Journal of Engineering Mechanics

This paper proposes a novel framework that combines high-fidelity mechanics-based nonlinear (hysteretic) finite-element (FE) models and a nonlinear stochastic filtering technique, referred to as the unscented Kalman filter, to estimate unknown material parameters in frame-type structures. The proposed identification framework updates nonlinear FE models using spatially limited noisy measurement data, and it can be further used for damage identification purposes. To validate its effectiveness, robustness, and accuracy, this framework is applied to a cantilever steel column representing a bridge pier and two-dimensional steel frame. Both structures are modeled using beam-column elements with distributed plasticity and are subjected to a suite of earthquake ground motions of varying intensity. The results indicate that the material parameters of the nonlinear FE models are accurately estimated provided that the loading intensity is sufficient to exercise the parts (branches) of the nonlinear material model, which are governed by the material parameters to be identified, and the measured response quantities are sufficiently sensitive to the material parameters to be identified, especially when a limited number of measurements are considered.

The present paper deals with the hydrodynamic problems of dam-reservoir systems based on the scaled boundary FEM (SBFEM). It provides a novel boundary condition for hydrodynamic pressure simulation on the truncation boundary suited for dam-reservoir interaction analysis. A two-dimensional gravity dam-reservoir system is addressed, where the impounded water is extended to infinity. In the analysis, a truncation boundary is placed at a short distance away from the dam face in the upstream direction to reduce the computational effort. The proposed approach can conveniently consider the water compressibility, the absorption of the reservoir boundary, and the effect of the reservoir geometry. It is applicable to the earthquake excitation either in the stream direction or in the vertical/cross-stream directions. Additionally, it has the potential to extend to the three-dimensional case. Numerical examples are provided to validate the effectiveness of the proposed approach.

For exposed subsea pipelines, lateral buckles are usually artificially triggered as an accommodation technique to release thermal expansion by providing temporary flotation with controlled spacing. With the given buoyancy load, the pipeline will be uplifted initially and will then buckle laterally under a certain thermally induced axial load, which is called the bifurcation load. Previous studies have analyzed this problem and modeled the uplifted pipeline as a fixed-fixed end beam, assuming the lateral seabed stiffness is infinite. However, the lateral seabed stiffness is usually low because the seabed is formed from soft soils. Using analytical methods, this paper investigates the buoyancy load required to trigger lateral buckles along a pipeline, considering a seabed with finite elastic stiffness, and provides suggestions for the buoyancy design with an elastic-plastic seabed model using an example of finite-element analysis. It was found that the seabed condition significantly influenced the flotation design.

The estimation of the deflection of glass panels under wind pressure is important in designing external facades of buildings. A direct method to compute the large deflections of glass panels under uniform loads such as wind pressure is presented. The model allows for derivation of the load-deflection response of square or rectangular multilayered glass panels subject to uniform loads. The boundary conditions examined are those of four-point discontinuous supports or continuous supports along the four sides. With some assumptions on bending and membrane action of flat glasses, simple analytical expressions were derived from the elastic theory. Viscoelastic effects for multilayered glass panels with a thin interlayer of PVB were taken into account by introducing a time-dependent equivalent thickness, as suggested in the literature. The mechanical model based on the structural behavior of equivalent beams is also able to account for the flexural and membrane actions. Different experimental results available in the literature and relative to flexural tests on out-of-plane loaded, large-scale panels made of laminated structural glass were utilized to validate the model. The model was compared satisfactorily with numerical results obtained by using nonlinear FEMs and with existing analytical models. In almost all cases, the comparison was satisfactory, and the proposed model provides a useful design instrument that allows one to determine the minimum thickness of multilayered glass panels while respecting maximum allowable deflection and strength criteria.

In this paper, the displacements of a cylindrical shell with varying thickness and subjected to axial and external pressure are calculated analytically using the first-order shear deformation theory. The kinematics of the problem is defined by von Kármán theory, and the constitutive equation obeys Hooke’s law. The governing equations, which are a system of nonlinear differential equations, are extracted by applying the virtual work principle; the matched asymptotic expansion method of the perturbation technique is used to calculate the analytical solution. The effects of different load profiles and thicknesses on the results are investigated. Also, a comparison with the FEM is performed.

Traditional studies that applied the viscoelastic continuum damage (VECD) model to asphalt materials were mostly accompanied by uniaxial tests whose specimen size requirement provided challenges for mixtures cored from thin asphalt layers. In this paper, an analysis methodology is presented for determining the pseudostrains and damage characteristic curves of asphalt mixtures from both the horizontal (for tension) and vertical (for compression) axes of the indirect tension (IDT) specimen simultaneously. To validate this methodology, laboratory tests were conducted on asphalt specimens in the IDT mode. Complex modulus tests were conducted at various temperatures and frequencies to allow for the construction of a dynamic modulus master curve, which, in turn, was used to obtain the relaxation modulus. In addition, constant-crosshead strength tests were conducted at two temperatures (10 and 20°C) to evaluate the behavior of the asphalt specimens under growing damage. The results of the laboratory study revealed that the damage characteristic curve obtained for compression was located above its counterpart for tension. It also was found that the material showed more favorable damage characteristics at 10°C than at 20°C for both tension and compression and that the damage characteristic curves obeyed the time-temperature superposition principle, which is consistent with the uniaxial test results. These results indicate that the VECD model can be applied to the biaxial IDT test for characterizing tensile and compressive damage characteristic curves simultaneously.

Random processes, such as the wind velocity field and spatially varying ground motions, are usually simulated as $N$-variate stochastic processes by the conditional-simulation method. However, the temporal power spectral density (PSD) function of a single, simulated sample process may be different from the target PSD. Such differences can usually be assessed by the statistical errors (i.e., the bias and stochastic errors). Therefore, this paper investigates the bias errors and the stochastic errors of the PSD functions produced by the conditional-simulation method. For the bias errors, it was found that the conditional-simulation method might produce the nonzero bias error of the PSD functions in some cases. However, this usually does not occur in the unconditional simulation, and it should be avoided. To avoid the nonzero bias error of the PSD functions, a modified conditional-simulation method was proposed. It was verified using both the theoretical derivation and the numerical example. For the stochastic errors, the closed-form solutions for the PSD functions’ stochastic errors produced by a conditional simulation were given by the theoretical derivation and verified by numerical examples. Finally, the PSD functions’ stochastic errors produced by a conditional simulation were compared with those produced by an unconditional simulation using the spectral-representation method.

Understanding flow characteristics around air pockets is fundamental in the study of air entrainment and transport in pipelines. This study deals with the use of flow visualization technique (FVT) and high-speed particle image velocimetry (HSPIV) in exploration of the characteristics around stationary air pockets in horizontal-pipe flow. The air-pocket volume varies from 0 to 10.0 mL, and the air pocket is injected into a fully developed turbulent flow with Reynolds numbers between 17,000 and 18,400. In the plane of symmetry, the main flow features include (1) a horseshoe vortex upstream, (2) a stagnation point on the frontal interface, (3) a separation point and a separated shear layer beneath, (4) a reattached shear layer downstream of the reattachment point (for air-pocket volumes greater than 2.0 mL), and (5) a reverse-flow region downstream. The deformable air pocket in the turbulent flow causes streamwise random movements of both the stagnation and separation points around their mean positions. The flow pattern is categorized based on the occurrence of either separated flow or flow reattachment. Fully separated flow (Mode I) occurs at air-pocket volumes less than 2.0 mL. Intermittently reattached flow (Mode II) occurs if the volume is within 2.0–5.0 mL. Fully reattached flow (Mode III) is evident at volumes greater than 5.0 mL. Water particles on the air-pocket surface move with the adjacent flow, thus forming a slip boundary. The evolution of mean streamwise velocity beneath the air pocket demonstrates the formation of either a separated or a reattached shear layer. Using nonlinear regression analysis, appropriate characteristic velocity and length scales are determined to obtain similarity profiles in the separated shear layer beneath.

The principal purpose of this paper is to develop a new analytical method for the shear lag problem of T-beams. The flange slab of a T-beam is simplified into a plane stress plate. A system of Hamilton dual equations of the flange slab using symplectic elasticity theory can be obtained. Analytical solutions of the flange slab under uniform or linear shear force applied on the entire slab length are obtained by solving the equation. Based on the solution, the cause of the positive shear lag and negative shear lag phenomenon are studied in this paper. Theoretical research indicates that the method is superior for analyzing the shear lag phenomenon. The analytical solution has been validated by comparison with the beam solution and FEM solution of a cantilever T-beam subjected to a uniform load.

Ramming paste is a carbonaceous porous material used in Hall-Héroult cells. It is baked in place under varying loads. To model the cell mechanical behavior during its lifespan, it was necessary to develop a constitutive law that included ramming paste creep behavior. A three-dimensional (3D) nonlinear viscoelastic-viscoplastic constitutive law was devised and developed to model the primary and secondary creep stages of baked paste. The model consisted of two parts (i.e., viscoelastic and viscoplastic). Each creep mechanism was based on the existence of a dissipative potential for the hydrostatic and deviatoric parts. Analytical solutions were presented for linear creep behavior. For the nonlinear case, the deviatoric part of the viscoelastic behavior could be obtained numerically, and all other parts analytically. Finally, model parameters were identified for paste baked and tested at different temperatures. A pattern search algorithm was used to optimize the model parameters. A comparison of the results gained from the model with experimental results showed that the devised model well represented the nonlinear viscoelastic-viscoplastic behavior of the paste baked at 250°C and tested at room temperature. In addition, the model was able to predict the qualitative creep behavior of the paste baked at 350, 560, and 1,000°C and tested at 300, 300, and 25°C, respectively.

This study proposes a laboratory-scale experimental setup that simulates the dynamics of a full-scale stay cable with a transverse damper, allowing the full-scale performance of different types of dampers to be evaluated and compared in a laboratory setting. This test scheme resembles hybrid testing in that it retains the physical structural component for experiments, whereas it uses an equivalent model to represent the large-scale main structure instead of the numerical simulation used in a hybrid test. For the cable, a feasible model is presented that is composed of two flexible beams that are suspended vertically, fixed at their upper ends, and connected at their lower ends by a rigid mass; the damper is attached transversely to one of the beams. The setup is modeled as tensioned beams for clamped-guided supports with a terminal mass and an intermediate damper for dynamic analysis, and the characteristic equation is formulated from dynamic-stiffness matrices of beam segments. Modal properties of the setup are found to be similar to those of a cable-damper system, based on which the equivalence conditions are presented. The established equivalence between a cable and a model is further found to be roughly independent of damper properties for practical applications, therefore ascertaining the potential of this setup for a cable-damper study. Experiments have also been conducted in which the damper performance is evaluated on a full-scale cable and on a beam-mass assembly designed to represent the first vibration mode of the stay cable. The amplitude-dependent damping ratios achieved in the laboratory setting are found to be comparable to those achieved for the full-scale cable, suggesting the applicability and effectiveness of this experimental setup.

A generalized theory for the viscoelastic behavior of bitumen is presented. The mathematical model incorporates strain rate and temperature dependency as well as nonmonotonic loading and unloading with shape recovery. The model is phenomenological in nature. It can be calibrated using a relatively limited set of experimental parameters that are obtainable by uniaxial tests. It is shown that the mathematical model can be represented as a special nonlinear form of the Burgers model. This facilitates the derivation of numerical algorithms for solving the constitutive equations. A numerical scheme is implemented in a customized material model subroutine in a finite-element analysis (FEA) program. Simulation results are compared with the results of uniaxial and indentation tests on bitumen.