Characterizing fatigue cracks using active sensor networks

Current identification techniques towards small-scale damage (e.g., fatigue cracks) making use of nonlinear features of acousto-ultrasonic (AU) waves are confronted by two inherent bottlenecks: (1) the inefficiency in quantitatively pinpointing the location and severity of small-scale damage, though most approaches are able to infer its existence qualitatively; (2) the use of bulky probes, moving back and forth to generate and acquire AU waves. Bearing in mind the above twofold bottleneck, a damage characterization approach, in conjunction with the use of an active piezoelectric sensor network, is reported in this chapter, based on the authors’ intensive research in this connection over the years. The reported approach characterizes individual fatigue cracks quantitatively. From fundamental modeling to experimental verification, this study has achieved insight into generation of nonlinearity in AU waves induced by fatigue cracks. A diagnostic imaging algorithm is employed to facilitate an intuitive presentation of identification results in images. Experimental validation is carried out by quantitatively evaluating multiple cracks of small dimensions in a fatigued aluminum plate, showing satisfactory accuracy. This study has led to a characterization approach for fatigue cracks in a quantitative manner using embeddable piezoelectric sensor networks. This will be beneficial to implementation of structural health monitoring able to identify small-scale damage at an embryo stage and evaluating its growth. Compared with existing methods, the developed method (1) makes use of embedded sensor networks that is conducive to online structural health monitoring; (2) evaluates fatigue cracks quantitatively; (3) enables detection of multi-cracks; and (4) presents identification results in intuitive images.