Investigations for damage detection using the guided waves remain challenging due to the lack of an effective damage size evaluation strategy. This paper presents a sparsity-based damage detection method, which allows for damage localization and size estimation synchronously. It is demonstrated that the complex damage-reflected waves contain rich information about the damage shape, while they are not distinct and seriously overlapped, making them indistinguishable. Hence, a sparse lamb wave estimation strategy is proposed to decompose the multiple components of the damage-reflected signals. On this basis, the damage width is estimated by determining the corresponding time of flight of different reflection components. Meanwhile, the experimental validation is conducted on an aluminum plate with artificial damage, which indicates the method proposed can characterize the damage accurately.
 

Ultrasonic guided waves,damage localization,damage size estimation,sparse estimation,reflection components