimate the degradation of insulation and predict the remaining useful life of coaxial cables. In the proposed approach, the degradation process of the insulation is monitored by using an optimized JTFDR incident signal in an accelerated thermal aging test. CWT is adopted to obtain the time-frequency scalogram of the reflected JTFDR signals. The features are then extracted from the CWT magnitude scalogram to represent the degradation level of the insulation. A particle filter based fault diagnostic and prognostic approach is developed. A series of experiments are conducted, and the results demonstrate the effectiveness of the proposed method.