Analysis of the wave propagation paths in numerical reinforced concrete models
Introduction
Acoustic emissions (AE) are caused by a sudden internal release of strained energy, for example due to material fracture [1], [2]. The released energy manifests in elastic waves radiating away from the source [3]. Sensors (primarily piezoelectric transducers) record the wave motion on the surface caused by these elastic waves. The recorded signals can be used subsequently as input for various analyses (e.g., source localization; see [4], [5], [6], [7], [8], [9], [10], [11], [12], [13]). The wave propagation path is of particular interest for a majority of the AE analysis application possibilities. Methods relying on the traditional Geiger algorithm [4], [8] assume a straight wave propagation path and a constant (homogeneous) velocity model. Such methods can also be combined with a heterogeneous velocity model to achieve better localization performance in heterogeneous materials [5], [14]. However, the assumed straight wave propagation path can differ from the actual fastest wave propagation path. Therefore, wave propagation in general and estimating the fastest wave propagation path in particular is an integral component of recent studies published in [15], [16]. Since computer capacity and processor speed rapidly increase, the physical relation between AE and elastic wave propagation can be treated numerically. For example, solutions of elastodynamic equations can be calculated immensely faster and be visualized almost plastically [17]. Parallel processing and high-speed computer cluster enable the simulation of complex elastic wave propagation in different geometries and with variable material models. In concrete modeling, some promising simplifications have been made that allow taking into account heterogeneous material properties. Sophisticated non-commercial tools for the numerical simulation of elastic wave propagation arose such as the elastodynamic finite-integration technique (EFIT) [18]. The equations of motion were discretized on a numerical time domain scheme to model elastic wave propagation in isotropic, anisotropic, homogeneous and heterogeneous media, which is based on a velocity-stress formulation on a staggered-grid. The EFIT scheme was validated for isotropic, homogeneous and unbounded media by discussing dispersion relation and convergence criteria. Schubert [19], [20] reconsidered the EFIT scheme and developed a randomly distributed concrete model that is employed in a modified version in recent scientific works to be shown later. Saenger et al. [21] presented a modified version of the classic velocity-stress formulation, discretized on a staggered-grid with finite differences (FD) method [22], [23]. Kocur et al. [24] applied the rotated staggered-grid FD method to simulate elastic wave propagation in numerical models obtained from X-ray computed tomography (CT) of real uncracked and cracked concrete cuboids [25]. Synergies were illustrated between numerical simulations in X-ray CT data for a concrete specimen and a self-developed purely numerical concrete model. The influence of concrete constituents such as aggregate grains and air voids on the wave propagation behavior was studied.
Section snippets
Problem statement
In the past decade, the employment and advancement of innovative ultrasonic testing methods such as AE analysis has attained noticeable attraction. Further progress of AE analysis, and in particular successful AE source localization, depends on how the wave propagation behavior is modeled and taken into account. Air-filled cracks represent impenetrable barriers for elastic wave propagation in solids; waves have to bypass the crack. The most common source location estimation methods (i.e.,
Elastic wave propagation in solids
In an isotropic and homogeneous infinite space, elastic waves emanated from a point-like source propagate spherically along a straight path (see Fig. 1a). In solids, pressure waves (P-waves) and shear waves (S-waves) occur, which independently propagate with their characteristic wave velocities and respectively [3]. Shear waves do not occur in ideal gases (). The wave velocities can be written in terms of the Young’s modulus density and dynamic Poisson’s ratio as
Numerical simulations
Numerical wave propagation simulations are an excellent tool to investigate and illustrate the interaction of waves with boundaries. In order to study the wave propagation behavior of elastic waves in reinforced concrete, numerical simulations using a staggered-grid finite-difference technique in the time domain [22], [23], [30], [31] were performed. In particular, a rotated staggered-grid scheme [21] was used. One of the attractive features of the staggered-grid approach is that the various
Numerical setup
The aim of this application is to corroborate the findings on the wave propagation paths to achieve a better performance with the FastWay algorithm compared to both Geiger methods. A simplified numerical model of a reinforced concrete cuboid ( mm) as shown in Fig. 12 was implemented. The concrete was modeled as a homogeneous material (single-component NCM). The concrete cuboid was reinforced with a single Ø30 mm steel bar. A notch of 10 mm width across the entire width of the
Discussion and conclusions
Contrary to both, the homogeneous (original) Geiger and the heterogeneous Geiger method, the FastWay algorithm is capable of considering the influence of cracks on wave travel paths as long as they are represented in the velocity model (i.e., the velocity matrix ). The knowledge of the exact (and fastest) propagation paths leads to significant improvement of the accuracy of estimated source location in general (see sources S2 and S4 in Fig. 15). In particular, the improvement is evident if
CRediT authorship contribution statement
Stephan Gollob: Conceptualization, Formal analysis, Investigation, Writing - review & editing. Georg Karl Kocur: Conceptualization, Writing - review & editing.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgments
The entire research was carried out at the Institute of Structural Engineering of the ETH Zurich in Switzerland. This research project was funded by the Swiss National Science Foundation SNF (grant number 200021_153371). Constructive anonymous reviewers have improved the quality of this work.
References (40)
- et al.
Acoustic emission for monitoring a reinforced concrete beam subject to four-point-bending
Constr. Build. Mater.
(2007) - et al.
Toward a probabilistic acoustic emission source location algorithm: a bayesian approach
J. Sound Vib.
(2012) Acoustic source localization
Ultrasonics
(2014)- et al.
Time reverse modeling of acoustic emissions in a reinforced concrete beam
Ultrasonics
(2016) - et al.
Defect mapping in pipes by ultrasonic wavefield cross-correlation: a synthetic verification
Ultrasonics
(2018) - et al.
A novel multi-segment path analysis based on a heterogeneous velocity model for the localization of acoustic emission sources in complex propagation media
Ultrasonics
(2017) - et al.
Numerical modeling of elastic wave propagation and scattering with EFIT - elastodynamic finite integration technique
Wave Motion
(1995) Numerical time-domain modeling of linear and nonlinear ultrasonic wave propagation using finite integration techniques–theory and applications
Ultrasonics
(2004)- et al.
Modeling the propagation of elastic waves using a modified finite-difference grid
Wave Motion
(2000) - et al.
Elastic wave propagation in a segmented X-ray computed tomography model of a concrete specimen
Constr. Build. Mater.
(2010)
Application of time reverse modeling on ultrasonic non-destructive testing of concrete
Appl. Math. Model.
Numerical methods to determine effective elastic properties
Acoustic Emission and Related Non-destructive Evaluation Techniques in the Fracture Mechanics of Concrete
Acoustic Emission Testing, Basics for Research - Applications in Civil Engineering
Wave Propagation in Elastic Solids (Applied Mathematics and Mechanics)
Herdbestimmung bei Erdbeben aus den Ankunftszeiten, Nachrichten der K. Gesellschaft der Wissenschaften zu Göttingen
Math.-Phys. Klasse
Analysis of source location algorithms, Part I: overview and non-iterative methods
J. Acoust. Emiss.
Analysis of source location algorithms, Part II: iterative methodes
J. Acoust. Emiss.
Active source localization in wave guides based on machine learning
Ultrasonics
Cited by (11)
Optimal arrangement of AE sensors for prestressed hollow slabs based on AE propagation characteristics
2024, Journal of Building EngineeringCluster-based sensor selection framework for acoustic emission source localization in concrete
2023, Measurement: Journal of the International Measurement ConfederationEnergy reflection and transmission characteristics of acoustic waves in a functionally graded piezoelectric plate immersed into non-viscous fluid
2022, Journal of Sound and VibrationCitation Excerpt :The knowledge of the reflection and transmission behavior of acoustic waves at the boundary between piezoelectric materials and water is a major problem in designing acoustic transducers, which in most cases are piezoelectric ceramics, for use in under water imaging [10]. The transmission and reflection of plane harmonic acoustic waves in isotropic media and plates [11,12], multilayered structures [2,6,13-18], functionally graded panels [19], piezoelectric [20,21] and magnetoelectroelastic [22] structures have been investigated earlier. The transmission and reflection characteristics of acoustic waves in various structures immersed into viscous [23] or nonviscous [24–26] fluid were also studied.
Research on structural sound source localization method by neural network
2023, Eurasip Journal on Advances in Signal Processing