Progress in instrumentation, computer hardware, and inversion methods is encouraging the development of more advanced guided wave tomography techniques, especially for nondestructive testing of plate structures to characterize corrosion. An experimental S0 tomography performance assessment in the membrane regime is reported. One of the main interests of the fundamental membrane regime is that in this regime, waves are propagated over long distances. A 2 mm thick steel disk containing calibrated sharp artificial defects (flat bottom holes) is tested in both reflection and extinction modes. A reconstruction algorithm derived from the membrane approximation is presented. We expose a complete reflection mode inversion approach that includes beam inversion, waveform deconvolution, and thickness loss calibration. Non-linear correction factors are introduced and discussed for quantitative imaging. A width-regularity-depth description of defects is introduced to put the results into perspective with other defect geometries. The results show the relevance of the inversion method to enhance the imaging performance with regard to defect localization and sizing. Crucial points concerning instrumentation such as coupling, signal-to-noise ratio, excitation mode, coupling, selection of frequency, are also discussed.
Guided Wave Tomography is a nondestructive imaging technique that consists in inverting guided wave propagation data to localize defects. In particular, this technique should provide quantitative information about the corrosion state of metallic plates by reconstructing a thickness
map from diffraction or time-of-flight measurements. In this paper we first present an analytical framework for corrosion profile reconstruction considering the 1D case. Due to the fact that, in practice, the low frequency ultrasound range (typically 50 to 100 kHz) is used for long range inspections,
the first-order shear deformation approximation is relevant for plate thicknesses encountered in metallic structures. This leads to an analytical description of guided wave phenomena: diffraction, refraction and mode conversion, for 5 modes: A0, S0, SH0, A1 and SH1. The validity of an analytical
approach to modeling thickness loss defects, in particular the validity of the first Born approximation, is discussed by comparing with elastodynamic numerical results. The comparison results show that the nonlinear behavior with depth increase, or width increase, of the defects (distortion)
can be fully described using a multimodal high order Born series. Consequently, a consistent iterative inversion Born series based algorithm can be used to deal with the reconstruction of strong thickness losses.
Guided Wave Testing (GWT) is now increasingly being used for the non-destructive testing of large structures. Petrochemical tank bottoms are particularly subject to corrosion. Today, the monitoring of corrosion in storage tanks is probably one of the most challenging applications of GWT because of the size and the complexity of such medium. This article deals with the main physical issues of this application, in particular, the prediction of the elastic field propagating in the tank wall and bottom, and the prediction of guided wave scattering by joints. Drawing on current research work, a representative numerical configuration is defined. The global diffusive effect and the local diffraction effect of the lap joints are studied. The methodology have led us to focus on the low ultrasound frequency range: from 10 to 50 kHz. A quantitative evaluation of the scattering by lap joint is carried out by means of simulations and experiments on elementary reduced scale joints. Dynamic range computations make it possible to get prior knowledge about the minimum signal-to-noise ratio (SNR) required to get information about the tank bottom, for a given configuration, leading to a SNR-resolution trade-off. The method can be extended to larger tanks. The results confirm that the use of GWT technology for long-range active testing and imaging of storage tanks is promising.
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.