Journal of Structural Engineering

Eleven half-scale experiments were conducted to investigate the nonlinear reversed cyclic behavior of coupling beam subassemblages in a new type of hybrid coupled wall system for seismic regions. In this new system, coupling of concrete walls is achieved by posttensioning steel beams to the walls using unbonded posttensioning tendons. Different from conventional hybrid coupled walls, the coupling beams are not embedded into the walls. Each test specimen described in this paper includes a steel coupling beam and the adjacent concrete wall regions at a floor level. Steel top and seat angles are used at the beam-to-wall connections to yield and provide energy dissipation. The test parameters include the angle leg thickness, angle gauge length, beam flange cover plate, posttensioning steel area, and beam depth. The results demonstrate excellent stiffness, strength, and ductility characteristics of properly designed subassemblages under cyclic loading, with considerable energy dissipation concentrated in the angles. The specimens were able to sustain large nonlinear displacements without significant damage in the beams and the wall regions; demonstrating that these structural components would require little repair, if any, after a severe earthquake. The residual deformations upon unloading of the specimens were negligible due to the restoring effect of the posttensioning force.

This paper uses the experimental results presented in a companion paper to develop a seismic design approach for unbonded posttensioned hybrid coupled wall subassemblages. The test results are also used to verify and refine analytical models for the subassemblages. The proposed design approach includes guidelines for the selection of the coupling beam cross-section and the posttensioning steel area, as well as the design of the top and seat angles, their connections, and the design of the concrete wall contact regions.

Results from an experimental study on the seismic response of two hybrid RC column-to-steel (S) beam (RCS) roof level T-connection subassemblies are presented. The primary variables investigated were: (1) anchorage requirements in column longitudinal bars; and (2) confinement requirements in RCS roof level T-connections. In the first subassembly the column bars were anchored by providing a development length of $12\text{bar}$ diameters and an end mechanical anchor with twice the diameter of the bar, while connection confinement was provided by steel band plates wrapping around the column regions just above and below the steel beam. This specimen exhibited a stable response up to 4.0% drift, the displacement at which a column bar anchorage failure occurred, which led to excessive bar slip through the connection with the associated decay in stiffness and strength. In the second specimen the development length for the column longitudinal bars was increased to $20\text{bar}$ diameters and a larger bearing area for the mechanical anchors was provided through steel channel sections running transversely above the top flange of the steel beam. The behavior of the second specimen was characterized by large column inelastic rotations, good stiffness retention and energy dissipation capacity, and limited joint damage. Thus, experimental results indicate that properly designed RCS roof level T-connections exhibit a stable behavior under large inelastic deformation reversals, such as those induced by strong earthquakes.

The behavior of high strength concrete tied columns was investigated. A total of 11 columns were tested under monotonic loading having two different eccentricities. The main variables investigated were: (1) the mechanical ratio, (2) the spacing, and (3) the configuration of transverse reinforcement. Test results indicate that, under monotonically increasing axial load, the cross ties with a 135° hook at one end and 90° bent at the other end provide ductility comparable to ties supporting the intermediate longitudinal bars. Also the use of four different equivalent rectangular stress blocks (ERSBs) in predicting the ultimate strength of members made of high strength concrete was evaluated. It is concluded that calculations made using the ACI 318-02 stress block always gives an upper bound, whereas a lower bound is obtained by the CSA A23.3-94 approach. Capacities calculated using the ERSB proposed by Bae and Bayrak, and Ozbakkaloglu and Saatcioglu also agree quite well with the test results.

A new formulation for a robust, yet simple, uniaxial material model for reinforcing steel subjected to arbitrary loading is presented. The model provides steel stress as an explicit function of the total steel strain and can account for nonlinear monotonic envelope curves, i.e., with a yield plateau and nonlinear strain hardening, degradation of the yield limit as a function of the plastic strain history (Bauschinger effect) and strain hardening phenomena. Suggested model input parameters for typical reinforcing steels used in the United States are provided and the characteristic behavior of the model is illustrated through comparison with experimental results.

Fatigue design criteria are necessary to design steel corrugated web girders for highway bridges. The paper presents research on the fatigue life of steel bridge I-girders with trapezoidal web corrugations. Eight large-scale test girders were fabricated from HPS 485W steel and fatigue-tested in four-point bending. The web-to-flange fillet welds were made using semiautomatic gas metal arc welding (GMAW) or robotic GMAW. Fatigue cracks initiated in the tension flange at the web-to-flange fillet weld toe along the inclined web folds and adjacent bend regions, and propagated in the flange. The results demonstrate that steel corrugated web I-girders exhibit a fatigue life that is longer than that of conventional steel I-girders with transverse stiffeners. For design of corrugated web I-girders, the Category $B^{\prime }$ design curve of the AASHTO LRFD specifications is recommended for finite life fatigue design calculations, and a value of $96.5\mathrm{MPa}$ $\left(14.0\mathrm{ksi}\right)$ is recommended for the constant amplitude fatigue limit.

Most current design specifications employed throughout the world prescribe the use of so-called interaction equations for the case of beam–columns. Most often these interaction equations are optimized for use with the case of I-shaped cross sections bent about the major principal centroidal axis. The current study focuses on the case wherein an I-shaped member is loaded in compression and bending about the minor principal centroidal axis. It is shown that the current AISC interaction equations can be improved on in terms of their ability to predict failure in these types of members. Alterations to the existing AISC interaction equations are suggested for improving on strength predictions relative to this case.

Overlap tubular K-joints are generally regarded as having higher axial strength than similar simple gap K-joints due to their more direct load transfer between braces. Due to congestion at the nodes in steel jackets, overlap K-joints are used as key structural elements in many existing offshore platforms. However, relatively little experimental research has been done to quantify their strength and this has prevented engineers from taking full advantage of overlapping in the design of new structures as well as in the life extension assessment of existing platforms. This paper presents the results of a numerical study on gap and overlap K-joints under in-plane bending. The parametric study reveals new insight into how the behavior and strength varies across the practical range of geometrical parameters. Based on the finite element strength database, a capacity equation has been derived and presented herein using nonlinear regression analysis techniques.

A test program was performed on aluminum alloy columns of square and rectangular hollow sections that were compressed between fixed ends. In total, five series of tests were conducted that included columns with both ends transversely welded to aluminum end plates using the tungsten inert gas welding method, and columns without welding of end plates. The tests were performed over a range of column lengths in order to obtain a column curve for each series of test. Heat-treated aluminum alloys of 6063-T5 and 6061-T6 material were used for the column specimens. Nonwelded and welded material properties were obtained from various coupon tests. The column specimens failed by local buckling, overall buckling, as well as interaction of local and overall buckling. In addition, some welded columns failed in the heat-affected zone. The test strengths were compared with the design strengths predicted by the American, Australian/New Zealand, and European specifications for aluminum structures. Reliability of the design rules for transversely welded aluminum columns in the three specifications was examined using reliability analysis. It is shown that the design strengths predicted by the three specifications are conservative for the nonwelded columns. For the welded columns, the design strengths predicted by the American and Australian/New Zealand specifications are generally conservative, whereas the design strengths predicted by the European Code are relatively quite conservative. It should be noted that the design strengths predicted by the American and Australian/New Zealand specifications are unconservative in some cases for the welded columns.

An analytical study of buildings without and with seismic damping systems was conducted to investigate the dynamic response of secondary systems. The buildings were represented by steel frames designed by the procedures described in the 2000 NEHRP Recommended Provisions. Nonlinear response analyses were performed for the undamped frames and the damped frames equipped with different energy dissipation systems and for far-field, near-field, and soft-soil earthquake histories. The study shows that significant benefits may be provided by viscous damping systems in terms of reduced floor spectral accelerations and floor absolute velocities.

An earlier procedure that estimates the peak deformation in base-isolated buildings is extended to include overturning and thereby estimate the peak axial forces in individual isolators. Such tools can be used as part of a design procedure to predict and subsequently eliminate bearing tension by modifying the design. The procedure is based on nonlinear response history analysis of an isolated block using an advanced bearing model that incorporates the relation between axial load and bearing response, known as axial-load effects. Rocking of the structure and bearing axial-load effects are found to have little influence on the peak lateral bearing deformation; median response spectra are within 10% of those when rocking is neglected entirely. Furthermore, bearing axial-load effects can usually be neglected in determining the maximum and minimum bearing axial forces; cases are identified where the error in neglecting such effects exceed 10%. Because the structure has been modeled as rigid, the limitations of the procedure should be assessed for superstructure designs that allow significant structural deformation.

A sampled data control algorithm taking into account external excitations was developed for structural engineering applications. The structure was modeled as a continuous system controlled by a digital control system. In addition to the feedback gain obtained using a conventional control design method, a linear feedforward gain was proposed to improve control efficiency. In order to ensure a satisfactory control performance, an adaptive scheme was developed to adjust this feedforward gain in real time according to the measurement history. The conventional sampled data linear quadratic regulator design was used as a baseline to evaluate the effectiveness of the proposed control algorithm. Some practical issues such as the effects of sampling frequency, time delay and actuator dynamics were addressed. This method was validated on the examples of two multistory shear beam lumped mass models. The numerical examples show that combining both the feedforward and feedback terms can improve the control efficiency significantly as compared to the case with feedback control only. It is also shown that the improvement on the control efficiency obtained with the proposed method is still noticeable when time delay is considered and compensated properly.

A step-by-step procedure for linear analysis of building structures with rigid floor beams is developed and presented. The method outputs the maximum story drift envelope, and consists of a series of explicit formulas that do not require iteration. The basis of the method is an internally damped shearbeam model. Drift demand spectra are used to provide information on the ground motion hazard of the site. In addition to the drift demand spectra, two additional spectra are incorporated, which describe the higher mode behavior. To test accuracy, three sample buildings structures (3-, 9-, and 20-story buildings at Los Angeles) are analyzed for five sample earthquake records using the proposed procedure, and conventional time history analysis. A comparison of the results shows that the proposed procedure is accurate enough for quick evaluation of seismic performance.

A new procedure to derive the distribution of peak pressure and load coefficients from individual sample records is applied to wind tunnel records obtained from a generic flat-roof model tested at the Univ. of Western Ontario (UWO). The initial step of this procedure requires the identification of the appropriate marginal probability of these records. The corresponding distribution of the peaks is then obtained with the use of the standard translation process. Predicted load coefficients over variable tributary areas as determined from UWO records and based on a preset probability of non-exceedence are compared with the provisions of the ASCE 7 standard. Based on the open-terrain observations, the code provisions generally correspond to relatively low levels of nonexceedence of 84% or less. The ASCE 7 suggested method to estimate the peak pressure coefficients for the suburban environment based on those for the open terrain could be successfully applied to the UWO observations. This, however, is generally not true when performing a similar estimation procedure for peak load coefficients from the Clemson experiments.

A genetic algorithm is presented that can simultaneously optimize size, shape, and topology of skeletal structures, including both trusses and frames. The algorithm is unique because it finds multiple optimum and near-optimum topologies in a single run. The algorithm was executed on a bridge example where it found both traditionally recognized bridge topologies as well as less familiar topologies. It was also executed on two standard test problems as well as on a plane frame example. This algorithm presents the designer with more choices and more information than algorithms that converge to a single optimum design.

In Turkey, especially in the Eastern Black Sea region, many roofs collapse or damage due to heavy snowfalls exceeding the maximum loads for design have been experienced. Roof snow loads specified for structural design in modern standards are calculated as the product of a design ground snow load and the conversion factors that depend on the roof properties. The aim of this study is to create a map to give precisely the variation of the ground snow loads in the Eastern Black Sea region so that design of roof snow loads in the region can be estimated accurately. To accomplish this purpose, the probability plot correlation coefficient test was examined with three different cumulative distribution functions (Gumbel, Lognormal, and Weibull) to determine the best fit of the theoretical probability distribution with the snow data of 32 Turkish State Meteorological Service weather stations located in the Eastern Black Sea region and adjacent cities. After the 50-year ground snow loads were determined using the data collected from each station, the ground snow load map of the region was produced using a geographic information system. The approaches in Turkish codes of TS 498 and TS 7046 were also discussed along with the results obtained in the study.