Journal of Engineering Mechanics

A new simplified contact model aimed at capturing the load transfer and recovery length in parallel steel wires, commonly used in main cables of suspension bridges, is presented. The approach is based on placing elastic–perfectly plastic spring elements at the contact region between the objects. These springs have varying stiffness (Model I) or yielding (Model II) depending on their proximity to the clamping loads. Their stiffness or yielding is highest when they are closer to this force, and it decays when they are farther away from the clamp. This decayed behavior is assigned according to Boussinesq’s well-known solution to a point load (applied on a half space). Both models converge quickly compared with a full contact model and recover Coulomb friction law on a two-dimensional (2D) benchmark problem. Moreover, when the same properties are chosen for all springs (disregarding Boussinesq solutions), the models reduce to the classical shear-lag model, which for high clamping (point) loads gives inaccurate results. The spring models are validated experimentally on a seven-wire tightened strand. In this case study, the outer wires are axially pulled, whereas the middle wire, slightly shorter than the outer wires, experiences no direct applied axial load. However, because the strand is radially fastened at several locations, the axial load is transferred to the inner wire by an interfriction mechanism between the wires. The strains at the center points of the outer and inner wires are measured via neutron diffraction for different clamping loads, showing that the inner wire is capable of recovering most of the load.

A numerical formulation for the nonlinear quasi-viscoelastic (creep and shrinkage) analysis of steel-concrete composite beams that are curved in their plan is developed. The creep behavior of the concrete is considered by using the viscoelastic Maxwell-Weichert model, in which the aging effect of the concrete is taken into account. Geometric nonlinearities and the partial shear interaction that exist at the deck-girder interface in the tangential (or longitudinal) direction and in the radial (or horizontal) direction owing to the flexibility of the shear connectors are considered in the strain-displacement relationship. The modeling based on the developed formulation is validated by comparisons with available results reported in the literature. The effects of initial curvature, partial interaction, and geometric nonlinearity on the time-dependent behavior of curved composite beams are illustrated in selected examples.

With the aid of the response surface methodology, which is a combination of mathematical and statistical techniques, this paper presents a method for updating a finite-element model based on the measured static responses of structures. Unlike in the traditional model updating procedure, original finite-element models are replaced with response surface models constructed using the uniform design. By this means the complexity of a structure can be easily expressed by explicit functions with low dimensions. A parameter scope shrinking technique is also proposed to construct response surface models. The proposed method is verified against a numerical beam and an experimental full-scale continuous box-girder bridge. It is demonstrated that the proposed response surface–based finite-element-model updating in structural statics has the advantages of easy implementation, high cost-efficiency, and adequate updating accuracy. Once the response surface model that is formulated explicitly is constructed, no finite-element calculation is required in each optimization iteration during updating. Therefore, such finite-element model updating can be easily implemented in practice with available commercial finite-element analysis packages.

The elastic solution of an orthotropic composite beam under the influence of uniformly distributed loading is obtained numerically. Two ends of the beam are rigidly fixed, and the fibers are oriented along and perpendicular to the direction of the length of the beam. An efficient finite-difference computational scheme based on displacement potential formulation is used to analyze the present mixed-boundary-value elastic problem. The effect of several important parameters such as beam aspect ratio and fiber orientation on the elastic field are investigated. Solutions are presented mainly in the form of graphs and deformed shapes. Finally, the reliability as well as superiority of the present computational scheme is discussed by comparing the present solution with the corresponding finite-element solutions.

This paper describes a new test system that relies on sudden strain energy release to subject cementitious composite specimens to high-strain-rate tensile loading. The proposed system is (1) more cost effective to build and smaller than existing systems; (2) can be used to test larger specimens, which is an important advantage for cementitious materials; and (3) can be conveniently adjusted to achieve a broad range of strain rates. The theoretical potential of the device is discussed and equations that describe the operation of the system are developed and used to identify influential variables. A computational simulation model of a prototype system is presented and exercised to quantitatively explore the influence of key variables on the accuracy of measured results. The new system has the potential for conducting accurate high-strain-rate testing of cement-based materials in direct tension.

The nonlinear behavior of cable-supported bridges is governed by the geometric nonlinearity of cables, which is attributable to sag and sliding effects at the saddle. In a cable-stayed bridge with a midspan saddle, and in all suspension bridges, cable sliding can occur at the saddle under extreme forces, such as those caused by an earthquake or typhoon. However, the conventional method of analysis of cable-supported bridges does not consider the effect of cable sliding at the saddle; instead it regards those cables as fixed. This assumption might lead to a misunderstanding of the global structure system. The goal of this study is to develop a three-dimensional (3D) elastic cable finite element that considers the sliding effect and uses a geometric nonlinear cable finite element based on elastic catenary theory. In this study, two types of sliding were considered: the roller sliding condition without friction and the frictional sliding condition. These were formulated to derive the nodal force vectors and tangential stiffness matrices. To validate the proposed 3D cable sliding element, experiments were conducted for both sliding conditions, and results were compared with calculations of the amount of sliding and displacement at the loading point. In addition, a cable-supported structural system was analyzed to investigate the characteristics of a realistic structure with cable sliding. Overall calculations using the 3D cable sliding model were in very good agreement with the measured values.

Current estimates of seismic structural fragilities are commonly made on the basis of finite collections of actual or virtual ground-acceleration records that are scaled to have the same scalar intensity measure, for example, peak ground acceleration or pseudospectral acceleration. This paper models seismic ground-acceleration records by samples of Gaussian processes $X(t)$ and constructs scaled versions $\tilde{X}(t)$ of $X(t)$ by following current procedures. This analysis shows that $\tilde{X}(t)$ and $X(t)$ have different probability laws, so that fragilities on the basis of $\tilde{X}(t)$ provide limited if any information on the seismic performance of structural systems, that is, fragilities on the basis of $X(t)$. The usefulness of current fragility estimates on the basis of scaled seismic ground-acceleration records is questionable, and scaling ground motions is not recommended.

Hot-mix asphalt (HMA) contains a significant amount of mineral aggregate, approximately 95% by weight and 85% by volume. The aggregate sphericity, orientation, and angularity are very important in determining HMA mechanical behaviors. The objective of this study is to investigate the isolated effects of the aggregate sphericity index, fractured faces, and orientation angles on the creep stiffness of HMA mixtures. The discrete-element method was employed to simulate creep compliance tests on idealized HMA mixtures. Two user-defined models were used to build 102 idealized asphalt-mix digital specimens. They were the R-model and the A-model, short for a user-defined rounded aggregate model and a user-defined angular aggregate model, respectively. Of the 102 digital specimens, 84 were prepared with the R-model to investigate the effects of aggregate sphericity and orientation, whereas the remaining 18 were built with the A-model to address the effect of aggregate angularity. A viscoelastic model was used to capture the interactions within the mix specimens. It was observed that (1) as the sphericity increased, the creep stiffness of the HMA mixture increased or decreased, depending on the angles of aggregate orientation; (2) as the angle of aggregate orientation increased, the creep stiffness of the HMA mixture increased, with the rate depending on the sphericity index values; and (3) compared with the sphericity index and orientation angles, the influence of aggregate fractured faces was insignificant.