Journal of Engineering Mechanics

Ultrasonic guided-wave technology has been successfully applied to detect multiple types of defects in pipes. However, cross-sectional deformation, which is a common defect, is less studied when compared with structural discontinuity defects in pipes. In this paper, the guided wave is employed to detect cross-sectional deformation. First, the effect of sectional deformation parameters on the reflection of guided waves is analyzed using a series of three-dimensional (3D) finite-element (FE) models, and the deformation parameters affecting the reflection are examined in light of the physics of the guided waves based on the FE results. The results show that the reflection occurs at the start of the cross-sectional deformation, while the subsequent gradual deformation region does not cause reflection. The reflection coefficient is dependent on the axial deformation severity, and the mode conversion ratio is dependent on the circumferential deformation extent. Second, an experimental study was conducted to evaluate guided-wave reflection characteristics due to pipe cross-sectional deformation in a realistic situation. Test pipes with local and overall deformations were manufactured, and the reflection from both types of deformation was investigated experimentally. The results show good agreement between the experimental measurement and the FE prediction. Two quantitative parameters, axial deformation rate $\delta$ and circumferential deformation rate $\beta$, are defined to represent the cross-sectional deformation, and are found to correlate well with the reflection coefficient and mode conversion ratio. The ratio of $\delta /\beta$ is suitable to be used to judge the deformation type.

A continuum particle model correlates the interatomic potential of a crystal lattice with the elastic moduli of the solid, in which discrete atoms are modeled by perfectly bonded continuum particles, named singum, to simulate singular forces by stress in continuum. A singum particle occupies the space of the Wigner Seitz cell of the atom lattice. As the first step, a cutoff of the interatomic potential at the bond length was used to simplify the model, which was demonstrated by a two-dimensional (2D) graphene monolayer. The mass, momentum, and energy equivalence between the discrete system and continuum was investigated, and the effective elasticity of the singum was derived from the interatomic potential. The model was extended to face-centered cubic lattices and generalized to polycrystals by orientational average for three-dimensional (3D) isotropic elasticity. The deterministic relationship between the interatomic potential and singum elasticity creates a method to construct a new interatomic potential directly from the elastic moduli. Using orientational average, the isotropic elasticity of the singum for general atom lattices was obtained. When the singum potential was calibrated by polycrystal diamond, it provided a reasonable estimate of the elasticity of the 2D graphene. This simplified singum model provides a clear physical and mechanical correlation from the interatomic potential to elasticity and can be extended to long-range interatomic interactions.

This paper presents a probabilistic analysis on the stochastic failure modes of reinforced concrete beams under fatigue loading according to a series of experimental tests done previously. The analysis is featured by a recently proposed stochastic fatigue damage constitutive model for concrete material and the probability density evolution method. The simulated failure modes and deflections of the beam show good agreement with the experimental results. It is also discovered that the occurrence probability of the shear failure of the beam will increase and that of the flexural failure will decrease with the rise of the upper level of fatigue load.

In the present work, the rate-dependent stress-strain responses of limestone and dolomite, sedimentary rocks, are determined by using a split Hopkinson pressure bar device under a high loading rate. The petrographical, physical, and mechanical responses of the rocks are studied through detailed experimental and numerical analyses. For experimental analyses under a high loading rate, the intact rock specimens are prepared in five different specimen sizes to understand the effect of specimen size on the stress-strain and peak stress responses. It is seen that specimen dimension does not have a significant impact on the dynamic stress-strain response of both rocks; however, the dynamic-to-static peak stress ratio (DIF) varies between 1.8 to 8.1 at any specimen size. Because split-Hopkinson pressure bar (SHPB) specimen dimensions are small compared to any field rock mechanics problem, the rate-dependent rock mass properties are also determined for each rock using Hoek Brown criteria. The calibration of a Johnson-Holmquist (JH-2) constitutive model is done by simulating the SHPB tests on both rocks, and model parameters are determined. Blast analysis in tunnels in the two rocks is performed using the coupled Eulerian Lagrangian (CEL) tool in ABAQUS 2017, the Jones-Wilkins-Lee (JWL) equation of a state for 50 kg trinitrotoluene (TNT) explosives, and JH-2 model parameters for rocks. Stresses and displacements are reported.

To simulate the behavior of fiber-reinforced concrete, an accurate modeling of the concrete and fiber phases as well as the fiber–concrete interaction is required. This study proposes a new model to account for fiber-matrix interactions based on micromechanics. The concrete matrix is simulated using the lattice discrete particle model (LDPM), a mesoscale model for heterogeneous materials. The fiber is explicitly represented using elastoplastic beam elements. The fiber–concrete interaction algorithm features a slideline model and a constitutive model to describe the bond-slip relation. The slideline model includes a tie algorithm for interactions between concrete and slideline, and a penalty constraint between the slideline and the fiber. The proposed model is examined by simulating different types of fiber pullout tests. These simulations examine the model validity in capturing the fiber–concrete interaction both in the direction orthogonal to the fiber and in the direction of pullout. Also, the proposed model is calibrated and validated by comparing numerical simulations with experimental data from the literature, including fiber pullout under confinement, pullout of hooked fiber, and pullout of inclined fiber that leads to matrix spalling. The good agreement between the simulation results and the experimental data in terms of force versus displacement curve demonstrates the effectiveness of the proposed approach in simulating fiber–concrete interaction. In addition, using the proposed model, the contributions from different toughening mechanisms can be quantitatively compared.

The kinematic analysis method based on generalized inverse theory has been used in the engineering field. However, the traditional numerical procedure is found to be extraordinarily time consuming when the number of unknowns increases. This paper presents an improved approximation approach (IAA) of the kinematic analysis method to trace the rigid motion of mechanisms with both high accuracy and efficiency. It simplifies the procedure of obtaining the generalized inverse into the process of solving linear equations. Accordingly, the whole calculation can be executed with sparse matrix format. To demonstrate the progress of the IAA, the numerical examples are examined through a comparison with the conventional methods. The results indicate that IAA shows distinct improvements in precision and efficiency. The improvements are found to be clear and effective, especially for systems with multiple DOFs.

Viscoelastic materials are used widely for seismic mitigation and vibration suppression in civil engineering, and have obvious advantages in energy dissipation. Viscoelastic materials and devices are often used in a thermoelastic coupling state. This paper investigated the thermodynamic behaviors of a viscoelastic plate for vibration control based on the generalized thermoviscoelastic theory. The plate was fixed on both surfaces and subjected to a moving heat source on the left. The governing equations of the plate were formulated considering the fractional order derivative heat conduction, material nonlocal effect, and temperature-dependent properties. The problem was solved numerically using the Laplace transform and its inverse transformation. The distribution of the nondimensional temperature, stress, and displacement along the thickness direction of the plate were determined and graphed. The impacts of the heat source moving speed, the fractional parameter, nonlocal effect parameter, and temperature-dependent parameter were discussed and analyzed. The thermodynamic behaviors of the plate are affected greatly by the considered parameters.