The influence of mechanical noise in an Acoustic Emission (AE) testing still obscures its successful application in monitoring various structures and systems. While advances in pattern recognition algorithms are helpful to differentiate relevant data from captured noise, the algorithms fail if the characteristics of relevant data are unknown. A better scientific understanding of the characteristics of elastic waves due to damage mechanisms in a structure is needed for developing a quantitative measurement approach of damage (e.g. frequency content, type, orientation) based on the AE method. The results of small scale coupon tests cannot be directly used in large scale structures due to the influence of boundaries on propagating waves detected by the AE sensors. In this paper, a steel plate is modeled using an absorbing boundary condition and spectral elements in order to understand the direct wave release from damage without influenced by reflections. The selected absorbing layer is perfectly matched layer (PML), which is designed such a way that wave reflections from boundaries back into the solution domain regardless of frequency or angle of incidence are prevented through providing a stable solution and satisfying the Sommerfeld radiation condition. The displacement-based and timedomain equations of PML are utilized. The numerical results of the two-dimensional metal plate with absorbing boundary condition are validated with extended-geometry numerical models and experimental results.
Understanding the propagating elastic wave characteristics in materials is the foundation for quantitative Nondestructive Testing methods based on wave propagation such as guided wave ultrasonic and acoustic emission.
The conventional finite element formulation requires very fine meshing and small time steps to prevent the dispersion pollution at high frequencies. The spectral finite elements reduce the required degrees of freedoms and the computation of time integration for dynamic finite element models via using high order orthogonal polynomials
to define the locations of nodal coordinates. In this study, the advantage of spectral elements over conventional finite
elements for frequencies up to 400 kHz is demonstrated on plane stress model of a structural steel plate. The excitation frequency is varied from 60 kHz to 400 kHz. The Legendre orthogonal polynomials with the orders of 3,
4 and 5 are selected. The required h refinements (i.e. element size) to eliminate the numerical error for three
polynomial orders are identified. The results provide a guide for selecting the element sizes for different polynomial
orders. The validity of the spectral element formulation is demonstrated via comparison with conventional finite
element results.
KEYWORDS: Chemical elements, Sensors, Finite element methods, Scanning electron microscopy, Wave propagation, Acoustic emission, Signal detection, 3D modeling, Numerical analysis, Matrices
The influence of mechanical noise in an AE testing still obscures its successful application in monitoring various
structures and systems. While advances in pattern recognition algorithms are helpful to differentiate relevant data from
captured noise, the algorithms fail if the characteristics of relevant data are unknown. The ability to accurately model
elastic waves using numerical methods offer a potential to understanding of frequency content of elastic waves.
However, the oscillatory nature of the wave equation requires fine meshing for a stable numerical approximation using
classical finite element models. Considering the size of civil structures, numerical modeling of full scale geometry is not
feasible. In this study, spectral element approach is implemented for modeling elastic waves in sub-scales. The transfer
function of a typical piezoelectric sensor is taken into consideration for identifying the output signal detected by the AE
sensor in relation to the input signal and the transfer function of the medium. The approach is demonstrated for 1D and
2D structures and compared with conventional finite element model using COMSOL Multiphysics program. The
comparison includes numerical efficiencies and computation times of spectral element and classical finite element.
Aluminum alloy 7075-T6 is a commonly used material in aircraft industry. A crack usually initiates at the
edge of a fastener hole, and it can affect the maintenance schedule and reduce the life of an aircraft structure
significantly. The fatigue property of the material has been researched widely to develop methods and models
for predicting fatigue crack growth under random loading. From the point of damage tolerance design, the
inspection technique of a crack for an aircraft structure is very important because it can be used to determine
the inspection period of the aircraft structure. The acoustic emission (AE) technique is a nondestructive
testing (NDT) method that is able to monitor damage initiation and progression in real time. Understanding
the early stage of AE signature due to the damage progression using small scale laboratory samples requires
non-traditional data analysis approaches. In this study, 1mm thick Al-7075-T6 plates were tested under
monotonic and fatigue loading. The initiation of damage progression using AE data was identified based on
improved linear location algorithm and the result was verified using elasto-plastic finite element model. The
improved location algorithm integrates dispersive characteristics of flexural waves and threshold independent
approach to pick up the wave arrival time. In this paper, AE results in comparison with FE model under
monotonic and fatigue loading will be presented. The comparison of traditional and improved location
approaches will be shown. The approach for implementing the laboratory scale results in the large scale field
testing will be discussed.
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.