Detection and localization of corrosion using the combination information of multiple Lamb wave modes

Themulti-modeanddispersionnature of Lamb waves means that a variety of modes with individual mode structures and distinct dispersion behaviors would propagate in the structures simultaneously. The existence of a corrosion would result in thickness reduction, which means the frequency-thickness product under a specific excitation would also decrease. Due to dispersion diversity, the interaction of each individual Lamb mode at the corrosion may be distinct, i.e., the velocity varies in different extent and even in opposite trends. In this paper, the combination of multiple modes, rather than a single sensitive mode, is used for structure diagnosis. Specifically, two Lamb modes both sensitive to corrosion but with opposite variation trends are taken and the corrosion index is defined on the ratio of their time-of-flight, so as to enhance the sensitivity to corrosion and eliminate the influence of difference path lengths in the sensor network. On this basis, a probabilistic reconstruction algorithm is introduced for corrosion detection and localization. Since the two modes are extracted from the same wideband Lamb wave response, the proposed method is baseline-free. The influence of mode conversion on the effectiveness of the proposed method is discussed. Ultimately, the performance of the proposed method is demonstrated by an experimental example. The results show that the defect could be correctly identified and accurately localized.