Algorithm Design on Filtering Signals in Cable Defect Magnetic Flux Leakage Non-destructive Testing System

摘要

Cable defect magnetic flux leakage non-destructive testing system is designed according to the characteristics of cable-stayed bridge cable nondestructive testing. The measured results show that the defects of the magnetic flux leakage signals are often accompanied by a lot of noise signals, which leads to signal characteristics no significant. In filtering signal algorithm, first to do the smoothing and filtering processes, then to obtain the signal by characteristic differential of limitation after quantization process characteristic, at least to acquire the object internal defects by the BP neural network. Simulation results show that after dealing with the filtering algorithm of magnetic flux leakage signal characteristics significantly, which provides a reliable basis for the nondestructive testing and analysis of the bridge cable wire brake defect.