Journal of Structural Engineering

Structural cables are widely used on large-scale structures such as bridges and stadia around the world. This paper presents a review of current research and development associated with the thermal and mechanical performance of structural cables when subjected to fire loading. The findings of this study highlight key knowledge gaps, which are subsequently used to propose the future research agenda. The particular focus of the review is on large-diameter cables, composed of both multiple strands and multiple wires, which are commonly used on structures of concern to civil and structural engineers. It is determined that the fire performance of structural cables is influenced by several factors such as the cable shape, diameter, metallurgical characteristics, uniformity of thermal exposure conditions, time-dependent effects of fires, and their terminal characteristics. The thermal expansion coefficients and reduction of material properties provided in EN 1992-1-2 and other research studies may not be appropriate for exposed prestressed cables because they were developed for tendons encased in concrete construction and creep is not accounted for. In particular, the material characterization at elevated temperatures may not necessarily be the same as that of normal high-strength steel owing to the different cold-working process of cable wires. Very limited studies have investigated experimentally or numerically the fire performance of structural cables with a shape and diameter different to the typically studied seven-wire strands used for concrete construction, leading to a number of knowledge gaps. The experiments conducted so far have shown that for unprotected cables a significant thermal gradient develops and early failure occurs with an increase in the cable diameter leading to an increase in its failure time. A range of simplified and advanced methods have been proposed in the literature; however, these generally lack extensive validation against experimental evidence. Numerical studies have shown that the presence of air cavities between strands can lead to larger thermal gradients in particular for unprotected cables because the cavity acts as insulation, unlike protected cables, where the temperature is more uniform within the section. The testing procedure of PTI DC45.1-12, which is the only available standard for stay cables, lacks extensive verification and validation to ensure that the aims of the standard are achieved. Further research is required to confirm that the two-phase procedure adequately captures all underlying physics under realistic fire exposures and its applicability for cables with characteristics different from the sample tested.

High-strength steel (HSS) and ASTM A1035 Grade 690, in particular, have been gaining popularity in the last two decades due to its considerably higher strength and corrosion resistance compared to conventional steel, i.e. ASTM A615 Grade 420. The stress-strain response of HSS is different from that of conventional Grade 420 steel because it lacks a well-defined yield point and yield plateau. Consequently, extensive research studies have been carried out to evaluate the performance of HSS reinforcement in structural concrete to examine the adequacy of current design provisions when a design yield strength of up to 690 MPa is used. In order to accommodate HSS reinforcement, the American Concrete Institute (ACI) released Design Guide for the Use of ASTM A1035/A1035M Grade 100 (690) Steel Bars for Structural Concrete (ACI 439.6R), whereas the AASHTO LRFD Bridge Design Specifications initiated the National Cooperative Highway Research Program (NCHRP) Project 12-77 to evaluate the provisions relevant to the use of high-strength reinforcing steel and other grades of reinforcing steel having no discernable yield plateau. This manuscript provides a synthesis of these two reports in a systematic way along with the other relevant research works. In addition, a comprehensive comparison between the requirements of ACI 439.6R, AASHTO, and ACI 318, relevant to the use of HSS reinforcement, is presented. Specified yield strengths for HSS, when used in different applications set by the three codes, are provided in a summary table. The provided table is intended not only for comparison purposes but also to facilitate the use of HSS reinforcement by designers because it contains all of the important and most recent relevant clauses in the three codes. Lastly, future research recommendations are proposed to revise and increase the specified yield strength of HSS in certain applications to enable designers to take full advantage of the potential benefits of HSS reinforcements.

The wind uplift resistance of wood roof sheathing panels has been the subject of many engineering studies due to the high probability that roofs are subjected to large uplift loads during intense wind events. However, few studies have focused on determining the out-of-plane wind resistance of wall sheathing for light-framed wood structures, which are integral to the construction of external walls in single-family residential housing. The objective of this research is to experimentally determine the out-of-plane wind resistance of wood framed walls as used in light-framed wood structural systems. Tests were conducted at the University of Florida on $1.22\times 2.44\mathrm{m}$ ($4\times 8\mathrm{ft}$) wood wall panels consisting of oriented strand board (OSB) and/or plywood sheathing nailed to wood studs. The panels were fabricated in accordance with the minimum requirements of the Florida Building Code. Each panel was subjected to a monotonic-increasing static pressure using a step-and-hold pressure test sequence until the panel failed. The mean out-of-plane structural resistance capacity of wood panels attained was 4.21 kPa (88.0 psf), and there was no statistical difference in performance between plywood and OSB sheathing panels. As expected, when results were compared with previous results from tests on wood roofs panels (conducted by the third author), the mean failure pressure of the wall panels was higher than the failure pressure of the roof panels. The ultimate nail-withdrawal load failure of the fasteners (by far the most common failure mechanism) in the wall sheathing was actually 21% lower than for the roof panels. Finally, the research has for the first time contributed experimentally determined fragility functions for wood wall panel systems and compared these to previously available fragilities for roof panels. Roof panels in the corners and along roof edges are more likely to fail in high wind events than are wall panels. The study shows that the failure probabilities for wall corner panels (in Zone 5) can exceed that of roof sheathing panels located in the field (Zone 1) of the roof. These results are important contributions for probabilistic modeling of wood framed building and for use in component-based wind-damage prediction models in hurricanes and in tornadoes.

Mass timber panels (MTPs) are a new generation of engineered wood panels that are available in large plane dimensions to facilitate fast floor construction with the obvious environmental benefit of being from a renewable material. In floor construction, concrete slab or topping is often applied over the MTPs to improve various performance attributes, including structural, acoustic, and vibration serviceability. A mass timber panel-concrete (MTPC) composite floor system often consists of a MTP connected to the concrete layer with mechanical fasteners and a sound insulation layer in between. The capacity of this type of composite system mostly depends on the strength of the connection, and often tests are performed to characterize connection properties required for structural design. In lieu of testing, analytical models can be developed to calculate connection properties based on component properties. To that end, two analytical models were developed for solid and layered timber by characterizing all possible kinematical failure modes for directly predicting the strength of a connection with inclined screws and an insulation layer. According to Johansen’s yield theory, the strength of a laterally loaded connection is controlled by the dowel-bearing effect of the fastener in timber, but joints with an inclined screw have a more complex behavior because of the combined bearing and withdrawal action of the screw. In the developed models, both the dowel-bearing and withdrawal action of the screw are considered along with the bending capacity of the screw and friction between the members. Both models were experimentally validated with a wide range of material properties. It was found that the models are capable of predicting the mode of failure of a connection and the load-carrying capacity within 10% of the experimental value.

The assessment of anticipated losses due to damage of both structural and nonstructural components is now recognized to be a key component in the performance-based seismic design or retrofit of buildings. Current performance-based seismic loss estimation procedures are building case-specific, and they do not easily allow for the integrated optimization of structural and nonstructural interventions in a particular building. The main objective of this paper is to develop a general optimization procedure within the performance-based earthquake engineering (PBEE) framework developed by the Pacific Earthquake Engineering Research (PEER) Center and implemented through the FEMA P-58 methodology. Available optimization procedures that can be applied to the PEER-PBEE framework are first reviewed, leading to the selection of the genetic algorithm for this purpose. The implementation of the genetic algorithm within the PEER-PBEE framework, considering integrated structural and nonstructural seismic upgrades, is described. The seismic retrofit case study of a 3-story steel moment-resisting frame archetype building is then conducted for the three different target metrics included in the PEER-PBEE framework: (1) an economic target metric, (2) a downtime target metric, and (3) a casualty-reduction target metric. The results of the case study indicate how this optimization process, based on the genetic algorithm, quickly and reliably converges to different allocations of resources for structural and nonstructural components depending on the target metric selected. Finally, a parametric study on the effects of the owner’s expected internal rate of return and building occupancy time on the optimum retrofit solutions illustrates the utility of the optimization process beyond the single case study.

Although extensive bond models have been developed for use in the numerical simulation of uncracked reinforced concrete, no simplified method exists providing satisfactory accuracy and efficiency for pre-cracked concrete. This paper intends, therefore, to explain bond failure mechanisms in pre-cracked concrete—as compared to intact concrete—using a simplified theoretical model. The main bond failure mechanisms considered in this study involve: (1) crushing a wedge-shaped concrete block using reinforcing bar ribs, and (2) tearing off the concrete between two adjacent ribs. Based on these scenarios, analytical expressions are derived to predict the average bond strength for uncracked concrete, in which the bearing angle, the rib face angle, the rib height, the rib spacing, and the friction coefficient between surfaces are the key parameters. A modified version of this model is proposed to predict the maximum bond strength of rebars embedded in pre-cracked concrete by introducing a reduction factor of surrounding confinement caused by the pre-cracking phenomenon. An experimental program was also conducted to validate the proposed models. Experimental results emphasize the crucial impact of the pre-cracking phenomenon on both the bond strength and the failure pattern. Analysis results show that the bearing angle and surrounding confinement by concrete cover are crucial parameters controlling bond failure of rebars in pre-cracked concrete. The results also indicate that as the crack width corresponding to the low confinement increases, rib sliding is expected to occur as an illustration of weak interfacial strength. The proposed bond mechanism models are also in good agreement with the experimental observations.

This paper first proposes a high-strength (HS) steel-moment frame structure with replaceable damage-control fuses, which utilized the very large elastic deformation capacity of HS steel and damage-control functions of fuses. The damage is basically concentrated to fuses even under very rare earthquakes, and the damaged fuses can be easily replaced after earthquakes. A test program including 11 full-scale specimens was carried out to study the seismic behavior of HS steel beam-to-column joints either with or without fuses. The investigated parameters include different types of fuses, different thickness of the fuse plate, and different ways to strengthen the out-of-plane stiffness of fuse plates. The failure mode, deformation progress, hysteretic responses, skeleton curves, bending strengths, and energy-dissipation capacity of these full-scale tests were reported, analyzed, and discussed. Test results showed that the HS steel joints with fuses were able to provide large damage-free deformation. The application of proposed fuses in the HS steel-moment frame effectively shifted the damage from beam-to-column welds to replaceable fuse plates. Meanwhile, the shape of fuse plate exhibited marginal influence, but its thickness has significant influence on the T-joint. Formulas were proposed to estimate the yielding and ultimate bending strength of T-joint with fuses. Finally, a step-by-step procedure is given to design the proposed fuses.

The seismic behavior of elliptical hollow section (EHS) beam-column members was examined in this study. A comprehensive experimental investigation was conducted on 14 concrete-filled and four bare steel EHS beam-column specimens under combined compression and cyclic bending. The main testing parameters were tube wall thickness, axial load ratio, and bending direction. Among other findings, it was revealed that the bare steel specimens failed in local buckling, whereas the concrete-filled members exhibited more diverse failure modes, including local buckling, a fracture of the steel section following local buckling, and an abrupt fracture of the steel section with no evident local buckling. The failure modes largely depended on the bending direction. Due to the beneficial effects of material hardening and steel-concrete interaction, the moment resistance obtained from the test was consistently larger than the predicted plastic moment resistance. In addition, the specimens exhibited satisfactory ductility and energy dissipation performance, and both characteristics could be effectively improved by either increasing the compactness of the steel section or by applying the concrete infill. The strength, ductility, and energy dissipation characteristics of the specimens subjected to biaxial bending fell between the cases of major-axis and minor-axis bending. Based on the test results, the existing design methods for predicting the moment resistance of the considered members were evaluated. A preliminary design equation for ductility prediction was also proposed.

This paper examines the influence of framing action and slab continuity on the hysteretic behavior of composite steel moment-resisting frames (MRFs) by means of high-fidelity continuum finite-element (CFE) analyses of two-bay subsystems and typical cruciform subassemblies. The CFE model, which is made publicly available, was thoroughly validated with available full-scale experiments and considers variations in the beam depth and the imposed loading history. The simulation results suggest that beams in subsystems may experience up to 25% less flexural strength degradation than those in typical subassemblies. This is because of local buckling straightening from the slab continuity and framing action evident in subsystems. For the same reason, beam axial shortening attributable to local buckling progression is up to five times lower in subsystems than in subassemblies, which is consistent with field observations. While the hysteretic behavior of interior panel zone joints is symmetric, exterior joint panel zones in subsystems experience large asymmetric shear distortions regardless of the employed lateral loading history. From a design standpoint, it is found that the probable maximum moment in deep and slender beams ($d_b\ge 700\mathrm{mm}$) may be up to 25% higher than that predicted by current design provisions with direct implications to capacity design of steel MRFs. The 25% reduction in the shear stud capacity as proposed by current seismic provisions is not imperative for MRFs comprising intermediate to shallow beams and/or featuring a high degree of composite action ($\eta >80\%$) as long as ductile shear connectors are employed.

This paper investigates the influence of the cylindrical charge characteristics, i.e., the length-to-diameter ratio, orientation, and initiation configuration, on the blast loads (peak overpressure and maximum impulse). Three different initiation configurations were considered, i.e., at the center, at one end, and at both ends. Numerical models were developed for the spherical and cylindrical charges. The numerical results were analyzed along the gauge lines at different azimuth angles around the cylindrical charges. Some important observations were made. Neglecting the charge shape effect, the peak overpressure (maximum impulse) in the near field generated from centrally initiated cylindrical charges can be underestimated by a factor as high as 3.0 (1.9). Therefore, the cylindrical charge shape should be explicitly modeled in the numerical simulations for the blast-resistant design of protective structures subjected to near-field detonations. It is confirmed that the shock front heals into a spherical one in the far field. Hence, the blast loads generated from the spherical charge of the same mass can be used. In addition, the influence range, beyond which the charge shape effect can be neglected, is $5.7\mathrm{m}/{\mathrm{kg}}^{1/3}$ for the impulse, which is about twice that for the overpressure ($3.2\mathrm{m}/{\mathrm{kg}}^{1/3}$). In general, the maximum values of blast loads resulted by the three considered initiation configurations can be sorted in descending order, i.e., $\text{at}\text{one}\text{end}>\text{at}\text{the}\text{center}>\text{at}\text{both}\text{ends}$. The initiation configuration at one end (at the center) has the largest influence range of $3.9\mathrm{m}/{\mathrm{kg}}^{1/3}$ ($5.7\mathrm{m}/{\mathrm{kg}}^{1/3}$) for the overpressure (impulse) among the three considered initiation configurations.

This paper reports on an experimental study of six full-scale partially grouted masonry walls (PG-MWs), and evaluated the accuracy of international design codes based on the test results. The main test parameters were the horizontal spacing between vertical grout cells and the horizontal reinforcement ratio. All test walls had identical thickness, height, and length, and were constructed using concrete masonry units (CMUs). The walls were subjected to a constant axial load and incrementally increasing in-plane lateral cyclic displacement applied to the top of each wall. The horizontal spacing between the vertical grouted cells ranged from 610 mm (24 in.) to 1,219 mm (48 in.), and the horizontal reinforcement ratio ranged from 0.127% to 0.254%. Results showed that the strength of the PG-MWs was very sensitive to the spacing between vertical grouted cells, and that the contribution of shear reinforcement to the shear strength was lower than what is commonly adopted in design codes, especially for the walls with higher ratios of horizontal reinforcement. The accuracy of current design codes in predicting the strength of the tested walls was investigated, and possible modifications were proposed. Disregarding the effect of grout spacing resulted in an inaccurate estimation of the strength in the current design codes. Consequently a simple grout spacing modification factor was introduced which could significantly improve the strength prediction in PG-MWs. Using a test database of PG-MWs collected from the literature, it was shown that the proposed method could effectively improve the shear strength prediction and reduce the scatter of the predicted strength.

The enhanced shear strength of transversely stiffened shear panels compared to the strengths of the unstiffened has been well recognized in the Specification for Structural Steel Buildings (AISC 360) and the North American Specification for the Design of Cold-Formed Steel Structural Members (AISI S100). The requirements for the transverse stiffeners to secure such strength enhancement was similar in early versions of the two specifications, but the Specification for Structural Steel Buildings has recently adopted a whole new approach to reflect more closely the actual mechanistic behavior of stiffeners in the context of steel plate girders. The paper presents a series of shear tests on cold-formed steel beams with various transverse stiffener sizes followed by finite-element nonlinear collapse simulation to demonstrate the applicability of transverse stiffeners in the context of cold-formed steel members. The experimental series consists of six tests on 200-mm-deep C-shaped sections having angle intermediate transverse stiffeners screw-fastened to the web also reveals the conservatism of the transverse stiffener design rules in the AISI S100-16 specification. Based upon the experimental and numerical results, the paper recommends the adoption of the corresponding provisions taken from the AISC 360-16 specification.

Mass plywood panels (MPP) are a new engineered wood material produced by Freres Lumber Company, Inc. in Lyons, Oregon. This paper presents a study of the lateral-load response of MPP walls connected to different base conditions by two hold-downs on each face. Four base conditions were designed and tested: (1) steel beam, (2) MPP base, (3) a proof of concept MPP base with a bearing plate at each toe of the walls, and (4) MPP base with toe screws. These varied base conditions represent balloon frame and platform constructions as lower and upper bound responses, respectively. The impact of base conditions on the lateral-load response was investigated through a series of in-plane cyclic tests on 76.2 mm thick, $\mathrm{1,220}\times \mathrm{2,440}\mathrm{mm}$ MPP wall panels. The test results followed idealized upper and lower bound responses and showed potential for using MPP wall panels as a lateral force resisting system for structures. Using the test results, two models (ASCE 41-13 trilinear and Pinching4 hysteretic) are calibrated and utility of model parameters are illustrated.

Testing of 21 single-sided partial-joint-penetration (PJP) groove weld specimens was performed to evaluate the effects of three parameters: (1) root opening, (2) weld reinforcing profiles, and (3) directionality of loading. The specimens consisted of either 25-mm (1-in.) or 51-mm (2-in.) thick ASTM A572 Gr. 50 (345 MPa) steel plates welded in the shop with E70T-1C/9C electrodes using gas shielded flux-core arc welding (FCAW-G). Most of the specimens (18) were cruciform in shape and loaded the PJP groove weld in the transverse orientation. Three remaining specimens were orientated such that the PJP weld was loaded longitudinally. Most of the specimens (16 out of 21) failed through the PJP weld. The five remaining specimens failed in the plates or fillet welds outside of the tested PJP welds. The resulting failure load is then compared to the design equations of AISC 360. The observation that the reinforcing fillet did not enhance the strength of the PJP-welded joint is explained by the transition of the tensile fracture surface to shear fracture surface.

Installation of open-joint roof paver systems with pedestals on flat roofs leaves a cavity between pavers and roof deck, allowing the development of cavity pressure. The cavity suction on roof pavers reduces the net wind uplift and overturning. This study conducted pressure measurements of the paved roof of a large-scale building model using the University of Western Ontario’s WindEEE Dome. The air-permeability factor was evaluated as a ratio of directional envelopes of peak net-pressure coefficients of the paved roof to peak external pressure coefficients of the bare roof. The air-permeability factors varied with pavers’ location across the roof, and wind uplift load decreased 40% for pavers in the corner and edge roof zones. Moreover, pressure tests conducted on roofs with high perimeter parapets showed that air-permeability effects of pavers and shielding effects of perimeter parapets can be treated as independent aerodynamic factors for evaluating the net uplift wind loads on rooftop pavers. Therefore the overall effect can be obtained simply by multiplying the roof pressure on bare flat roofs by air-permeability and parapet factors.

Braces made of square hollow structural sections (HSSs) in special concentrically braced frames (SCBFs) are likely to fracture under expected demand levels. This study, therefore, examines the additional seismic demands on column splices induced due to brace fracture. First, the study introduces the brace fracture models that were utilized for predicting fractures of square HSSs. Second, the selected 5-story and 13-story SCBFs with two different splice locations permitted in the seismic design practice were assessed under an ensemble of ground excitations by introducing different fracture prediction models. Third, the splice demands in terms of axial force, bending moment, and their combined effects were discussed with and without considering brace fracture. The study finds that brace fracture has a negligible impact on tensile force demands on column splices. In return, flexural demands, and the combined tensile force and flexural demands, are shown to be strongly influenced by brace fracture. It is also shown that the splices in SCBFs are vulnerable to undergo inelastic deformations when a brace fracture occurs, which is against the intended design objective.

The load transfer mechanism in cold-formed steel (CFS) bolted moment connections is mainly through the bolt group in the web of beam elements, which may lead to relatively large bimoment and warping deformations. While the bimoment effects can be considered in the Direct Strength Method (DSM), ignoring the fact that the bolt-group length in the conventional design process can lead to nonconservative solutions. This paper presents an alternative analytical design approach using Eurocode 3 (EC3) effective width method to determine the ultimate flexural strength of CFS bolted moment connections by considering bimoment effects. The results compare very well with previously published experimental test data as well as detailed finite-element models developed in this study. It is shown that a short bolt-group length may lead to up to 25% reduction in the flexural strength of the CFS bolted connections. However, a longer bolt-group length generally results in a moment capacity almost equal to the flexural strength of the CFS channel section. Shape optimization is then conducted using a genetic algorithm (GA) to improve the flexural capacity of the connections by taking into account the bimoment effects. The main design variables are considered to be the relative CFS beam cross-sectional dimensions, while the plate slenderness and dimension limits suggested by EC3 as well as a number of manufacturing and practical end-use constraints are incorporated as design constraints. It is found that, compared with standard cross-sectional dimensions, the optimized sections can improve the flexural strength by as much as 36% for a bolt-group length equal to the depth of beam element.

The ductility of a beam is important in reinforced concrete member design at the ultimate limit state, especially in resisting dynamic loads such as those from earthquakes or blasts. Concrete confinement reinforcement, such as stirrups or tubes, are widely used in structures and can significantly enhance the ductility of concrete beams. However, this confinement effect is normally ignored in current design standards, limiting the ability to design specifically for ductility or to estimate the ductility of existing structures. In this paper, a novel concrete passive stress/strain relationship based on the application of partial interaction and shear friction theories is simplified to a rectangular stress block for flexural analyses. This confined concrete stress block is then applied to quantify the moment/rotation of a hinge where both the confinement of the concrete and hinge lengths are quantified through mechanics. The aim of this paper is to provide a mechanics based approach for quantifying the ductility of RC beams that can be used to develop simple design rules without the need for large amounts of member testing.

A back analysis method based on multioutput least-squares support vector regression machine (MLSSVR) and improved differential evolution algorithm (IDE) is proposed to estimate multiple mechanical parameters of concrete dams. Based on the uniform design method, representative combinations of mechanical parameters are generated. Using these combinations, calculated hydrostatic displacement component differences are obtained through the finite-element method (FEM). The calculated hydrostatic component differences and mechanical parameters are then used to train MLSSVR models, with the model parameters selected by IDE. This allows for establishing the mapping relationship between dam displacements and corresponding mechanical parameters. The hydrostatic component differences separated from prototype observed data are then substituted into the model to obtain back analyzed mechanical parameters. If predefined terminating conditions are not satisfied, new combinations are added to the training set, and the training process continues until satisfied. The proposed back analysis method was successfully implemented to the Jinping I arch dam. Results indicate that the proposed method has high precision and strong generation ability.

The local instability of steel plates is the predominant seismic failure mode for steel bridge piers. However, the application of shell-based finite-element (FE) models in an engineering design presents significant difficulties, and existing simplified calculation methods have their own insufficiency. In this paper, single-column square-section steel bridge piers were studied to propose an equivalent hysteretic model for the fiber elements of steel piers that can consider the local instability behavior of steel plates. Using the length of the half-buckling wave at the bottom of pier as the length of effective damaged zone ($L_d$), average stress–strain curves of the cross-section points over the $L_d$ were extracted from the calculation results of a fiber-shell hybrid model. The two-surface constitutive model of material was modified by introducing the equivalent elastic modulus and skeleton curve and reducing the bounding surface radius to make the hysteresis and average stress–strain curves coincide. An equivalent hysteresis model for the fiber elements in the effective damaged zone was proposed, and the process for identifying model parameters was introduced. Using Q345qC structural steel as an example, regression formulae for equivalent hysteresis model parameters were established. Based on the comparison with calculation results of a hybrid element model, the improved fiber model can accurately calculate the seismic response of a structure under horizontal unidirectional or bidirectional earthquakes and significantly save calculation cost. The proposed equivalent hysteresis model can provide a method for calculating the elastoplastic seismic response with a high accuracy and good availability for the seismic design of steel piers under horizontal bidirectional earthquake actions.

This research focuses on evaluating the seismic properties of corrosion-damaged RC columns strengthened by bonded steel plate (BSP) and high-performance ferrocement laminate (HPFL). The mechanical properties are experimentally investigated based on various combined loading cases. A total of nine square RC columns with identical structural properties are produced, and five of them are corroded and poststrengthened with BSP and HPFL. The effects of the combined loadings, bearing capacity, failure mode, skeleton curves, hysteretic curves, ductility, stiffness degradation, energy dissipation, and plastic hinge characteristics are analyzed. In particular, a new method for calculating the skeleton curves under different failures modes is proposed and agrees well with the experimental results by comparison.

This paper presents the results of full-scale flexural testing of deconstructable composite beams that are part of a sustainable composite floor system. In this system, steel beams are attached with precast concrete planks utilizing deconstructable clamping connectors to achieve composite action as well as design for deconstruction. The beam specimens were designed with varied levels of composite action and demonstrated very ductile behavior. It is shown that the elastic stiffness of the beams can be conservatively estimated using a lower-bound moment of inertia, and the ultimate flexural strength of the beams closely matches the strength predicted by standard design provisions of the American Institute of Steel Construction (AISC). The AISC effective-width provisions can be utilized to calculate the effective widths of the deconstructable composite beams. Under moderate loading, the specimens behaved as composite beams with full interaction due to the friction generated at the steel–concrete interface.

This paper presents the results of an experimental investigation into the punching shear behavior of an innovative simple and effective shear connection method between steel-tubular column and flat-concrete slab using welded shear studs, steel plates and bent-up rebars. Two full-scale tests were carried out, one on a center column–slab assembly and one on a corner column–slab assembly. Detailed numerical simulations of the two tests were performed to supplement the experimental observations to provide full explanations of the behavior of the shear connection system, including crack initiation, final crack position, and punching shear critical perimeter. The results indicate that for calculating punching shear resistance of the slab, the proposed shear connection system works as an enlarged column. The enlarged column dimension can be taken as the distance between the lower bends of the bent-up rebars. The experimental results of critical punching failure perimeter and punching shear resistance agree with those calculated using the the European standard for the enlarged column for both tests. The punching shear resistance for the new shear connection system is demonstrated to be sufficient for practical applications.

This article describes an extensive experimental program that was conducted to characterize comprehensively the nonlinear sectional force-displacement response of bolted steel angle semirigid connections. A total of 139 tests were performed to investigate the influence of angle thickness, position of the column bolts, bolt grade, and bolt pretension. The effect of loading sequence was also examined by applying static and dynamic monotonic and cyclic loading. The observed behavior and failure modes are described. Key parameters defining the load-deformation response of the specimens are quantified, including yield strength, initial stiffness, sharpness of the transition between initial and yielding phases, postyielding stiffness properties, peak forces, and deformations at failure. Monotonic test results were used to define a four-parameter power model that reproduces the force-deformation response of the angles. The influence of each geometric parameter, the bolt type, and the loading protocols on the behavior of the bolted angles was also described. The tests showed that strength and stiffness increased when angle thickness increased, and decreased when the distance between the heel and the column bolt increased. The force-deformation response of monotonic tests can be used to predict the backbone of cyclic and seismic responses. The angles subjected to cyclic and seismic tests, however, developed stable hysteretic response characterized by gradual softening, strength degradation, and smaller ultimate deformations compared with those subjected to monotonic loading.

In this study, five double-plate connected concrete-filled steel tube (CFST) composite columns were tested under a uniaxial eccentric compressive load. The test phenomena, load–axial displacement curves, strain analyses, and column curve analyses were examined. The effects of different eccentricities and loading directions were compared with respect to the double-plate connected CFST composite columns under uniaxial eccentric compressive load. ABAQUS 6.14-4 software was used to develop models of the double-plate connected CFST composite columns. Load–axial displacement curves and failure modes obtained from the finite element analysis (FEA) models and tests were compared in this study. The FEA results, which were in good agreement with the test results, can well simulate the uniaxial eccentric compression performance of double-plate connected CFST composite columns. Based on the superposition principle and calculation method of the slenderness ratio as per the Standard for Design of Steel Structures (GB50017-2017), the authors proposed a formula for the stable bearing capacity of the double-plate connected CFST composite columns. In addition, a new theoretical formula for the bearing capacity of the uniaxial eccentric compression specimen was proposed. A comparison between the calculated N–M correlation curves and the test and FEA results indicated that the calculation results exhibited good safety and can provide ideas for engineering design.

This paper presents the first experimental and numerical investigation of the full-range shear behavior of a nonprismatic overhanging beam with concrete slabs and steel trapezoidal corrugated webs. The results showed that the traditional calculation hypothesis that the steel trapezoidal corrugated webs resist the total vertical shear in a section was not applicable to nonprismatic members. The inclined bottom concrete slab shared a significant portion of the shear force in the region with a larger hogging moment; thus, the actual shear force experienced by the steel trapezoidal corrugated webs subsequently decreased. The dominant factor of the shear force redistribution between the concrete slabs and the steel trapezoidal corrugated webs was the Resal effect, which was also a crucial factor contributing to the differences in shear behavior between prismatic and nonprismatic beams with steel trapezoidal corrugated webs. In addition, eigenvalue buckling analyses of geometric imperfections and nonlinear buckling analyses using the Riks algorithm were carried out for a nonprismatic beam with steel trapezoidal corrugated webs considering material and geometric nonlinearities. The experiment also indicated that the cracking and crushing of the concrete slabs was another important factor affecting the ultimate bearing capacity of the test beam. A significant finding that may invalidate intuitive knowledge was that the shear buckling of the steel trapezoidal corrugated webs did not occur at the support section, although this section bore the maximum shearing force and bending moment and had the highest section depth. Because of the Resal effect, the maximum shear stress and shear buckling in the steel trapezoidal corrugated webs were located at a certain distance from the support.

Controlled rocking steel braced frames (CRSBFs) are low-damage lateral force resisting systems that mitigate structural damage through a controlled rocking mechanism. Critical to the safety and low-damage nature of these systems are the floor-to-frame connections that allow the CRSBF to rock without imposing the associated uplifts on the adjacent gravity framing. This paper introduces three potential connection details through which the seismic forces are transferred to the CRSBF as the primary lateral force resisting system. The connections were tested at 60% scale between a one-story CRSBF subassembly and representative tributary gravity framing, with inertial and restoring forces simulated using hydraulic actuators for cyclic static testing. The experimental results show that all three connections are able to transfer the necessary loads while undergoing the displacements that are expected during a large earthquake, with some differences in the resistance that develops. These results are complimented by numerical simulations that are shown to be in good agreement. Design recommendations for the three proposed connections are also presented.

Partially restrained beam-column connections such as bolted double web with top and seat angle connections can be used in the gravity load system of steel buildings to develop secondary moment frame action and enhance seismic collapse prevention. To assess the gravity beam-column connection role in lateral reserve capacity, a comprehensive test program was conducted at Ecole Polytechnique Montreal to characterize the nonlinear hysteretic behavior of bolted angle connections subjected to simultaneous gravity shear and rotational demand from the moment frame action. Fifteen full-scale beam-column subassemblages of four different geometries were tested to characterize their deformation pattern and failure modes. The influence of the following critical parameters on the connection hysteretic behavior was investigated: the top and seat angle geometrical parameters, the beam and column sections, the gravity shear load, and the loading history. The envelope curve of each test was mathematically characterized to numerically reproduce the connection behavior. The tested connections exhibited large ductility and significant moment capacity, which translated into a substantial energy dissipation capacity. The use of bolted angles for gravity framing connections could thus be beneficial by enhancing the reserve capacity and providing lateral resistance to mitigate seismic building collapse.

An extensive series of experiments in a large boundary layer wind tunnel (BLWT) was carried out at the University of Florida (UF) Natural Hazard Engineering Research Infrastructure (NHERI) Experimental Facility to investigate the effects of upwind terrain on surface pressures acting on a generic low-rise structure. Three scaled models of the Wind Engineering Research Field Laboratory (WERFL) building were subjected to 33 boundary layer flows through precise control of an automated roughness grid called the Terraformer. The roughness configuration of the Terraformer was adjusted to simulate a variety of upwind terrains ranging from marine to dense suburban. Surface pressures were obtained for the three models (1:20, 1:30, and 1:50), three wind directions (0°, 45°, and 90°), 16 roughness element heights, and two element orientations (wide and narrow edge windward). The complete dataset is publicly available on the NHERI DesignSafe cyberinfrastructure (CI) web-based research platform. The experimental data records can be reused to study the effects of upstream terrain on the aerodynamic loading acting on low-rise structures. Further, the inclusion of different model sizes enables investigation of scaling effects in the BLWT. The DOI for the dataset is 10.17603/DS2W670.

Due to the high-frequency content of ground motions in central and eastern United States (CEUS) sites, the US Nuclear Regulatory Commission has modified their regulatory guides to consider these effects in the evaluation of nuclear power plants. Ground-motion response spectrum at CEUS has high frequency content as opposed to a typical design spectral shape. The purpose of this case study is to evaluate the seismic response of squat RC walls subjected to seismic ground motions with response spectra having acceleration amplifications in the high-frequency zone. Incremental dynamic analyses are performed on three selected RC walls subjected to acceleration time histories compatible with spectra for CEUS sites. Several performance indicators are evaluated at different damage levels such as drifts, spectral accelerations at fundamental periods, and base shears. Fragility functions are developed to have a better idea of the damage caused by the ground motions with different spectra in the walls. Maximum drifts and the highest probability of exceeding the drift limits studied in this paper were obtained with the lowest-frequency-spectrum case.