Journal of Structural Engineering

Recent advances in nonlinear analysis of reinforced concrete structures have been based primarily on fiber-based discretization of the member cross section at locations of expected inelastic action. However, the accuracy of a fiber-section model is almost entirely dependent on the ability of constitutive material models to represent the overall inelastic behavior of the member. The postyield response of an RC element is controlled by the longitudinal reinforcing steel and the confinement provided by the transverse reinforcement. While adequately confined concrete exhibits stable hysteretic behavior, poorly confined sections degrade rapidly initiated in part by buckling of the longitudinal reinforcement. Particular attention must, therefore, be given to the stress-strain model used to represent the behavior of reinforcing steel bars. The development and validation of an advanced material model for reinforcing bars in RC members is presented in this paper. First, a base material model to describe the primary cyclic stress-strain relationship of reinforcing steel is developed. Using available concepts to initiate bar buckling, low-cycle fatigue fracture, and cyclic degradation, a generic phenomenological material model is developed and implemented in an open-source computational platform. The effectiveness of the new reinforcing bar model is validated through comparison with available experimental data.

Most of the existing mechanical models for shear strength and strain calculation of beam–column joints do not explicitly account for bond conditions of the longitudinal reinforcement of the connected members that are anchored inside the joint region. When considered, bond is accounted for through empirical parameters. In this paper, an analytical method for the definition of the strain distribution along the anchorage of the column longitudinal reinforcement inside a bridge monolithic connection is proposed. The anchorage capacity of the main column reinforcement is considered to be affected by two independent mechanisms. The capacity of the first mechanism depends on the average bond stress along the elastic part of the anchorage that is calculated by a mathematical solution without taking into account the beneficial effect of the transverse reinforcement. The capacity of the second mechanism depends on the coefficient of friction arising between main column bars and transverse reinforcement that encloses and restrains the anchorage. The coefficient of friction is empirically set in relation to the position of the transverse bar; either along the elastic part of the anchorage or in the part where yielding penetrates inside the joint. A parametric evaluation of the main design parameters affecting the anchorage capacity is conducted next. The proposed model is finally used in calculating the joint shear response of selected bridge connection specimens for which reported data of strain distribution along anchorages exist. The calculated values correlate successfully with the associated experimental data.

Depending on the relevant material properties, failure of grouted anchor anchorage to concrete can result from pullout of concrete cones, debonding at either anchor-grout or grout-concrete interface, fracture of anchor, and combination of some of these failure modes. The intention of the present study is to present an analytical method for analyzing the concrete cone failure during the process of interfacial debonding by using deformation compatibility conditions at the anchor-grout and grout-concrete interfaces. The tensile stress in the anchor and interfacial stresses at the two interfaces are formulated at different loading stages. It is assumed that interfacial debonding modeled as an interfacial shear crack only appears at the anchor-grout interface and is initiated from the loaded end. The concrete cone and interfacial debonding failures are analyzed according to the sizes and material properties of the anchorage. Three different failure modes are considered, including concrete cone failure at the unloaded end without interfacial debonding, interfacial debonding plus concrete cone failure at the same height as the interfacial shear crack tip, and interfacial debonding plus concrete cone failure at the unloaded end. In each case, the load for the initiation of the concrete cone failure, and the cone angle are determined analytically. Subsequently, material and structural parameters are adopted to study their influences on the failure modes and calculated results. It is found that if the embedment length is relatively long, the concrete cone failure occurs within a local scope. If the concrete diameter is large enough, the failure mode is the interfacial debonding without concrete cone failure. The conclusions provide useful references for engineering practice such as designing grouted anchor anchorage to concrete in light of different needs and failure modes.

In this paper a numerical and an experimental study on reinforced concrete box culverts (BC) behavior is presented. A BC under a $9.5\mathrm{m}$ high embankment was instrumented and observed during the construction period. Numerical analyses were then performed using a finite-element code capable of considering the construction sequence, the nonlinear behavior of the reinforced concrete structure, and an elastic-plastic behavior for the soil and the interfaces. Once the computed results were in good agreement with the observed behavior, a parametric study was developed for the identification of the main parameters influencing the interaction mechanism and to evaluate the BC structural performance up to failure. The influence of the BC nonlinear behavior on the interaction mechanism, both in service and ultimate limit state conditions, was analyzed and discussed. It is concluded that the soil pressures and the BC nonlinear behavior are directly related and should be reflected in the design stage in order to achieve a more rational and economical design.

In the paper corrosion-induced crack initiation and propagation are investigated experimentally and numerically, with particular emphasis on quantifying the proportion of corrosion products that are dissipated within the concrete pores and cracks, thus reducing the pressure exerted by corrosion products on the surrounding concrete. Initially, experimental data on crack initiation and propagation obtained from accelerated corrosion tests of reinforced concrete slabs are presented. A comparison of finite-element model results and experimental data is used to estimate the amount of corrosion products penetrating into concrete pores and cracks, which is an essential parameter for prediction of corrosion initiation and propagation. It was found that the amount of corrosion products penetrating into the concrete pores before crack initiation is larger than that obtained by other researchers. The paper also showed that corrosion products do not fully fill corrosion-induced cracks in concrete immediately after their initiation as the cracks are being filled gradually over time and the thicker the concrete cover the longer it will take to fully fill a crack.

Seismic design relies on inelastic deformations through hysteretic behavior. However, this translates into damage on structural elements, permanent system deformations following an earthquake, and possibly high cost for repairs. An alternative design approach, proposed in the past, is to concentrate damage on disposable and easy to repair structural elements (i.e., “structural fuses”), whereas the main structure is designed to remain elastic or with minor inelastic deformations. The implementation of the structural fuse concept into actual buildings would benefit from a systematic and simple design procedure. Such a general procedure is proposed here for designing new or retrofitted structures. The proposed structural fuse design procedure for multi-degree-of-freedom structures relies on results of a parametric study (presented in the paper), considering the behavior of nonlinear single degree of freedom systems subjected to synthetic ground motions. Nonlinear dynamic response is presented in dimensionless charts normalized with respect to key parameters. The proposed design procedure is illustrated as an example of application using Buckling-restrained braces as metallic structural fuses. This example is used in an experimental project (which is described in a companion paper) as a proof of concept to the developed design procedure.

In a companion paper, an alternative design approach was proposed to concentrate damage on disposable and easy to repair structural elements (i.e., “structural fuses”), while the main structure is designed to remain elastic or with minor inelastic deformations. To verify and validate the developed design procedure, an experimental project was conducted on the shaking table at the University at Buffalo, which consists of a three-story frame designed with buckling-restrained braces (BRBs) working as metallic structural fuses. This experimental project also assesses the replaceability of BRBs designed as sacrificeable and easy-to-repair members. These BRBs are connected to the frame by removable and eccentric gusset plate, especially designed to prevent performance problems observed in other experimental research. Design and behavior of this type of connection is also investigated here. Another objective of this test is to examine the use of seismic isolation devices to protect nonstructural components from severe floor vibrations in buildings designed per the structural fuse concept. The seismic isolation device consists of a bearing with a spherical ball rolling in conical steel plates, also called a ball-in-cone system. This type of seismic isolator was installed on the top floor of the frame model, and its response in terms of acceleration and displacement is investigated as part of this investigation.

The close proximity to the New Madrid Seismic Zone and the significant population and infrastructure presents a need to evaluate the seismic provisions in building codes adopted in Memphis, Tenn., and other central U.S. cities. This investigation identifies potential deficiencies for a typical four-story reinforced concrete (RC) frame building designed to current code requirements in the central United States. The structure was designed for three sets of seismic provisions: the 2003 International Building Code (IBC), the 2003 IBC with local Memphis amendments, and the 1999 Standard Building Code (SBC). Special seismic detailing was required for the IBC and SBC designs, whereas the amended IBC design required only intermediate seismic detailing. The nonlinear structural analysis for two earthquake recurrence intervals (10 and 2% in $50\text{years}$ ) indicates special seismic detailing can provide a significant improvement in the seismic performance of similar low- to midrise RC buildings in the central United States.

The infrequent nature of earthquakes in the Central and Eastern United States (CEUS), and the fact that none with intensity comparable to the New Madrid sequence of 1811-12 has occurred in the past century, have caused the earthquake hazard in the CEUS to be neglected until quite recently. The performance of reinforced concrete (RC) frames in the CEUS may be deficient if they are subjected to earthquakes that are believed to be plausible in the New Madrid Seismic Zone (NMSZ). This study presents a set of probability-based tools for seismic vulnerability and risk assessment of such gravity load designed (GLD) RC frames, and uses these tools to evaluate the seismic vulnerability of RC frames that are representative of the building inventory in Memphis, Tenn.—the largest population center close to the NMSZ. This evaluation indicates that traditional GLD RC frames have a substantial probability of suffering life-threatening damage or collapse in the event that they are subjected to earthquake demands stipulated in recent seismic design regulations such as ASCE Standard 7-05.

Structural control using energy dissipater devices is emerging as a heavily researched strategy in earthquake engineering. Among several control systems, semiactive control is usually possible and efficient. In this research, a semiactive energy dissipating bracing system based on a viscous damper is proposed. In the conventional bracing systems, it is assumed that the braces can buckle under compression. Therefore, a semiactive on-off brace strategy is implemented to improve the conventional brace performance. Further, an energy absorbing mechanism is implemented. In the proposed system, the buckling of the member is prevented by implementing a one-way valve device. The permanent story drifts at the end of excitations are negligible. In this mechanism, any actuator and/or large power supply are not needed, instead, just battery-size power supplies to switch the one-way valve mechanical system on or off. The developed mechanism is composed of a viscous damper, a length correction control system, and a normal wind bracing on each floor. Despite the fact that this system does not get control command from the outside, it is called a semiactive control system. The hysteresis loop of the braces is almost similar to ductile tensile steel members. Therefore, the system is full capacity design. Nonlinear dynamic analysis of the system has been carried out and structural behavior has been investigated. In addition, damage indices are determined in order to compare the viscous tensile bracing system with the conventional bracing systems.

A test building was constructed in Fredericton, New Brunswick and was used to collect pressure and force coefficient data. A 1:200 scale model was also fabricated and tested for a total of 15 angles of wind attack at the Building Aerodynamics Laboratory of Concordia University. Mean and peak local and area-averaged pressure coefficients were measured and finite-element analysis via a 3D linear model was performed in order to compute force coefficients for the selected wind directions. Pressure distribution comparisons between the full-scale and the wind tunnel results showed generally good agreement with small discrepancies attributed to the direction fluctuations of the full-scale records. Force coefficient comparison between the full-scale and the finite element analysis results showed good agreement for the mean values. However, significant overestimation of the fluctuating dynamic component of the foundation force is apparent when this is derived from the fluctuating roof pressures, in comparison to the directly measured values in the field. The experimental critical local pressures (suctions) tend to be higher (lower) in comparison to the ASCE 7-05 provisions, although area-averaged pressures and suctions seem to agree much better. The results can be used to provide realistic provisions for wind design of building structural systems and foundations.

As the number of high-rise buildings increases, the floor-to-floor height has been considered as a significant component due to limited city areas. To reduce the floor-to-floor height, a newly developed composite beam using asymmetric steel section with web opening was proposed and experimentally explored. Shear connectors are not used for the proposed composite beam. Instead, longitudinal shear strength is obtained through the bond strength of the interface between concrete and steel, and the bearing strength of the opened web area. In this paper, the flexural behavior of the proposed beam was assessed using the simple beam test. For comparison, a slim floor beam was tested. A bare steel beam was also tested to evaluate resistance under construction loads. The proposed system showed satisfactory horizontal shear resistance and good composite behavior.