Composite materials are becoming more popular in the aerospace industry, because of their physical properties. In quality assurance and in-service inspection, there is a need for fast, non-contact, high quality, non-destructive inspection techniques. The most common approach is to perform the inspection using water-coupled high frequency transducers. Full wavefield techniques are promising to replace the conventional inspection approach. However, these are currently performed by a laser vibrometer setup, which has drawbacks. As an alternative, a low-cost MEMS sensor array and dedicated processing scheme are presented enabling fast inspection of large samples. This inspection approach uses a piezoelectric actuator to excite the composite or metallic part with Lamb waves. An array of MEMS sensors records the energy that radiates into the surrounding air. A dedicated processing scheme will translate the measured wavefield into a thickness map of the inspected part. For composite parts, material's anisotropy needs to be taken into account for accurate thickness mapping. In principle all relevant defects show up as local thickness reductions. The results in this paper are obtained with a MEMS-sensor array of 128 elements capable of detecting ultrasound up to 250 kHz at a typical stand-off distance of 100 mm. Defects up to 6 mm in diameter could be detected in thick panels, and defects as small as 2.5 mm could be detected in thin panels. A full-size fuselage experiment shows that the method is also suited for fast inspection of large inspection areas.
It is well known that source deghosting can best be applied to common-receiver gathers, while receiver deghosting can best be applied to common-shot records. The source-ghost wavefield observed in the common-shot domain contains the imprint of the subsurface, which complicates source deghosting in common-shot domain, in particular when the subsurface is complex. Unfortunately, the alternative, i.e., the common-receiver domain, is often coarsely sampled, which complicates source deghosting in this domain as well. To solve the latter issue, we propose to train a convolutional neural network to apply source deghosting in this domain. We subsample all shot records with and without the receiver ghost wavefield to obtain the training data. Due to reciprocity this training data is a representative data set for source deghosting in the coarse common-receiver domain. We validate the machine-learning approach on simulated data and on field data. The machine learning approach gives a significant uplift to the simulated data compared to conventional source deghosting. The field-data results confirm that the proposed machine-learning approach is able to remove the source-ghost wavefield from the coarsely-sampled common-receiver gathers.
The estimation of the upgoing wavefield based on a ghost model is well-known as receiver deghosting. The shape of the rough and dynamic (time-varying) sea surface and the varying receiver depth should be included to carry out this step correctly. However, the exact shape of the dynamic sea surface is not commonly measured. Therefore, the ghost model is often inaccurate, introducing artifacts in the deghosted result. We have developed a data-driven, iterative, receiver deghosting method that includes the estimation of the dynamic shape of the sea surface and the varying receiver depth. The deghosting algorithm is based on, respectively, wavefield propagation from the varying receiver level up to the dynamic sea surface and wavefield propagation from the dynamic sea surface down to the varying receiver level. The depth of the receivers relative to the height of the dynamic sea surface is estimated and updated at each iteration. We validate the deghosting method on 2D simulated data recorded on a variable depth and 2D simulated data recorded with a dynamic sea surface. Our deghosting method gives a significant uplift on the simulated data given that the initial ghost model is inaccurate. On the field data, recorded with a midwater, quasistationary cable, our deghosting method is able to correct the receiver depth and provides an accurate deghosting result. The performance of the deghosting method is similar to the performance of a multicomponent decomposition method in this realistic scenario.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.