Journal of Engineering Mechanics

Silos in the form of a cylindrical metal shell are commonly elevated to provide access to the space beneath, permitting the contained materials to be directly discharged. A few discrete column supports at evenly spaced intervals are commonly used. However, the structural design of discretely supported cylindrical shells presents a variety of challenges. The presence of discrete supports results in circumferential nonuniformity in the axial compressive stress as well as a progressive vertical decay above the support. Several approaches can be adopted in design depending on the severity of the nonuniformity of the stresses. Relevant research to date has focused mostly on the behavior of cylinders supported on brackets, local forces at the base, or stiff ring beams. The use of intermediate ring stiffeners to provide circumferential uniformity in the axial membrane stresses has long been recognized, but few studies have given a clear view of the practical requirements for such rings. In this paper, a combination of base and intermediate ring stiffeners is explored to develop a practical and cost-effective solution that leads to more uniformity in the axial membrane stresses above the intermediate ring stiffener. For the purposes of obtaining a simple analytical solution, the cylindrical shell is subjected to the fundamental harmonic of the column support and analyzed using membrane theory. It is shown that an ideal location exists for an intermediate ring stiffener such that the axial membrane stress above this ring is circumferentially completely uniform. The ideal location of this ring is determined analytically and is expressed in terms of the basic geometric variables. This ideal ring location is then independently verified using many linear finite-element analyses. A further study explores the effect of placing the intermediate ring stiffener below the ideal location. The results are presented in a manner that makes them suitable for direct adoption into traditional design specifications.

A unified approach for the aeroelastic analysis of bridge decks is presented. The structural aerodynamic transfer functions are approximated using a second-order polynomial with the coefficients determined through minimization of the weighted error between the exact and approximate transfer functions. The exact aeroelastic transfer function is used as the weighting function to define the weighted error. Using the proposed approximation, the dependence of the structural aerodynamic transfer function on frequency is eliminated, and a popular time-marching algorithm is adopted for aeroelastic analysis in the time domain. For the frequency-domain analysis, a complete set of modal frequencies and modal shapes can be evaluated in a single eigenvalue analysis. The efficiency of the analysis is enhanced with the truncated modal decomposition in the structural eigenvector space. The proposed method is verified for an idealized cable-supported structure with two extreme types of deck sections. Approximating the structural aerodynamic transfer function works well, even in the case of a bluff H-type section. The proposed method substantially simplifies the aeroelastic analysis procedure, while yielding consistent results for frequency- and time-domain aeroelastic analyses by using the same equation of motion.

In this work, a soft matter cell model is proposed to simulate the cellular cytoskeleton network and motions of intermediate filaments during cell contact and adhesion, in an attempt to explain mechanical information exchange between cells and their extracellular environment. In particular, the cell is modeled as liquid-crystal elastomers. A microscale adhesive model has been introduced to describe the interaction between receptors and ligands. A Lagrange-type mesh-free Galerkin formulation and related computational algorithms for the proposed cell and adhesive contact model have been developed and implemented. A comparison study with experimental data has been conducted to validate the parameters of the cell model. By using the soft matter cell model, the soft adhesive contact process between cells and extracellular substrates with different stiffness has been simulated. The simulation shows that cell motion includes gliding as well as rolling forward along the substrate during the spreading process.

Brillouin scattering–based fiber-optic sensors have been used in a number of applications, including monitoring of pipelines and well-casing damage in the oil and gas industry. The major attribute of these sensors is their capability for distributed sensing of strains along large sections of structures. For these reasons, Brillouin-based fiber-optic sensors have great potential for civil structural applications, such as in bridges. This study reports the investigation of the Brillouin optical time-domain reflectometer for assessing damage in structural members. The primary advantage in using the Brillouin optical time-domain reflectometer is the single-fiber installation and multidamage-detection capability for the entire structure. The challenge in using a Brillouin optical time-domain reflectometer for detection of localized damage is to overcome the effects of system noise and Brillouin frequency-peak shift distortions. The compound effects of noise and frequency-peak distortions result in dispersion of the localized strains and concealment of crack locations. A signal-processing approach based on the multiresolution analysis (MRA) of the distributed strain data by using the stationary wavelet-transform (SWT) method is introduced in this study. This approach was employed to enhance data quality and to extract the damage-related features from the distributed strain data. The process involved decomposition of the strain data and extraction of the approximate and detailed coefficients of the signal. The MRA of the data involved further transformation of the approximate coefficients to extract the crack features. Evaluation of the method was accomplished by using signal-processing methodology for detecting the damage locations in a 15-m-long beam with simulated damage during four-point bending experiments. By using the wavelet-transform approach, it was possible to detect the location of the simulated damage sections with joint opening displacements larger than $50\hspace{0.17em}\mu \mathrm{m}$ when the damage spacing was greater than the spatial resolution. The methodology developed in this study provides a quantitative method for detecting damage locations and a qualitative measure of crack intensity. Further research is necessary to examine the efficiency of the approach for more complex loading conditions and damage patterns.

Classical quasi-steady galloping analysis deals exclusively with cases of across-wind vibrations, leaving aside the more general situation where the wind and motion may not be normal. This can arise in many circumstances, such as in the motion of a power transmission cable about its equilibrium configuration that is swayed from the vertical plane as a result of the mean wind or in a tall slender structure in a skewed wind. Furthermore, the generalization to such situations, when this had been made, has only considered special issues. In this paper, the correct equations for the quasi-steady aerodynamic damping coefficients for a rotated system or wind are derived, and the differences from other variants are highlighted. Motion in two orthogonal structural planes is considered, potentially giving coupled translational galloping, for which previous analysis has often been limited or has even arrived at erroneous conclusions. For the two-degree-of-freedom case, the behavior is dependent on the structural as well as the aerodynamic parameters, in particular the orientation of the principal structural axes and the relative natural frequencies in the two planes. For the first time, the differences in the aerodynamic damping and zones of galloping instability are quantified between solutions from the correct perfectly tuned, well detuned, and classical Den Hartog equations (and also an incorrect generalization of the latter) for a variety of typical cross-sectional shapes. It is found that although the Den Hartog summation often gives a reasonable estimate for the actual aerodynamic damping, even in the rotated situation, in some circumstances the differences can be quite large.

Real-time hybrid simulation (RTHS) has increasingly been recognized as a powerful methodology to evaluate structural components and systems under realistic operating conditions. The concept of RTHS combines the advantages of both numerical analysis and physical laboratory testing. Furthermore, the enforced real-time execution condition enables testing of rate-dependent components. One of the most important challenges in RTHS is to achieve synchronized boundary conditions between computational and physical substructures. The level of synchronization, i.e., actuators tracking performance, largely governs RTHS test stability and accuracy. The objective of this study is to propose and validate a generalized procedure for multiple-degree-of-freedom (MDOF) RTHS. A loop-shaping $H_{\mathit{\infty }}$ robust control design strategy is proposed to control the motion of the actuators. Validation experiments are performed successfully, including the challenges of multiple actuators dynamically coupled through a continuum steel moment resisting frame (MRF) specimen. The resulting framework is further utilized to evaluate the performance of a magnetorheological (MR) damper in its effectiveness to mitigate structural vibration when the structure is subjected to dynamic disturbances, e.g., earthquakes.

This paper investigates the use of notched strip tensile tests, previously proposed, to determine the yield characteristics of stainless steel. The original concept is thereby improved in two important ways. First, digital image correlation techniques are used to allow an accurate measurement of strain rates in a very localized area. Second, new theoretical developments are presented that eliminate the need for measuring the applied load and allow the yield surface of an anisotropic material to be determined starting from a parameterized equation. An experimental program is described where the method has been successfully applied to a ferritic grade as well as an austenitic grade stainless steel. As a disadvantage, a relatively large scatter in test results appears to be intrinsic to the method of notched strip tests.

In the present work, by considering the inhomogeneity, solid skeleton viscosity, and compressibility of solid particles and fluid, the calculation formulations of reverberation ray matrix method (RRMM) is established for transient response of viscoelastic porous medium. The derivation process of the formulas and calculation method is described in detail herein. Feasibility and efficiency of the present approach are validated by comparing the solutions with those obtained from other methods. As numerical examples, by assuming that the material properties of the saturated poroviscoelastic media have an exponential law distribution along the thickness coordinate, the transient response, in terms of displacement of solid skeleton, total stress, and pore fluid pressure, are examined by using numerical inverse Laplace transform. The effects of inhomogeneity on transient response of gradient saturated poroviscoelastic media are demonstrated and discussed. The results show that there are significant effects of inhomogeneity on compressive wave propagation and that RRMM is an effective way for solving the partial differential equations with variable coefficients in a mathematic model of saturated porous media with gradient-variation properties.

A linear elastic fracture mechanics (LEFM) model for the engineering fracture design of cement sheath integrity at early ages in oil and gas well applications is proposed. The model considers the specific worst-case scenario of a single radial crack and estimates the energy release rate (stress intensity factor) that potentially develops in a cement sheath as a result of the buildup of eigenstresses at early ages, considering the loss of axisymmetry. Specifically, the model involves a two-step solution procedure: a chemoelastic stress solver and a LEFM solver using an Airy-stress function approach together with the method of complex variables. For the first, an appropriate constitutive model for cement slurries at early ages is required as a backbone for chemoelastic stress development. Second, the LEFM solver uses the stresses to estimate the energy release rate and the stress intensity, and thus provide a means to evaluate the driving force of fracture propagation and the fracture risk as a function of the degree of hydration. The functional relationships thus established between the LEFM quantities and the degree of the chemical reaction are expected to become an indispensable tool in safely designing well cements for operation under extreme bore hole conditions.

An analytical model was developed to describe the nonlinear in-plane deformations of timber floor diaphragms in historic unreinforced masonry buildings. Diaphragm deformation mechanics are detailed for both principal-loading directions (parallel to joists and perpendicular to joists) and are shown to produce differential equations of identical form. The model is formulated based on the assumption of pinned-end conditions, symmetrical configurations, and a uniformly distributed applied lateral load. The model is validated against full-scale diaphragm experimental data and is shown to predict response in both principal-loading directions with good accuracy. It is recommended that further research be undertaken to establish representative parameter values for the governing nail connection load-slip function.

A new method for dynamic response spectrum analysis of a structural system with interval uncertainty is developed. This interval finite-element–based method is capable of obtaining the bounds on dynamic response of a structure with interval uncertainty. The present method is performed using a set-theoretic (interval) formulation to quantify the uncertainty present in the structure’s parameters, such as material properties and cross-sectional geometry. Independent and/or dependent variations for each element of the structure are considered. At each stage of analysis, the existence of variation is considered as presence of the perturbation in a pseudodeterministic system. Having this consideration, first, a linear interval eigenvalue problem is performed using the concept of monotonic behavior of eigenvalues for symmetric matrices subjected to nonnegative definite perturbations. Then, using the procedures for perturbation of invariant subspaces of matrices, the bounds on directional deviation (inclination) of each mode shape are obtained. Following this, the interval response spectrum analysis is performed considering the effects of input variation in terms of the structure’s total response, which includes maximum modal coordinates, modal participation factors, and mode shapes. Using this method, for the problems considered, it is shown that calculating the bounds on the dynamic response is more computationally efficient than the combinatorial or Monte Carlo–solution procedures. Several problems that illustrate the behavior of the method and comparison with combinatorial and Monte Carlo–simulation results are presented.

Accurate modeling of tire–pavement contact behavior plays an important role in the analysis of pavement performance and vehicle stability control. A three-dimensional (3D) tire–pavement interaction model was developed using the FEM to analyze the forces and contact stresses generated during vehicle maneuvering (free rolling, braking/acceleration, and cornering). A pneumatic radial-ply tire structure with rubber and reinforcement was simulated. The steady-state, tire-rolling process was simulated using an Arbitrary Lagrangian Eulerian (ALE) formulation. An improved friction model that considers the effect of sliding speed on friction coefficients was implemented to analyze the effects of pavement surface friction on contact stresses, friction forces, and cornering forces. The results showed that the magnitudes and nonuniformity of contact stresses are affected by vehicle-maneuvering conditions. As the pavement surface friction increases, the tangential tire–pavement contact stresses at various rolling conditions (free rolling, braking/acceleration, and cornering) and the vertical contact stresses at the cornering condition increase. It is reasonable to use the constant friction coefficient when predicting tire–pavement contact stresses at the free-rolling condition or at the cornering condition with small slip angles. However, it is important to use the sliding-velocity–dependent friction model when predicting the friction force at tire braking.

Density and pressure have great influences on the strength and stress-strain behaviors of rockfill material. A series of triaxial compression tests were conducted for Tacheng rockfill material with different initial consolidated void ratios and initial confining pressures. From the experimental results, Tacheng rockfill material presents behaviors of strain hardening, postpeak strain softening, volumetric contraction, and expansion with a range of densities and pressures. For the sake of simplicity, the critical state line (CSL) is linear in the plane of the mean effective stress versus the deviatoric stress. A simple expression related to the initial consolidated void ratio was proposed for the CSL of rockfill material in the plane of the void ratio versus the logarithm of mean effective stress. A comprehensive equation incorporating a state parameter is proposed to describe the complex dilatancy behaviors of Tacheng rockfill material with a range of densities and pressures. The stress ratios at the dilatancy and the peak stress ratio states are both formulated as exponential functions of the state parameter. A bounding surface model that incorporates the proposed CSL in both planes of the mean effective stress versus the deviatoric stress and the void ratio versus the logarithm of mean effective stress was established. The model, which has a set of the same model parameters, can well predict the behaviors of strain hardening and volumetric contraction for rockfill material with a loose state or in a high confining pressure and the behaviors of postpeak strain softening and volumetric expansion for rockfill material with a dense state or in a low confining pressure.

A nonlinear formulation capable of accurately and efficiently predicting the nonlinear response of an axially restrained prismatic beam-column with initial geometric imperfections (i.e., initial curvature and connection eccentricities) and nonlinear temperature-dependent end connections under elevated temperatures is presented. The proposed model includes the effects of (1) an arbitrary thermal gradient along both the span and cross section of the member and (2) the nonlinear behavior and any stiffness and strength degradations of the beam-column material. However, high-temperature creep and shear deflection effects are not taken into consideration. The proposed model is generic allowing the implementation of any thermal regime and the use of any stress-strain-temperature curves of the beam-column material. The stress-strain-temperature curves of the material can be either theoretical or experimental providing a more realistic approach to the structural response of beam-columns subjected to different fire conditions. Two comprehensive examples are presented and discussed in detail showing the effectiveness and accuracy of the proposed iterative method on the nonlinear large-deflection behavior of slender prismatic beam-columns under high temperatures, including the combined effects of initial imperfections, end restraints, and nonlinear end connections.

This study investigates the flow field and the hydrodynamic forces induced by the tilting lift of a large object from a rigid porous seabed using an analytical approach. The flow in the corners between the object and the seabed is a corner flow with a low Reynolds number, and the porous media flow in the seabed obeys the Brinkman equations. The complete boundary conditions for viscous flow, including the continuity of velocities and stresses, are utilized at the seabed-water interface. The Helmholtz decomposition theorem, which decomposes the flow field into irrotational and rotational parts, and a perturbation expansion, are employed to solve the boundary-value problem. Perturbation solutions with first-order corrections are presented, indicating that the flow inside the wedged corner is a Stokes flow driven by the liftup of the object, and the flow within the permeable seabed is a boundary-layer flow. Furthermore, in the wedged corner, the flow induced by the slip velocity at the porous interface is similar to the flow in the paint-scraper problem of Taylor.

The shear lag effect is one of the very important mechanical characteristics of thin-walled box girders and has been studied over several decades. However, the generalized displacement adopted in many papers is not very simple or clear, and the analytical procedure is somewhat complicated. In this paper, a new method for analyzing the shear lag effect in thin-walled box girders is proposed in which the additional deflection induced by the shear lag effect is adopted as a generalized displacement to describe the shear lag deformation state. Based on the generalized moment for shear lag defined in this paper, the shear lag deformation state is separated from the flexural deformation state of the corresponding elementary beam and analyzed as a fundamental deformation state. The governing differential equation and boundary condition for the additional deflection are established by applying the principle of minimum potential energy, and the initial parameter solution to the differential equation is given. A very simple and convenient formula of shear lag warping stress is proposed that has the same form as that of the bending stress of the elementary beam. A finite beam segment element with eight degrees of freedom is developed to analyze the shear lag effect in complex continuous box girders with varying depth. An example of a simply supported concrete box girder is provided in which the results obtained by the present method are compared with those by FEM, finite-strip method, and other existing analytic methods. A three-span continuous box-girder model with varying depth is also analyzed, and the calculated results are in a good agreement with the test results on the whole, which validates the analytical method and the element presented.

The dynamic response of a rectangular plate on a viscoelastic foundation under moving loads with varying velocity is investigated. Three types of moving loads, including acceleration, deceleration, and uniform speed are discussed. Analytical solutions of the dynamic deflection of a plate with general boundary conditions subjected to moving loads are presented by means of dealing with the governing equation. In the numerical analysis, a plate with four edges simply supported is selected to study the dynamic response yielded by the moving loads. The deflection distributions in the range of the plate for the accelerated and decelerated load cases are illustrated. The time histories of the deflection response corresponding to different acceleration and deceleration levels are investigated in detail. The effects of the initial speed of the moving load and the damping of the foundation on the response are also discussed.

In this paper, a new type of multilayer feedforward neural network has been proposed based on inspiration from the Preisach model, which has been called the Preisach neural network (Preisach-NN). It is comprised of input, output, and two hidden layers. The input and output layers contain linear neurons, whereas the first hidden layer incorporates neurons called stop neurons, whose activation function represents a stop operator. The second hidden layer includes sigmoidal neurons. The subgradient optimization method with space dilatation has been adopted for training of the Preisach-NN as a nonsmooth problem. Although the proposed Preisach-NN could be mathematically identical to the Preisach model, tuning of the Preisach-NN is easier and also more general than that of the model. To assess their capability, Preisach-NNs are used to model two different types of hysteretic behaviors of Masing and non-Masing problems. The results presented and discussed in this paper show that the neural networks have been capable of learning the material behaviors successfully and with high precision.