High resolution non-destructive evaluation of defects using artificial neural networks and wavelets

This paper presents artificial neural networks (ANN) and wavelet analysis as methods that can assist high resolution of multiple defects in close proximity in components. Without careful attention to analysis, multiple defects can be mis-interpreted as single defects and with the possibility of significantly underestimated sizes. The analysis in this work focussed on A-scan type ultrasonic signal. Amplitudes corresponding to the sizes of two defects as well as the phase shift parameter representing the distance between them were determined. The results obtained demonstrate very good correlation for sizes and distances respectively even in cases involving noisy signal data.

► We modelled signals formed by reflection of waves from multiple defects.
► We used ANN – wavelets methods to analyse the signals for defect size and location.
► Very encouraging results were obtained for both sizes and distances.
► The methods can assist with higher resolution of multiple defects in components.
► Digital signal filtering was found useful to ameliorate noise effect in the predictions.