Journal of Structural Engineering

This paper investigates numerically the optimal tuning of Coulomb friction dampers on cables, where the optimality criterion is maximum additional damping in the first vibration mode. The expression for the optimal friction force level of Coulomb friction dampers follows from the linear viscous damper via harmonic averaging. It turns out that the friction force level has to be adjusted in proportion to cable amplitude at damper position which is realized by amplitude feedback in real time. The performance of this adaptive damper is assessed by simulated free decay curves from which the damping is estimated. It is found that the damping efficiency agrees well with the expected value at the theoretical optimum. However, maximum damping is larger and achieved at a force to amplitude ratio of 1.4 times the analytical value. Investigations show that the increased damping results from energy spillover to higher modes evoked by the amplitude proportional Coulomb friction damper which clamps the cable at its upper and lower positions. The resulting nonsinusoidal cable motion clearly violates the assumption of pure harmonic motion and explains why such dampers have to be tuned differently from optimal linear viscous dampers.

In this work, a large-scale magneto-rheological fluid (MRF) bypass valve damper for seismic mitigation applications is designed, constructed, and tested. The damper utilizes a simple design, with a modular bypass MRF valve, connected to a passive commercial seismic damper. The modular feature of the new MRF valve provides the flexibility of adding as much force as needed to the damper with minimal modification to an existing passive viscous damper. The performance of the MRF valve and the combined semiactive bypass MRF damper are examined through quasi-static and dynamic tests. It is demonstrated that the proposed modular MRF bypass damper can provide a maximum damping force over 200 kN which meets the high-force requirements of a semiactive seismic damper for large-scale structure vibration control.

Long-term monitoring on the degradation of concrete structures has been a major engineering concern in many parts of the world. On the basis of the distributed long-gauge fiber optic sensors recently developed at Ibaraki University, this paper is dedicated to proposing a method for parametric estimation of RC flexural members. Experimental investigations are first introduced to validate the ascendancy of the developed sensors over some traditional sensing techniques for health monitoring of a RC beam. A section fiber model with plane section assumption is then adopted to establish the correlations among loads, structural parameters, and structural macrostrain responses. Based on the assumption that all structural parameters within the gauge length of a long-gauge sensor are homogeneous in an average way, an idea of “dividing and ruling” is put forward by artificially dividing a member into several cells corresponding to the sensor gauge length and then treating each cell as a section fiber. From the view of long-term structural health monitoring, an integrated strategy is proposed to identify the structural parameters including curvature, crossing area of reinforcement, and average stiffness of RC flexural members. Experimental and analytical investigations on a RC beam are finally conducted to verify the feasibility and effectiveness of the proposed method.

In blast engineering, many designs begin with simplified hand procedures with the loading parameters determined based upon a reflective surface of infinite size. Individual structural members such as columns have finite widths and should be considered as finite surfaces for blast loading calculations. A study was performed to investigate the effect of finite flange width on blast loadings on structural components. The diffraction of a blast wave around the leading edges of the cross section and the propagation of rarefaction waves from the leading edges to the column centerline leads to a more rapid reduction in reflected pressure than that of an infinite surface: a process that is widely known as clearing. A series of analyses were performed using the computational fluid dynamics code Air3d. Peak reflected pressures are not changed by the “finiteness” of the section, although the reflected impulse can be substantially reduced by clearing. For a given charge mass, held constant for a range of stand-off distances, $R$ , impulse is approximately proportional to $1/R$ when considering an infinite surface. If clearing is considered, the reflected impulse is still proportional to $1/R$ , but can be 50% lower than the value computed for an infinite surface, which has significant implications for blast resistant design of structural components.

This work has two objectives. The first objective is to introduce structural engineers to the experiment design technique, whose future applications may include, for example, the total response of tall buildings, the strength reserve of statically indeterminate structures with nonlinear behavior, for which the duration of each run is large, and the aerodynamic response of a structure under a flow with a variety of simulated flow parameters. The second objective is to show, by using the experiment design technique, that wind load factors for tall buildings depend significantly upon uncertainties with respect to the natural frequencies of vibration and the damping ratios. Because this dependence is disregarded in the ASCE 7-05 Standard, wind load factors specified therein can lead to safety levels that are significantly lower for flexible buildings than for rigid buildings. An unexpected result of the study is that estimates of wind effects based on peak gust speeds entail nonnegligible errors. Consideration should therefore be given to replacing peak gust speeds by sustained wind speeds.

In this study, a parallel axial-flexural hinge model capable of representing postyield flexural behavior and considering interaction effects of axial force and moment is proposed for a simplified nonlinear progressive collapse analysis of welded steel moment frames. To this end, the load-resisting mechanism of the column-removed double-span beams was investigated based on the material and geometric nonlinear parametric finite-element analysis. A multilinear parallel point hinge model which captures the moment-axial tension interaction was then proposed. The emphasis was to develop a reliable and computationally efficient macromodel for practical progressive collapse analysis. The application of the proposed hinge model to nonlinear dynamic progressive collapse analysis was illustrated by using OpenSEES program. The accuracy as well as the efficiency of the proposed model was verified based on inelastic dynamic finite-element analysis results. The importance of including catenary action effects for proper progressive collapse resistant analysis and design was also emphasized.

This paper describes research results for an innovative curved highway bridge girder system, which uses I-shaped steel girders with hollow rectangular tubes as flanges. The increased torsional stiffness provided by the tubular flanges dramatically improves the structural behavior of the curved girders, resulting in substantially reduced deflection, cross section rotation, and stress compared to conventional curved I-shaped steel plate girders. In this paper, finite-element (FE) models for systems of curved tubular-flange girders are described. The models consider material inelasticity, second-order effects, initial geometric imperfection, and residual stresses. The girder systems are comprised of curved hollow tubular-flange girders (CHTFGs), cross frames between the CHTFGs, and a concrete deck. A parametric study is performed using the FE models to study the effects of web stiffeners, tube diaphragms, geometric imperfection, and residual stresses on the load capacity of three-girder systems with CHTFGs. Then, the results for the CHTFG systems are compared with results for corresponding conventional curved I-girder systems. The effects of the curvature, cross section dimensions, number of cross frames, and a concrete deck are investigated. The results indicate that the CHTFG systems are more structurally efficient than the corresponding curved I-girder systems.

Controversy has risen about the suitability of small-scaled pushout tests to satisfactorily model the behavior of shear connections in composite beams. However, the determination of the strength of the shear connectors from full-scale composite beam tests is expensive, complicated, and subject to modeling inaccuracies. The results of several composite beam tests incorporating trapezoidal steel decking are presented along with the results of companion pushout tests. These were especially designed to deliver insight into the ductility of the shear connection when subject to the so-called “brittle” failure modes of the concrete slab observed in small-scale pushout tests, and to examine the suitability of novel reinforcing measures such as waveform reinforcement components or stud performance-enhancing devices. The comparison of the shear connection performance in both the carefully conducted pushout tests where no additional lateral restraints were employed and the beam tests demonstrated that the failure modes experienced were identical, the load-slip behavior was also similar, and comparable connector strengths were achieved. The requirement of sufficiently ductile shear connection behavior in achieving a satisfactory composite beam performance was also established in these tests.

The paper investigates the design of continuous steel-concrete composite beams with compact cross section using the elastic analysis with limited redistribution. The permissible moment redistribution which satisfies the requirements of the ultimate limit state (collapse) and serviceability limit state (crack width in the concrete slab) was computed. An advanced finite element program accounting for all mechanical nonlinearities and time-dependent phenomena (creep and shrinkage of concrete) was used. An extensive parametric analysis aimed to determine the influence of several geometrical parameters on the permissible moment redistribution was carried out on propped cantilevers and fixed-end beams. The analyzed parameters include the shape of the steel profile, the ratio between the depths of concrete slab and steel beam, the steel to concrete area ratio, and the reinforcement percentage of the concrete slab. The analysis was limited to compact steel sections (AISC 360-05) or class 1 steel sections (Eurocode 3) and low ductility reinforcing steel (elongation at maximum load ${\epsilon }_{\text{ru}}=2.5\%$ ). The moment redistribution domain which satisfies the rotation compatibility in the critical sections, due to the attainment of the rupture of the reinforcement or the local buckling of the steel profile, and the control of cracking $\left(<0.3\text{mm}\right)$ in service was evaluated and compared with the limits recommended by current codes of practice. A proposal for the allowable moment redistribution domain according to the limits of the study was given.

Several codes of practice now support the use of the strut-and-tie method (STM) for the design of complex regions in structural concrete. In this method, a load-resisting truss is idealized and designed to carry the applied forces through these regions to their supports. The method assumes that the load can be carried in the manner envisioned by the designer and that the nominal design strength is at least equal to the calculated capacity of the idealized plastic truss. These assumptions are not always valid, particularly for nonductile and complex structures, as revealed by experiments in which some of STM designed structures have exhibited poor performance at service load levels and/or not been able to support their calculated nominal design strength. Thus, there is clearly a need for a convenient and reliable means of assessing the likely performance of complex regions designed using the STM. This paper presents an integrated STM design and computational framework that was developed to overcome the barriers to efficient design by the STM and effective design validation by nonlinear finite-element analysis.

This paper presents a study on the stability capacity and lateral bracing requirements of wood beam-columns subjected to biaxial eccentric compression loads. Material property tests and wood beam-column tests were conducted to provide input parameters and verification to a finite element method based numerical analysis model. The model provided good predictions of the maximum compression load, midspan lateral deflection and lateral bracing force of the wood beam-columns. A parametric study was conducted for the effect of the lateral bracing stiffness, load eccentricity and initial lateral deflections. The adequacy of the 2% rule-of-thumb, Winter’s and Plaut’s methods was also studied by comparing to the test results of the lateral bracing force. The developed model and its output can be used to improve the design methods of the lateral bracing systems of wood beam-columns.

The structural behavior of timber-concrete composite slabs is mainly governed by the shear connection between timber and concrete. When a timber-concrete composite structure is exposed to fire, it is of particular importance to know the changes in stiffness and strength that the shear connection is subjected to. For the calculation of the fire resistance of timber-concrete composite slabs a simplified design method was developed on the basis of the calculation model for mechanically jointed beams with flexible elastic connection given in EN 1995-1-1 and the reduced cross section method given in EN 1995-1-2. The effects of temperature on the mechanical properties of timber, concrete, and connection are taken into account by modification factors $k_{\text{mod},\text{fi}}$ . For the strength and stiffness properties of the screwed connections simplified formulae for the calculation of the modification factors $k_{\text{mod},\text{fi}}$ were developed, based on the results of an extensive experimental study on the fire behavior of the screwed connections. The modification factors $k_{\text{mod},\text{fi}}$ were expressed as a function of the side cover of the connections. In this way the temperature in the timber member does not need to be determined, making the calculation of the strength and stiffness properties of the screwed connection in fire much easier for the designer. The paper first describes and discusses the behavior and modeling of the screwed connection in fire, and then the simplified design method developed for the calculation of the fire resistance of timber-concrete composite slabs with screwed connections is presented. Finally, the result of the calculation method is compared to the result of a full-scale fire test.