Journal of Engineering Mechanics

The Vlasov–Leont’ev foundation model describes the soil displacement field as a product of separable functions and utilizes the calculus of variation to obtain the governing differential equations describing the motion of the structure/substructure and the soil. In this paper, the model is employed to solve a lateral pile dynamics (in a homogeneous linear viscoelastic soil) problem. The step-by-step formulation is demonstrated, and the results obtained are compared with existing formulations (Winkler-, cone-, approximate continuum–, and rigorous continuum–based formulations) having varying degrees of rigor and accuracy in the literature. The comparison is performed to evaluate the applicability of the present formulation for two class of problems (theories): half-space and soil stratum on a rigid base. It is found that the present formulation is applicable to problems in which the pile is embedded in a soil stratum on a rigid base (relevant to practice). Further, from the comparative results, discussions on the accuracy, computational efficiency, and ability to capture the mechanics of a laterally loaded pile–soil interaction problem are also provided. The advantages and limitations of the formulations are also included in the paper.

The nonlinear dynamics of rocking rigid bodies with simple geometries, such as rectangular blocks and cylinders, have been the focus of the rocking community over the last six decades. However, many objects that are prone to rocking or overturning do not conform to such geometries. These objects include museum artifacts and precariously balanced rocks in the natural world. Even in cases where the response of the rocking body is planar, the geometry of the body is much more complicated than the commonly studied geometry of a rocking block or a body with only two rocking corners. This paper introduces a complete model that can examine the planar motion of a body with an irregular in-plane polygonal geometry when subjected to a vibrational excitation, utilizing the geometry of the body as an input—for example, in the form of a stereolithography (STL) file. The model is used for studying the rocking response of an object while taking into account sliding and free flight. The problem is formulated and solved using Newtonian equations of motion, and impacts are treated as hard. A robust framework for integrating the occurring discontinuous equations of motion and for detecting transitions between patterns of motion and impacts, using MATLAB, is presented. Suitably chosen examples demonstrate the importance of accounting for the actual geometry of the studied rocking body, whose dynamic response is substantially richer than an object with simplified geometry.

The nonlocal (NL) and the strain gradient (SG) based equivalent continuum theories were proposed for nanomechanics by accommodating the long-range molecular interactions and to bypass the computationally expensive atomistic simulations. Nanostructures are often made of smart materials (e.g., piezoelectric, piezoceramic, and flexoelectric) for multifunctional properties. Applications of nanostructures in critical systems deserve accurate analysis. The buckling and postcritical behavior of an axially loaded, piezoelectric, nonlocal strain gradient (NLSG) thin cylindrical shells with functionally graded elastic properties were presented in earlier studies following Donnell’s approach, leading to a set of stiff, nonlinear, partial differential equations, accommodating the prebuckling nonlinearity and large postcritical deflection. This study extends the previous work by accommodating geometric imperfections in the formulation. Furthermore, a consistent thickness-wise distribution for the electric potential is also adopted following the Maxwell equation. The boundary layer (BL) concept is employed to solve the resulting nonlinear equations via asymptotic expansions of the regular and the BL fields. The solution is numerically illustrated on moderately short shells, illustrating the influence of the geometric imperfections. The critical load remains imperfection sensitive, yet the imperfections change an unstable postcritical path into a stable one. The influence of the functional gradation and the external electric fields on the imperfection sensitivity behavior are illustrated.

This paper reports two nonreflecting boundaries based on the Lysmer and Kuhlemeyer (L-K) absorbing boundary, respectively for absorbing symmetric and antisymmetric Lamb wave, which compensate for the weakness of the L-K absorbing boundary in the application for absorbing Lamb waves at the structural boundary. In this paper, the accurate and approximate formulas of the coefficients of the nonreflecting boundary for symmetric and antisymmetric Lamb waves are obtained by theoretical deduction based on the L-K method. The effectiveness and robustness of these nonreflecting boundaries are verified by comparing with the conventional L-K absorbing boundary in simulation cases. In addition, the limitations of these nonreflective boundaries are discussed.

Partially saturated soils are commonly encountered in practice. To facilitate the construction and maintenance of infrastructure, it is important to estimate the dissipation of partially saturated soils reasonably, especially for the nonsmoothing dissipation at the boundaries. In the present study, a mathematical model was established for the dissipation process of partially saturated soils by considering nonsmooth boundary conditions. The depth-dependent total stress distribution was associated with the proposed semi-analytical solution. The corresponding semi-analytical solution was obtained, and a series of scenarios were conducted to show the characteristics associated with the dissipation process. Compared with the existing analytical solutions, the derivation process is much simpler and does not use complicated mathematical methodologies. The proposed solution was verified as accurate by comparing it with analytical solutions in the literature. This study reveals that, with a nonsmooth boundary, some effects exist that hinder the pressure changes of water and air at corresponding boundaries. The pressure variations of water and air close to the nonsmooth boundary show a rebound feature. The characteristics of the depth-dependent stress distribution mainly influence the prophase of the dissipation process for both water and air.

This study investigates the use of electromagnetic (EM) energy harvesters for the dual purpose of energy scavenging and vibration mitigation of seismically isolated structures during earthquake excitation. Specifically, the study examines the transient seismic response of a harvester–structure system, with emphasis on the electromechanical coupling. The behavior of the structure is nonlinear with bilinear hysteretic characteristics, whereas the harvester includes a standard energy harvesting circuit, which induces circuit nonlinearity. To characterize the nonlinear response of the coupled harvester–structure system, the study employs dimensional analysis and considers both trigonometric pulse-type ground motions and historical earthquake records. The results revealed the (dimensionless) parameters that are critical (e.g., the short-circuit damping coefficient and resistance ratio of the electronic load to the coil of the harvester, among others), and others that are immaterial (e.g., the yield displacement of the isolator and the forward voltage drop of the circuit when sufficiently small) to the seismic performance of the coupled system. In general, the analysis shows that flexible and low characteristic strength seismically isolated structures allow for higher energy harvesting. Furthermore, the study investigates the optimal design of the energy harvesting circuit. It recommends a harvester with a high short-circuit damping ratio because it can produce more output energy, and at the same time, suppress the structural vibration and reduce the forces acting on the structure. The analysis also shows that the circuit nonlinearity should be considered for accurate output energy estimation. Overall, the results herein indicate that EM energy harvesters are a promising alternative for reducing the structural response while simultaneously scavenging energy from the seismically induced vibration.

Damage assessment of the built infrastructure forms a critical step in post-disaster response as it is necessary for estimating the severity and extent of the disaster impact, thereby ensuring effective and adequate recovery strategies. Contrary to traditional expert-driven approaches, recent trends show growing popularity in exploring more advanced alternate solutions, such as artificial intelligence (AI) and citizen science. One major current limitation, however, is the potential lack of reliability of these approaches. While recent efforts in the disaster research domain have successfully developed and demonstrated the use of AI and crowdsourcing-based solutions for large-scale post-disaster damage assessment, the inherent uncertainty associated with the adoption of such techniques for complicated subjective and expert-reliant tasks still hampers their practical implementation. This study aims to address this issue by reducing the uncertainty and increasing the consistency in post-disaster damage assessment by developing a novel crowd-AI framework that leverages the collective power of AI with citizen science. The framework comprises two modules: (1) an uncertainty-aware AI-assisted building damage classification module; and (2) a crowd-based probabilistic module for participatory damage assessment. Mainly, the framework uses AI predictions and the underlying uncertainty as prior knowledge in a Bayesian setting to achieve an enhanced crowd-based damage assessment. This paper presents a case study and validates that this innovative crowd-AI approach can reduce the uncertainty by as much as 83%, depending on the end-user’s uncertainty tolerance setting.

A nonviscous damping model uses convolution to account for the complete velocity history. The displacement response, velocity response, and acceleration response of a system with nonviscous damping are usually computed by modified direct integration methods. However, existing direct integration methods often come with strong numerical damping that reduces the accuracy of calculated responses. An improved composite direct integration method based on the Bathe method is proposed in this study. In this method, the convolution of the nonviscous damping is computed using Simpson’s rule to improve the accuracy. The combination of Simpson’s rule and the Bathe method almost doubles the efficiency of the calculation of responses. Nonviscous damping of two moments can be computed within one loop calculation of the convolution. Compared with existing direct integration methods, the accuracy can be improved using the proposed method. Moreover, a simplified method is proposed to improve the efficiency of response calculation for the exponential kernel nonviscous damping system. The simplified method can improve the efficiency of computation greatly. Examples are used to validate the performance of the extended composite direct integration and simplified method.

Abstract

The Shields curve underestimates the critical shear stress for incipient sediment motion in shallow flow conditions. This study starts with a consideration of sediment transport based on Bagnold’s concept of stream power. Then it is shown that the critical Shields number is related to the dimensionless sediment transport rate and friction factor. Finally, by defining the threshold condition using the pickup probability, the critical Shields number is presented as a function of relative roughness height for incipient sediment motion in shallow flows. The proposed formula was calibrated with experimental data collected from 13 sources in the literature.

Practical Applications

Gravel-bed rivers usually are characterized by shallow water and steep slopes, e.g., in mountain rivers. Sediment motion in such a case long has been a subject of research interest. To evaluate incipient sediment motion over a gravel bed, we propose applying the critical sediment transport rate. With the concept of stream power and pickup probability, the critical Shields number is theoretically related to the dimensionless sediment transport rate and flow resistance. Consequently, a new formula was developed for calculating critical Shields number based on the relative roughness height. The formula was calibrated using previous experimental results.

A fast three-dimensional (3D) Voronoi and voxel based mesostructure modeling method was proposed for asphalt concrete. This method combine the advantages of the Voronoi diagram, voxel meshing, and take-and-place techniques and can be divided into three stages. First, the specimen space was divided into Voronoi cells that provided clear boundaries for the aggregates to be generated. Then, the voxel model of the specimen space was established and the voxels within the Voronoi cells were regarded as the finite elements (FEs) of the aggregates via a voxel mapping process in the second stage. Both the space division process and the voxel mapping process were repeated until aggregates with a desired gradation were obtained. The last stage was to generate air voids by randomly deleting the asphalt mortar elements via a take-and-place strategy. The effectiveness of this method was validated by the complex modulus testing data of a real asphalt concrete. A common FE simulation workflow that generally separates the processes of establishing the geometry models and FE meshes was employed. The results showed that the proposed method only spent ten of seconds, less than 1% of the time required by the common workflow, to obtain the mesostructural FE models, but exhibited very similar mesomechanical results to those from the common method. Therefore, the proposed method enables a fast entry into the FE analysis stage and can be expected to be an efficient tool in simulating the mesomechanical behavior of asphalt concrete.

Composite structures are used commonly in civil engineering; a typical example is a bridge deck consisting of concrete slabs and steel girder beams, in which shear connectors are used to connect the concrete slabs and steel beams to form a composite structure. The structural performance of a composite structure is well understood to depend not only on the properties of the primary components (slabs and steel beams) but also on the properties and condition of the shear connectors. Therefore, in a structural health monitoring and damage identification process, it is imperative to distinguish the damages in the primary components and in the shear connectors. However, in the existing literature concerning damage assessment of composite structures, there generally is a lack of differentiation between the damages in the two distinctive groups of constituent entities, and oftentimes the damages are treated simply in terms of the gross flexural stiffness with the use of an equivalent Euler–Bernoulli beam. This could result in a false identification of the actual severity of the damages, and even in misleading results. In this study, the basic mechanics governing the equivalent flexural rigidity and its distribution in a composite beam were investigated analytically, and the effects of the essential differences between the component beam damage and the shear connector damage on the distribution of the flexural rigidity were examined using numerical simulations. On this basis, the feasibility of differentiating the two types of damage from a damage identification process using vibration information, namely the natural frequencies and mode shapes, was demonstrated by means of a finite-element model updating procedure.

The multiphase displacement of Newtonian and non-Newtonian fluids in a porous media has some important applications in enhanced oil recovery (EOR) and water resources contamination. In such flows, the viscosity difference between the phases and the surface tension could destabilize the flow and cause complex patterns to occur in the interface. According to the literature, the physics of chemical flooding inside a homogeneous porous medium is not comprehensively understood. The numerical simulations are mostly restricted to the simplified volume-averaging equations for the flow inside the porous media, such as Darcy’s law and its modifications (Darcy-Brinkman, Darcy-Forchheimer, Darcy-Oldroyd, Modified-Darcy) for both cases of Newtonian and non-Newtonian fluid flows. These simplified equations might be unsuitable/inaccurate, especially in the case of non-Newtonian fluids. There are few studies in which the full nonlinear momentum equations of Newtonian and/or non-Newtonian fluids (i.e., the Navier-Stokes and Cauchy equations) are solved for the case of immiscible two-phase flows inside a pore-scale micromodel of the porous media. Here, we studied this problem for cases of both displacement of Newtonian and non-Newtonian fluids using the level-set method, which is excellent for tracking the interface between the phases. Therefore, the results of these simulations are expected to be more realistic than solving the simplified volume averaging equations. Here, the oil displacement efficiency in the micromodel is evaluated by seawater, an aqueous solution of sodium dodecyl sulfate (an anionic surfactant), and a polystyrene solution as the shear-thinning non-Newtonian fluid. The effect of displacing fluid injection rate, viscosity ratio, mobility ratio, and power-law index $n$ of the non-Newtonian Carreau fluid model on the displacement parameters and Saffman-Taylor instability is studied in detail. The result indicates the effectiveness of shear-thinning non-Newtonian fluid within the sweeping range of the micromodel, compared to other discussed chemical factors. It was found that for the case of injected polystyrene at a flow rate of $0.5\mathrm{mL}/\mathrm{h}$, the breakthrough time increased to 34 s, compared to 21 s for seawater.

Programmable instability is crucial to the design of intelligent structures. Origami has inspired new methods of designing spatial truss structures with programmable instability. In this study, the finite-particle method was used to investigate the programmable instability modes of spatial structures based on Miura-ori, an origami pattern. Truss models of single-layer and multilayer Miura tube–based spatial structures were established. The instability process for each structure was simulated, and each structure’s stiffness, strain energy, and instability processes were studied. The instability modes of these structures were controlled through adjustment of two geometrical parameters: the angle in a plane and the angle between planes. The influence of these parameters on structural stability was analyzed. Finally, the numerical simulation results were compared with the results of experiments involving physical models. This paper proposes a new design for Miura tube–based truss structures with programmable instability.

Low-velocity impacts (LVI) during a structure’s service life may affect the mechanical properties of carbon-fiber–reinforced composites (CFRP) and, consequently, their structural performance. Accordingly, it is crucial to consider this issue in the design steps of composite structures. In the present study, experimental procedures and numerical modeling have been implemented to predict the damage induced by LVI and its influence on the residual compressive strength of carbon/epoxy laminates. For numerical simulation, a user-defined subroutine (VUMAT) is utilized in commercial finite element software ABAQUS/Explicit. The first step of numerical simulations corresponds to the collision of the impactor with the composite plate. Then, the impacted composite plate is subjected to a compression test to estimate the residual compressive strength. Experimental tests are completed by performing LVI tests to validate the numerical simulation results in three impact energy levels of 10, 15, and 20 J. The results reveal good agreement between experimental and numerical methods. According to the results, 18.1%, 28.7%, and 33.0% reductions in compressive strength of the carbon/epoxy composites were observed due to LVI with energy levels of 10, 15, and 20 J, respectively.

Cyclic loading is encountered in several practical geotechnical problems. Understanding and modeling cyclic soil responses are key to engineering analysis and design. As one of the most distinct features of soils, fabric anisotropy plays an essential role in the soil response to cyclic loading, such as pore-pressure development dependent on the interplay between fabric and loading direction, effective stress path inclination related to initial fabric, and stiffness variation associated with bedding-plane orientation. However, most previous clay models developed within the bounding surface framework use the rotational angle, a stress-ratio-type scalar, to describe fabric anisotropy and its evolution, which fails to comprehensively capture the anisotropic responses. In this study, a deviatoric fabric tensor, instead of the commonly used rotational angle, was used to describe the internal microstructure within the framework of anisotropic critical state theory. A scalar-valued anisotropic fabric variable quantifying the interplay between the fabric tensor and loading direction was used to account for the impact of anisotropy on both dilatancy and strength, aimed at simulating the typical ‘butterfly-shaped’ stress loops, and varying rates of stiffness degradation and pore-pressure accumulation of samples with different bedding-plane directions. The initial fabric tensor was also introduced into the elastic expression to replicate the inclined undrained stress paths, as well as the variational degree and direction of inclinations due to different bedding-plane orientations. The predictive capability of the proposed model was demonstrated by simulating three typical clays in undrained and drained conditions, with varying stress and strain amplitudes. The model can capture the major influences from the initial fabric anisotropy and its evolution of clay, i.e., the typical ‘butterfly-shaped’ stress loops and the bedding-plane direction-dependent effective stress path, the pore-pressure generation, stiffness variation, and strain accumulation.

This work designs a new compression-mode viscoelastic (VE) damper to avoid bonding interface failure in classical shear-mode VE dampers. However, most existing mathematical models based on the shear deformation at the macroscopic scale cannot fully characterize the damper due to the limitations in describing strong nonlinear behavior. Therefore, a new model is proposed from the microscopic perspective to predict the nonlinear characteristics of compression-mode VE dampers. First, a series of dynamic experiments were conducted to investigate the damper’s performance under different excitation conditions and temperatures. The results indicate that the damper exhibits strong energy dissipation characteristics and good deformation capacity in wide temperature and frequency ranges. Besides, the damper also shows obvious nonlinear behavior because the hysteresis loops and mechanical properties are significantly affected by excitation frequencies, displacement amplitudes, and ambient temperatures. Then, the new model is proposed by combining the fractional derivative theory and the micromolecular structure of VE material. The model can reflect the influence of material microstructure on the damper’s macroscopic mechanical properties, and each parameter of the model has a clear physical meaning. Finally, the accuracy of the model is verified by experiments. The results suggest that the model can effectively describe the nonlinear behavior of the new damper, both in terms of hysteresis loops and mechanical properties. Furthermore, since the derivation of the model does not depend on a specific damper form, the model is also suitable for analyzing other VE dampers that have similar nonlinear characteristics as the new damper. In future studies, to further improve the applicability of the new model, it is significant to optimize the model to reflect more nonlinear factors and study a structural dynamic analysis method based on the model.

Machine learning–based methods, especially deep learning methods, have achieved great success in seismic response modeling due to their exceptional performance in capturing nonlinear features. However, imbalanced features of a limited training data set can significantly decrease the prediction accuracy of machine learning models. Therefore, this study proposes a novel frequency-based clustering approach for ground motion selection to generate a balanced training data set to improve the data-driven surrogate modeling of bridges. The hierarchical clustering method was developed to suppress the redundant information on the basis of a wavelet analysis of ground motion records. The proposed method was validated by a benchmark finite-element model of a girder bridge, in which long short-term memory (LSTM) neural network was used to predict the seismic responses given ground motion excitations. Specifically, the prediction performances of LSTM surrogate models trained using different data sets have been compared, while the influence of time-frequency characteristics of ground motions has been discussed in detail. The results indicated that the proposed method can provide a balanced training data set with a uniform distribution of time-frequency characteristics and effectively improve the prediction accuracy of deep learning–based surrogate models.

It is greatly significant for structural safety in service to real-time adjust the structural posture. The prerequisite for this purpose is to monitor in real time the deformation displacements. To solve the problem, this study proposes an improved inverse finite-element method (iFEM) for reconstructing the deformation by using discrete strain measurement data. The technology is also known as shape sensing, which presents robust and efficient computational characteristics. This paper proposes an improved iFEM method for composite and sandwich closed-section beams, which employs the iFEM methodology as a general framework and the refined zigzag theory (RZT) as the underlying beam theory to establish the shape-sensing model. The main advantages of the proposed formulation are that the reconstruction model can not only avoid shear strain measurement problems, but also eliminate the influence of the shear correction factors on reconstruction accuracy. The proposed algorithm is implemented in a cantilever three-layered composite beam subjected to two loads. The numerical results demonstrate the superior capability and potential applicability of the proposed improved shape-sensing methodology for monitoring accurate displacement and stress field of laminated composite beam structures.