Method of Wire Insulation Fault Location Based on Wavelet Transform
Zhang Junmin1, Zhou Xiaomeng1, Wei Juan1, Cao Dashu2, Yao Hongyu2
1. Beijing University of Aeronautics and Astronautics Beijing 100191 China 2. Center of Aviation Safety Technology of China General Administration of Civil Aviation of China Beijing 100028 China
Abstract:A mathematical model of aerospace wire insulation faults is presented based on time domain reflectometry (TDR) in this article. Transmission in a defective lossy is simulated for step wave of different rise time. The results show that steep leading edge of incident signal is conducive to find out small insulation faults, and simultaneous attenuation of high frequency signals results in smooth reflected wave, which leads to errors in locating the starting point of peaks corresponded to faults. A TDR test platform is built to detect small insulation faults. Soft threshold de-noising method of wavelet is used to eliminate noise in received signals, and then the starting point of peaks corresponded to faults is ascertained by continuous wavelet transform. Test results indicate that wavelet method helps to identify failure points, and to improve the accuracy of faults location. Minimum length of faults found in experiment is 5cm and positioning error is less than 2%.