his paper presents results from an investigation of localized geometrical effects on crack-induced acoustic emission in welded plate-like structural components. The aim is to better understand the influence of weld geometry and precracking on the radiation patterns of guided waves. Attention is confined to the fundamental symmetric (S0), antisymmetric (A0), and shear-horizontal (SH0) guided modes. Numerical modeling is used to study cases of butt welds, T-joint welds, and finite-length full-penetration precracks. The numerical model is referenced against an analytical model which neglects these geometrical effects. Butt welds are shown to have a relatively small, and perhaps negligible, influence on radiation. T-joint welds exert a much stronger influence, especially behind the joint. Ahead of the joint, the S0 and SH0 modes are less affected. In the precrack case, mode conversion from a Rayleigh mode into the SH0 mode is shown to enhance primary radiation behind the precrack. An SH0 metric is therefore introduced, which shows crack-sizing potential up to about five plate-thicknesses.
KEYWORDS: Corrosion, Acoustic emission, Data modeling, Data analysis, Diagnostics, Visualization, Data acquisition, Sensors, Structural health monitoring
The potential of topological data analysis (TDA) to aid acoustic emission (AE) in revealing early signs of corrosion in prestressed concrete has recently been demonstrated by the authors. This paper serves to extend the experimental investigation of this structural health monitoring potential. The topological method was evaluated in accelerated corrosion testing of control and weathered prestressed concrete specimens. The results highlight the potential of TDA to aid in extracting corrosion information from AE data. Further, with TDA aiding traditional AE monitoring, there is potential for early and reliable indication of concrete cracking, prior to the appearance of external visual signs. In addition, the results demonstrate the potential generalizability of the method toward existing in-service prestressed concrete structures. Lastly, the AE-based corrosion indicators are combined with a hidden Markov modeling framework. The capability of the framework for automated corrosion diagnostics is demonstrated through training and testing between the two specimens.
This paper presents solutions for guided wave motion (Lamb and shear horizontal) due to tensile and shear cracks in an isotropic plate using elastodynamic reciprocity. Finite-length through-thickness cracks are considered via Huygens’ principle by representing them as a superposition of point cracks. Far-field solutions are then derived in order to simplify the results and facilitate a direct comparison of guided mode excitability due to various cracking modes. Relatively short- and long-length line cracking are compared to point cracking for the fundamental modes S0, A0, and SH0. It is shown that the A0 modal response is the most sensitive to crack length, with S0 and SH0 being relatively insensitive. Additionally, the radiation patterns of S0, A0, and SH0 are relatively insensitive to crack length. The results have applications in acoustic emission monitoring of plate-like structures, where modal responses may be used to characterize crack growth.
This study focuses on localizing and characterizing acoustic emission (AE) sources in metallic panels with rivetconnected doublers. In particular, a deep learning-based framework is proposed that first performs zonal localization with only one sensor and then depending on the zone in which the source occurs, either finds the coordinates of the source or characterize it based on its source-to-rivet distance. The performance of the framework is assessed in typical scenarios in which the training and testing conditions of the deep networks are not identical, and Hsu-Nielsen sources were carried out for validation.
This paper proposes a guided wave-based approach for monitoring stress redistribution in prestressing strands under corrosion. The stress dependence of wave velocity is leveraged for stress measurement, while targeting advantageous frequencies of higher-order wave modes to eliminate the geometric effects of corrosion. For practical longterm monitoring scenarios, where sensor reattachment/replacement may introduce artificial noise, a technique for eliminating such effects (called modal modulation) is also proposed. To demonstrate the approach, accelerated corrosion testing was carried out on a strand while actively generating and receiving higher-order modes. The strand was subjected to 29 cycles of accelerated corrosion (reaching 45% mass loss), with the 29th cycle terminated at the simultaneous fracture of three peripheral wires. The measurements from several higherorder modes were processed into a single estimate using a data fusion approach. The data fusion estimate showed good agreement with the measured stress values even under significant surface roughness (up to 20% mass loss), and especially under a large stress increase due to fracture. To evaluate the modal modulation technique, stress estimations obtained without applying the technique were also shown, which yielded incoherent results.
This paper presents a numerical approach based on spectral methods for the computation of guided ultrasonic wave modes in stressed elastic plates and rods. The approach is applicable to Lamb modes in anisotropic plates and longitudinal modes in isotropic rods under uniform stress. The proposed approach computes the modeshapes and the full complex dispersion spectrum (real frequency, complex wavenumber), accommodating both propagating (real wavenumber) and non-propagating (complex wavenumber) modes. Numerical results are presented for plates composed of fiber-reinforced graphite/epoxy (GREP) and plates and rods composed of Hecla 17 steel. The results are used to investigate and compare the effect of stress on the dispersion curves for plates and rods, while demonstrating the computational efficiency of spectral methods. The convergence rate is demonstrated, showing spectral convergence.
A sparse reconstruction localization method is proposed, which is capable of localizing multiple acoustic emission events occurring closely in time. The events may be due to a number of sources, such as the growth of corrosion patches or cracks. Such acoustic emissions may yield localization failure if a triangulation method is used. The proposed method is implemented both theoretically and experimentally on large diameter thin-walled pipes. Experimental examples are presented, which demonstrate the failure of a triangulation method when multiple sources are present in this structure, while highlighting the capabilities of the proposed method. The examples are generated from experimental data of simulated acoustic emission events. The data corresponds to helical guided ultrasonic waves generated in a 3 m long large diameter pipe by pencil lead breaks on its outer surface. Acoustic emission waveforms are recorded by six sparsely distributed low-profile piezoelectric transducers instrumented on the outer surface of the pipe. The same array of transducers is used for both the proposed and the triangulation method. It is demonstrated that the proposed method is able to localize multiple events occurring closely in time. Furthermore, the matching pursuit algorithm and the basis pursuit densoising approach are each evaluated as potential numerical tools in the proposed sparse reconstruction method.
This paper presents an imaging technique to locate damage in plate-like structures by permanently attached piezoelectric transducers (PZT) capable to generate and receive guided ultrasonic waves. The technique is based on a model capable of predicting envelope of scattered waves. Correlations between the estimated scattered waves and experimental data are used for image reconstruction. The approach is validated on an aluminum plate and results are compared with two common imaging algorithms, that is, Delay and Sum (DS) and Minimum Variance (MV). Damage is simulated by placing two magnets on sides of the plate. It is shown that the inclusion of Lamb wave reflections improves the localization accuracy while making use of fewer number of sensors possible.
Helical guided waves in pipelines are studied under the effects of pressurization stresses from a contained liquid. The pipeline is approximated by an “unwrapped” plate waveguide, and a transfer matrix method is used to solve for guided wave velocity and attenuation dispersion curves in a multilayered plate waveguide subject to an arbitrary triaxial state of initial stress. The matrix-based model is able to incorporate both elastic and viscoelastic solid materials, as well as approximate non-uniform distributions in initial stress through the thickness of a waveguide. Experiments on a steel pipe filled with pressurized water are carried out to validate the modeling approach.
In this paper an algorithm for acoustic emission source localization in cylindrical shell structures is presented. The proposed algorithm is based on the propagation of uncertainty through the Unscented Transform. Time of arrival of desired wave modes and wave velocity are measured parameters, whose uncertainties are processed through the algorithm, which provides mean and covariance statistics for the predicted location. Results of the algorithm using the Unscented Transform are compared to a Monte Carlo simulation, and this is accomplished through the Kullback-Leibler divergence. The results support a strong correlation between the two, however, the Unscented Transform demonstrates superior computational speed.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.