Modified minimum variance imaging of Lamb waves for damage localization in aluminum plates and composite laminates

Abstract Lamb wave minimum variance imaging is a promising method for visual damage identification and localization with a sparse transducer array. Imaging performance of minimum variance is highly dependent on the design accuracy of look-direction to describe amplitude relationship of array reflection signals. Look-direction is the combination of a directivity reflection pattern and an attenuation with propagation distance. However, reflection pattern is closely related to damage parameters (e.g. type, orientation and size) and these parameters are usually unknown beforehand. Therefore, accurate design of look-direction is difficult or even impossible, and design error can significantly degrade imaging performance. To overcome this limitation, a modified minimum variance method is proposed in this study. Besides amplitude information, waveform information is integrated into algorithm imaging procedure. Correlation coefficient between local signal and excitation waveform is calculated to generate the distribution of weights for diagonal loading. Diagonal loading weight is an adjustable coefficient in minimum variance algorithm to control the tolerance for look-direction error. With larger weights for potential damage locations, tolerance for inaccuracy in look-direction is increased, and imaging performance is accordingly improved. Experiments on both aluminum plates and composite laminates are carried out to demonstrate the performance improvement of modified method.