Feature Extraction and Analysis of Magnetic Non-destructive Testing for Wire Rope

In order to ensure the safe operation of wire ropes, non-destructive testing methods are being applied to inspect wire ropes increasingly. In this paper, an intelligent non-destructive testing instrument, consisting of the detection probe and signal analysis computer system, is presented. Any geometrical discontinuity in magnetic permeability in the magnetized inspection wire rope will cause the magnetic field to leak out of the rope. The flux leakage fields conceive information on anomalies, such as broken wires, abrasion, corrosions, are measured by Hall sensors. Peak-to-peak difference and phase difference of the signal are used as signal characteristic values, and are extracted using numerical algorithms. A Back Propagation (BP) neural network is proposed for the identification of defects in wire rope. Performance of the system is then verified experimentally. The discrimination of the instrument for several typical defects can reach 90%.