Cable Fault Location and Signal Separation Based on Continuous Wavelet Transform

Based on the time-frequency domain reflectometry for cable fault detection, the original signal collected by oscilloscope is the superposition of reflection signal and reference signal. The signal has a certain duration in the time domain, the partial overlap of the reflected signal and the reference signal will cause measurement errors and affect the accuracy of the positioning results when a close reflection occurs. In this paper, the time-frequency reflection method based on continuous wavelet transform is used to deal with the short circuit fault signal of high temperature superconductor cable in room temperature environment in 20 meters. Continuous wavelet transform uses complex Gaussian wavelet as the parent wavelet. It obtains the position of reflection point by calculating the modulus maximum of wavelet coefficients to represent the signal similarity. It also separates the reference signal from the measurement signal according to the linear superposition of wavelet transform. The experimental results show that the method can eliminate the influence of reference signal on fault location. The maximum value of wavelet coefficient module is used to locate the fault, and the absolute error is less than 2%.