The Automation And Quantitative Analysis For Ferromagneic Material Nondestructive Testing

It is more difficult to test the micro-cracks on the rough surface of ferromagnetic material than to test the cracks on the smooth ones by use of leakage magnetic field. The micro-crack signal tested by leakage magnetic was very difficult distinguished from the signal on rusty surface caved in. After the special differential sensor was used to test the micro-cracks, on the surface of rusty ferromagnetic rods, we could find the peak of some cracks were higher than surface caved in. The real depths of these cracks were above 0.8mm and the wide were about 0.5mm. However the signals were not identical in testing the same micro-crack repeatedly, because it is impossible to ensure that the tracks and their starting points were same every scanning and there were noise in the testing process. But there were peak scopes of testing signal for every mi. cro-crack. To determine the depth real micro-crack according these peak scopes, we must have the aid of some modern analysis method such as the wavelet theorem. The wavelet analysis for signal filter is more effective than others. Since the depth of micro-crack is a key factor for the service life of a ferromagnetic rod, it is important to determine the exact depth of a micro-crack on rusty ferromagnetic rod We can pick up all peak and wide values of the same micro-crack signal and regard them as time-dependent peak and wide signals containing the noises. Then the exact criterion of micro-crack can be got by use of wavelet analysis. Only when the peak value of one micro-crack detected, by testing system will correspond with one of the believable range of exact criterion, the depth of real micro-crack can be determine.