Impact imaging of aircraft composite structure based on a model-independent spatial-wavenumber filter

The spatial-wave number filtering technique is an effective approach to distinguish the propagating direction and wave mode of Lamb wave in spatial-wave number domain. Therefore, it has been gradually studied for damage evaluation in recent years. But for on-line impact monitoring in practical application, the main problem is how to realize the spatial-wave number filtering of impact signal when the wave number of high spatial resolution cannot be measured or the accurate wave number curve cannot be modeled. In this paper, a new model-independent spatial-wave number filter based impact imaging method is proposed. In this method, a 2D cross-shaped array constructed by two linear piezoelectric (PZT) sensor arrays is used to acquire impact signal on-line. The continuous complex Shannon wavelet transform is adopted to extract the frequency narrow band signals from the frequency wideband impact response signals of the PZT sensors. A model-independent spatial-wave number filter is designed based on the spatial-wave number filtering technique. Based on the designed filter, a wave number searching and best match mechanism is proposed to implement the spatial-wave number filtering of the frequency narrow band signals without modeling, which can be used to obtain a wave number-time image of the impact relative to a linear PZT sensor array. By using the two wave number-time images of the 2D cross-shaped array, the impact direction can be estimated without blind angle. The impact distance relative to the 2D cross-shaped array can be calculated by using the difference of time-of-flight between the frequency narrow band signals of two different central frequencies and the corresponding group velocities. The validations performed on a carbon fiber composite laminate plate and an aircraft composite oil tank show a good impact localization accuracy of the model-independent spatial-wave number filter based impact imaging method. (C)2015 Elsevier B.V. All rights reserved.