Journal of Structural Engineering

A procedure for the design of buckling-restrained braced frames is presented. The procedure uses nonlinear time history analysis and optimization. Results from this procedure are compared to those from the commonly used equivalent lateral force procedure. Results are used to develop design curves and formulas for rapid determination of cross-sectional areas for buckling-restrained braces in each story. One interesting result is that the optimum variation of area from story to story is nearly linear for most cases studied. This result is not predicted by the equivalent lateral force procedure, and can only be discovered through optimization and nonlinear time history analysis.

Recent earthquakes such as Loma Prieta, Northridge, and Kobe have demonstrated a need for a new design philosophy of bridge piers that avoids damage in order to ensure postearthquake serviceability and reduce financial loss. Damage avoidance design (DAD) is one such emerging philosophy that meets these objectives. DAD details require armoring of the joints; this eliminates the formation of plastic hinges. Seismic input energy is dissipated by rocking coupled with supplemental energy dissipation devices. In this paper the theoretical performance of a DAD bridge pier is validated through bidirectional quasi-static and pseudodynamic tests performed on a 30% scale specimen. The DAD pier is designed to rock on steel–steel armored interfaces. Tension-only energy dissipaters are used to increase tie down forces and further reduce dynamic response. The seismic performance of the DAD pier is compared to that of a conventional ductile pier. Results show that one can have 90% confidence that the DAD pier will survive a design basis earthquake without sustaining any damage, whereas for the conventional design substantial damage is sustained.

A new beam-to-column connection for earthquake-resistant moment-resisting frames is introduced. The connection has a beam bottom flange friction device (BFFD) and posttensioned (PT) high-strength steel strands running parallel to the beam. The BFFD provides energy dissipation to the connection and avoids interference with the floor slab. The PT strands produce self-centering connection behavior. The connection behavior requires minimal inelastic deformation of the connection components and the beams and columns, and requires no field welding. A series of seven large-scale tests were performed to investigate the effect of the BFFD friction force, connection details, and the loading history on the performance of the connection under cyclic loading. The test results indicate that the BFFD provides reliable energy dissipation, and that the connection remains damage-free under the design earthquake.

This paper presents research on seismic design of steel special moment resisting frames (SMRFs) with supplemental elastomeric dampers. The SMRF properties and damper design criteria are varied to illustrate their effects on the supplemental damper design. The SMRFs are based on current seismic design practices, and the supplemental dampers are designed using a simplified design procedure (SDP). The SDP allows the structure with dampers to be designed for specified performance objectives, where the structure is intended to reach selected performance levels, defined in terms of story drift and member response, at specified seismic input levels. Nonlinear dynamic time history analyses show that SMRFs with dampers designed using the SDP achieve the specified seismic performance objectives, and that elastomeric dampers are more effective when used in a more flexible SMRF. The SDP permits various seismic performance objectives and corresponding design criteria to be specified. Criteria that allow some inelastic behavior $(R=2)$ , but limit the story drift to 1.5% under the design basis earthquake are shown to lead to the most effective damper design.

Cracking is a very common phenomenon in reinforced concrete (RC) members, dramatically affecting the members’ properties, such as stiffness, deformation capacity, and permeability. However, cracks develop randomly, making their simulation difficult, especially under combined loading conditions. Using numerical simulation results, the width of a crack can be estimated by integrating the cracking strain, along the crack spacing. Cracking strain can be obtained from a finite element analysis with the well-known smeared crack approach. To use such an approach, however, a reliable simulation of the crack spacing is needed. In this paper, a method is proposed to predict the average crack spacing occurring in RC shear walls subjected to lateral demands. The method extends earlier work, which is based on both a strength and fracture energy criteria. The extension allows the method to be applied to RC members under a plane stress condition with orthogonally placed reinforcement. Eleven low aspect ratio RC shearwall specimens from three experimental data sets subjected to 49 different lateral demand levels are extracted from the literature and used to validate the model. From these data sets, a formula to determine the bond strength between the concrete and reinforcing steel is regressed. Evaluation of the model against the experimental data sets indicates that the method can predict the average crack spacing with reasonable accuracy not only in the crack initiation stage, but also in the stabilization stage.

An experimentally based relationship of the cyclic local bond stress–slip response of steel bars corresponding to splitting mode of bond failure is developed. The relationship is composed of a monotonic envelope response and cyclic bond degrading response. The model accounts for both unconfined and moderately confined concrete and incorporates the effect of several critical bond parameters such as the diameter of steel bars, the ratio of concrete cover to bar diameter, the concrete compressive strength, the type of confinement such as steel ties, fiber-reinforced concrete (FRC), and fiber-reinforced polymer jackets, and the area or content of confining reinforcement. Results predicted by the model showed very good agreement with the test data. The bond–slip relationship can be utilized for evaluating the seismic behavior and bond degradation response of bond-critical regions in reinforced concrete members when confined internally with steel ties or fiber reinforced concrete or retrofitted using external FRP jackets.

A softened strut-and-tie macromodel able to reproduce the flexural behavior of corbels in plain and fibrous concrete and with the presence of steel bars, including softening of compressed struts and yielding of main and secondary steel bars, is presented in this paper. The main focus of the proposed model is the determination of the load-deflection curves of corbels subjected to the coupled effects of vertical and horizontal forces bearing in mind the influence of the type of concrete (normal or high strength), of the fiber percentage and of the arrangement and percentage of the main and secondary (horizontal stirrup) steel bars. A validation of the proposed model is made therefore with reference to some of the experimental data available in the literature and also to numerical results obtained by using a nonlinear finite-element program (ATENA-2D) able to analyze the flexural behavior of deep members. Finally some design considerations on the use of fibrous reinforced concrete in RC corbels are given.

A new method is proposed for inhibiting the buckling response of steel braces. The strengthening technique involves attaching a core comprised of mortar blocks to the braces and then wrapping the entire system with carbon fiber-reinforced polymer (CFRP) sheets. An advantage of this method is that the technology can be applied in situ by a single individual with minimal training. Test results of seven specimens subjected to reversed axial loading show that it is feasible to achieve buckling restrained response up to 2% interstory drift. Other than the number of longitudinal fiber layers, it is observed that performance of the strengthening scheme depends upon the size of the core material, the presence of bond between the steel plate and core material, the existence of extra stitch plates for double angle members, and the presence of transverse CFRP layers at the member ends. It is also observed that double angle braces benefit more from the strengthening scheme than single angle braces, which are prone to premature out-of-plane buckling.

Single equal angle steel beams are often used as lintels to support eccentric loading acting normal to one flange. This loading causes combined biaxial bending and torsion, which is not allowed for in most steel design codes. Instead, approximate methods based somewhat loosely on past research studies have been used to develop design approximations and tables. This paper reviews past research on single equal angle beams used as lintels and develops an improved method of predicting their strengths which includes the effects of initial twist rotations, eccentric loads, and large twist rotations, and utilizes the plastic capacities of compact beams. The strengths predicted are significantly higher than those of previous approximations. More accurate strength approximations are proposed, and suggestions are made for serviceability design.

Complementary to recent studies on the cross-section behavior of elliptical hollow sections, investigations into the member response have also been performed and described in this paper. The first stage of these investigations involved a series of precise column buckling tests on hot-finished steel elliptical hollow sections. In total, 24 flexural buckling tests about the minor and major axes were carried out. Specimen sizes and lengths were carefully chosen in order to cover a wide range of both cross section and member slendernesses. The nondimensional column slenderness of the test specimens varied between 0.19 and 1.58. Measurements were taken of specimen geometry, global initial geometric imperfections, and tensile material properties. Key results from the tests including the full load-deformation histories have been presented and discussed. The test results have also been supplemented by 158 numerically generated results, allowing a wider range of geometries to be investigated. Design rules for the member buckling resistance of elliptical hollow section columns have been proposed and verified by means of reliability analysis. The results presented are the first member buckling tests on elliptical hollow sections, and represent part of the development of comprehensive structural design rules for these sections.

This paper presents the experimental and numerical investigation of aluminum alloy flexural members. The tests were performed on ten different sizes of square hollow sections subjected to pure bending. Material properties of each specimen were obtained from tensile coupon tests. A nonlinear finite-element model was developed and verified against the pure bending tests. Stress-strain relationships obtained from tensile coupon tests were incorporated in the finite-element model. The verified finite-element model was used for a parametric study of aluminum alloy beams of square hollow sections. A comparison of the experimental and numerical bending strengths with the design strengths calculated using the current American, Australian/New Zealand, and European specifications for aluminum structures was presented. The bending strengths were also compared with the design strengths predicted by the direct strength method, which was developed for cold-formed carbon steel members. Design rules were proposed for aluminum alloy square hollow section beams based on the current direct strength method. Reliability analysis was performed to evaluate the reliability of the design rules.

Data sets in civil and structural engineering are often statistically challenging. This is because the data are from one-of-a-kind systems such as buildings and other large facilities, as opposed to replicated systems as found in most other fields of engineering. Special care is required in developing predictive models from such data. Herein a database of building natural period and damping is used to provide a rich context for analyzing one-of-a-kind systems. The database is statistically nonconforming in three ways. The data are nested, where measurements from different excitation sources are obtained for each building; the data set is unbalanced with measurements unevenly distributed among different building categories; and the variability is nonuniform. Furthermore, the number of possibly relevant building parameters is large. The goal is to develop a relatively simple, yet general approach for deriving predictive models based on such statistically nonconforming data sets. The approach is based on the statistical framework of generalized linear models and is structured in a manner to allow for engineering insights into the model. In the companion paper, it is shown how this approach can be applied to develop comprehensive models for building natural period and damping.

There are two goals in this paper: (1) the presentation of a comprehensive database of over 4,000 fundamental period and damping measurements from nearly 1,000 buildings; and (2) the derivation of predictive models for period and damping. The modeling approach presented in the companion paper, which accounts for the statistically nonconforming properties that characterize building data, is used herein to derive models which include two intrinsic sources of variability: that associated with multiple measurements from different excitations on the same building and variability between buildings that are nominally identical. The period models are in general agreement with those currently used, with model parameters that are in terms of height, material, and lateral force resisting system. The damping models show a decrease in damping as building height increases and, for steel and reinforced structures, a twofold increase in damping for earthquake excitations as opposed to other, lower amplitude excitations.

Reliability-based assessment of seismic performance of wood shear walls needs to consider many uncertainties arising from earthquake ground motions, material properties, carried mass, and analytical models. Using a response surface method with importance sampling, this paper presents a seismic reliability analysis on eight types of diagonal-braced and structural-panel-sheathed shear walls which are commonly used in modern post-and-beam wood buildings. A recently developed wood shear wall model was used to simulate the seismic response of the shear walls. The shear wall model parameters were calibrated and verified using a reversed cyclic test database. Peak wall drift was chosen as the performance criteria to estimate the failure probabilities of the walls with respect to two performance expectations. The results presented herein showed that the seismic reliability of the structural-panel-sheathed walls was higher than that of the diagonal-braced walls.

The benefits associated with structural control include the mitigation of undesired structural responses and reduction in the probability of damage to structural components during seismic events. Structural control systems in current use depend on extensive wired communication systems to connect sensors and actuators with a centralized controller. While wired architectures are appropriate when control systems are small, the cost and installation complexity of tethered systems increases as the control system grows large (i.e., defined by hundreds of nodes). Alternatively, wireless sensors are proposed for use in large-scale structural control systems to keep costs low and to improve system scalability. Wireless sensors are capable of collecting state data from sensors, communicating data between themselves, calculating control actions, and commanding actuators in a control system. However, bandwidth and range limitations of the wireless communication channel render traditional centralized control solutions impractical for the wireless setting. While computational abilities embedded with each wireless sensor permit fully decentralized control architectures to be implemented, strategic utilization of the wireless channel can improve the performance of the wireless control system. Toward this end, this paper presents a partially decentralized linear quadratic regulation control scheme that employs redundant state estimation as a means of minimizing the need for the communication of state data between sensors. The method is validated using numerical simulations of a seismically excited six-story building model with ideal actuators. Additional experimental validation is conducted using a full-scale physical realization of the six-story building. A wireless sensor network commanding magnetorheological dampers is shown to be effective in controlling a multistory structure using the partially decentralized control architecture proposed.