Journal of Structural Engineering

Steel beams are susceptible to large deformation and capacity reduction when subjected to elevated temperatures. Determination of collapse loads for structural steel members under fire is critical for realistic assessment of system vulnerabilities. Such determination, however, should be conducted within a probabilistic framework that allows for the integration of various uncertainties influencing the behavior. In addition, the assessment should address realistic fire exposure that arises due to typical fire scenarios. To date, evaluation of the collapse load of beams in the presence or absence of uncertainties under nonuniform temperature is lacking. This paper presents a new analytical formulation, based on virtual work, for calculating the collapse load of beams subjected to nonuniform longitudinal temperature distribution. Moreover, a new probabilistic framework is devised to generate fragility surfaces for beams under combined fire and applied loads. Randomness in load and resistance including applied mechanical loads, compartment ventilation, compartment geometry, and thermal characteristics of surrounding surfaces are accounted for by including the statistical variations in these parameters. Comparisons of collapse loads calculated using the proposed approach and those calculated using a commercial finite element software show excellent correlation. The outlined framework can allow structural and fire engineers to rapidly evaluate the collapse load and mechanism of beams under a nonuniform temperature distribution.

A bond failure criterion, a wave propagation–based dowel action, and an inelastic, three-dimensional, damaged interface element are proposed. The present approach models the behavior of steel–concrete interaction under quasi-static loading and unloading. The decay of the steel–concrete interface is included for variable confinement levels. Existing bond models neglect either the bond path dependency or accurate unloading behavior. Both are addressed in this research. Existing methods model dowel action through empirical relations or contact-based algorithms. However, proper contact surfaces are tedious to define and difficult to stabilize because they may experience abrupt failures in cases involving major damage to concrete. To simulate dowel action, an alternative method to contact and empirical methods is proposed based on the concept of wave propagation. The present interface element is comfortably applied without remarkable complications. Comparison with test results from the literature demonstrates the validity of the proposed element.

Reinforced concrete beams tend to elongate after flexural cracking and yielding; however, the elongation is restrained by the surrounding structural components in a RC moment frame. Experiments were conducted on seven $1/2$-scale interior beam-column subassemblies to study the effects of axial restraint on the seismic performance of RC frame beams and beam-column joints without the presence of floor slabs. The test setup permitted applying axial restraint to beam ends and measuring the compressive axial force passively generated in the beams. Major test variables included beam flexural reinforcement ratio and axial restraining rigidity. Without axial restraint, the total beam elongation reached 3.75% of the beam height at 3% lateral drift. Under the considered levels of axial restraining rigidity, large compressive axial force developed in the beams, leading to an axial force ratio up to 0.25. The axial restraint increased both beam flexural stiffness and strength. Depending on the tensile reinforcement ratio, beam flexural capacity increased 40%–150% at 3% drift. Compared with the unrestrained specimens, the axially restrained specimens suffered greater damage in the beam plastic hinge regions and beam-column joints due to the dramatically increased shear demand, which can negatively impact the seismic performance of a RC frame.

A full-scale experimental study was carried out on a two-story steel frame to investigate the progressive collapse behavior of beam–column joints with cast steel stiffeners (CSS). The frame was divided into four vertical sections, and each section had a specific joint arrangement of (1) joints with CSS and (2) welded joints without stiffeners. Seven loads of increasing magnitude were applied to the frame. In each section, one of the columns had a device at the base that could be removed and replaced to simulate the failure and repair of this column; this column is referred to as an adjustable column. Strain changes and displacements were monitored during column removal. Moreover, three types of finite-element models were developed to simulate the test frame. The results showed that compared to a welded joint without stiffeners, joints with CSS reduced the strain change and deformation of the frame, as well as the dynamic strain magnification factor of strain change; concrete slabs played a significant role during the load redistribution in the sudden failure of the adjustable column.

Sliding isolation, as one of the modalities of base isolation, has demonstrated its value in seismic hazard mitigation. Conventional sliding isolation systems, however, may exhibit unacceptable large sliding displacements under severe earthquakes and may suffer a risk of unpredictable impact effect due to the insufficient isolation gap. A novel base isolation system that uses sliding hydromagnetic bearings has been proposed recently to overcome these shortcomings. These bearings comprise steel tubes with a pressurized internal fluid and attached permanent magnets, and slide over aluminum base plates also with attached permanent magnets. They minimize the friction between bearings and base plates, generate a damping force that reduces the bearings displacements to practical levels, and introduce a restoring force and a displacement constraint. In the present study, a sliding hydromagnetic isolator is designed, fabricated, and tested experimentally to assess its performance as a seismic protection system. Additionally, numerical simulations are carried out for quantifying the repulsive, damping, and friction forces involved. It is found from these studies that the applied loads on the hydromagnetic bearing does not produce fluid leakages, O-ring damage, or scratch marks on the base plates; the bearing’s friction coefficient does not exhibit a conventional friction-vertical load correlation and is therefore lower and more stable than in existing sliding isolators due to the effect of oil-solid interface; and the pressurized fluid significantly reduces the frictional force between bearing and base plate and facilitate thus the bearing’s sliding; the repulsive force increases dramatically with the bearing displacement and may effectively prevent bearings from sliding off their base plates.

This paper describes an experimental investigation of stub column tests of cold-formed steel C- and Z-sections with different stiffeners. The experimental program consisted of 30 fixed-ended column tests of 6 series of C-sections and 4 series of Z-sections. The configurations of stiffeners comprised simple edge stiffeners, simple lips with inward or outward return lips, and intermediate web stiffeners. The specimens were brake-pressed from high-strength zinc-coated grades G450 and G550 structural steel sheets. The material properties were measured through tensile coupon tests. The load-carrying capacities, load-end shortening responses, and failure modes of stub column specimens are presented and discussed in this paper. The test strengths and observed failure modes were compared with the design strengths and failure modes predicted by the Direct Strength Method. The comparison showed that the Direct Strength Method provides accurate and reliable design strength predictions for C- and Z-sections with different edge stiffeners but provides less accurate design strength predictions for C-sections with intermediate web stiffeners.

This paper reviews the experimental research on beam-to-column connections for composite special moment frames (C-SMFs) and presents an experimental database consisting of 165 tests conducted on these connections. The paper also summarizes the behavior of different types of connections and evaluates the connections using current standard requirements for composite beam-to-column connections. Based on the evaluations, double split-tee (DST) connections are recommended for C-SMFs with rectangular concrete-filled steel tube (CFT) columns, through-beam connections are recommended for C-SMFs with circular CFT columns, while connections using diaphragms are recommended for C-SMFs with both rectangular and circular CFT columns. Design provisions for the DST and through-beam connections are also provided, which include design equations to calculate the panel zone shear strength and flowcharts illustrating step-by-step design procedures. The paper will be useful and practical for engineers planning to design C-SMFs using current AISC seismic guidelines.

This paper summarizes key diaphragm response parameters observed during the shake table testing of a three-story precast concrete structure built at half-scale. The test structure geometry possessed an aspect ratio in plan and structural wall layout to form a diaphragm-critical condition. Each floor of the structure incorporated a different precast concrete diaphragm construction technique: untopped double tees, topped hollowcore, and topped double tees. A new seismic design methodology was used to design and detail the diaphragms. The test structure was subjected to input ground motions of increased intensity in a series. The series totaled 16 strong ground motions, consisting of design basis and maximum considered earthquakes for which the diaphragms were designed. The primary objectives of the testing program were to examine the behavior of precast diaphragms within the structural system, demonstrate the seismic performance of the precast diaphragms designed with the new design methodology, and calibrate analytical models for the wider research program. This paper describes diaphragm global and local responses. The responses were interpreted in terms of expected diaphragm performance. Design conclusions were drawn based on the observed diaphragm seismic response.

Modeling the various aspects of a disaster scenario and the interactions that occur between them is necessary to create a meaningful and high-fidelity simulation of the event. Using a publish-subscribe pattern for data management, a platform based on distributed simulation concepts is developed to address this challenge. The platform is designed to enable disparate models to interact together within the modeling environment. Each model, or simulator, represents one part of a disaster scenario and is viewed as a black box that subscribes to data from other simulators and/or publishes its results for other simulators to use. The framework is versatile and scalable and, as such, enables the developed system to grow through addition of simulators provided by other users. The new concepts are demonstrated through a case study of wind-induced progressive damage due to both the structural response of a building and external wind pressures. The potential of the framework for modeling community resilience is discussed.

A probabilistic calibration method is proposed for evaluating the accuracy and applicability of available deterministic models for the shear strength of shear-critical reinforced concrete (RC) columns. Shear mechanisms of RC columns were explored and a deterministic model for shear strength was developed based on the variable angle truss-arch model. Subsequently, a probabilistic shear strength model involving both aleatory and epistemic uncertainties was proposed based on the Bayesian theory and the Markov Chain Monte Carlo (MCMC) method. Meanwhile, the probabilistic characteristics of shear strength were determined according to the developed probabilistic shear strength model. Finally, a probabilistic calibration method was developed to evaluate the accuracy and applicability of available shear strength models under different conditions of the compressive strength of concrete, the transverse reinforcement ratio, the axial load ratio, the aspect ratio, and displacement ductility.

Drift loads on roofs are arguably the most important snow load from a structural engineering perspective in the US, as they are the governing load for most cases of snow related structural collapse. Design relations for both leeward and windward roof step drifts are currently available. They were based upon an analysis of roughly 300 case histories from insurance company files. The relations have been in use in the US for roughly the past 30 years. When first introduced, they were considered an improvement over the prior relationship wherein the peak drift load was simply a multiple of the ground snow load. That is, the inclusion of the upwind fetch was intuitively appealing. Note however that neither wind speed nor wind duration were in the aforementioned original insurance company files. Hence winter windiness at the site could not be included as a variable in the original regression analysis, upon which the current relations are based. The overall purpose of this paper is to introduce a new drift relation which incorporates a winter windiness parameter. It is shown that this new relationship improves the associated drift prediction capability in comparison to the current drift relations.

An innovative type of bolted beam-to-column connection by means of a Π-shaped connector welded by two flanges, one web and one end-plate is put forward in this study. Besides standard bolt holes, short-slotted bolt holes are considered in the flanges and web of the connector for the Π-shaped bolted connections. An experimental investigation of six Π-shaped bolted connections with the parameters of slotted bolt-hole sizes and treatment methods of the faying surfaces between the Π-shaped connector and the beam and two comparative connections under cyclic load was undertaken to provide insight into the cyclic behavior of this joint type. The results demonstrate that the presented Π-shaped connections are characteristic of large plastic rotation, good ductility, and high energy dissipation, meeting the seismic design provisions for stiffness, resistance, rotation, and ductility. A slightly slotted size of the bolt holes in the connectors can greatly improve the seismic performance of connections in terms of the rotation capacity, ductility, and energy dissipation, and consequently prevent or reduce the risk of local damage progression in the beam flanges and web through the small movement allowed for the beam end. This can increase the postearthquake reparability of steel structures with such connections. Treatment methods of the surfaces just affect the slip resistance, the corresponding displacement, and therefore the ductility of connections. Observations of test connection specimens show that the Π-shaped bolted connections have multiple failure modes and can be viewed as a combination of end-plate and T-stub connections. Based on this, a reliable and conservative formula for prediction of rotation capacity of this joint type is proposed.

Current shear stud fatigue capacity recommended for design of shear connectors is based on experimental data from push-out tests, and there is a need to experimentally evaluate the fatigue performance of shear connectors in steel-concrete composite beams. Therefore, this paper presents the results of four composite beams tested under fatigue with different degrees of shear connection. The fatigue testing was carried out to a million cycles under bending with two different amplitudes. The residual strength of steel-concrete composite beams was evaluated by performing static tests until failure after the completion of the fatigue tests for each of the specimens. The testing measurements included midspan deflection, slippage between the steel beam and concrete deck, shear connector strains, and strains in the concrete deck and bottom steel flange. Based on the experimental results, degrees of the shear connection between the steel beam and the concrete deck played a major role in controlling the longitudinal fatigue cracks in the concrete deck. The residual deflection and plastic slippage increased as the fatigue of cycles increased. This is caused by the damage in the concrete deck that developed around the shear studs, which led to a reduction in the composite stiffness. The shear connectors’ strain ranges gradually increased due to the accumulation of damage as the number of cycles increased. Based on the results of this paper, it is concluded that the current shear stud fatigue design curve predicts a lower fatigue life for the shear connectors.

Design requirements for passive fire protection of steel structures in the United States are primarily based on prescriptive approaches. Yet performance-based design has gained attention in recent years for its potential to unlock design solutions that are robust, cost effective, and applicable to complex architectural configurations. For example, prior research has shown that fire protection on selected secondary beam elements in composite floor systems is not necessary due to the development of membrane action in the concrete slab during fire. In addition, the evaluation of a structure’s performance under realistic fire scenarios, required in a performance-based approach, is enabled by recent developments in advanced computational modeling. However, currently there are no systematic guidelines to determine the reliability of a performance-based fire engineering design. This study provides a comparative analysis of the fire performance of a floor system designed following prescriptive and performance-based approaches. The floor system is adopted from a prototype steel–concrete composite office building. Further, a parametric study is conducted to investigate the effects of several parameters on the thermal-mechanical response including the modeling approach, fire curves, applied gravity loads, and hazard scenarios. Performance is measured using survival time at the structural system level, but also with predefined thresholds in deflection and reinforcement bar temperature. The results demonstrate that the performance-based design is robust, and verification of safety is not dependent on a particular demand value or performance measure. Most importantly, the performance-based design shows resistance when subjected to natural fires with large percentage fractile of fire load as well as in multihazard post-blast fire situations.

This study evaluates the economic viability of allowing for controlled inelastic deformation of special moment–resisting buildings during wind storms. This is achieved through a set of case studies involving steel buildings of varying heights made of special moment–resisting frames located in three different cities in the United States, with different intensity levels of wind and seismic hazards. The evaluation procedure utilizes the latest developments in the field of performance-based earthquake engineering but in a multihazard context. By comparing the capital costs of the buildings with the expected life-cycle losses, it is shown that controlled inelastic deformation of buildings under wind load may be economically advantageous, depending upon the properties of the structures and characteristics of both hazards. It is also shown that, irrespective of whether a building is designed following the current code provisions or allowing for controlled inelastic deformation under wind excitations in seismically active regions, neglecting one hazard during performance-based cost-benefit analysis of the building can result in very large errors in the predicted life-cycle losses.

This study developed a seismic design approach, which has no need of additional structural members, to uniformly distribute the peak interstory drift ratios along building heights. The generalized building model (GBM), which can effectively imitate a building, was used to reach this aim. The flexural-shear deformation factor of a GBM, $\alpha$, characterizes the deformation type of the corresponding building. Varying the deformation type of a building changes the distribution of the peak interstory drift ratios among stories. Therefore, the proposed optimization approach searches for the optimal value of $\alpha$, ${\alpha }_{\mathrm{opt}}$, so that the GBM’s normalized variation of peak interstory drift ratios is minimized. In response to ${\alpha }_{\mathrm{opt}}$, the member properties of the original building model are adjusted. One nine-story steel moment-resisting frame was used as the example building in this study. The elastic response spectrum analysis results showed that the peak elastic interstory drift ratios of the optimized building were less varied among stories compared with the original building. In addition, 13 of 20 nonlinear response history analysis results were consistent with the trends observed from the elastic response spectrum analyses. As a result, the effectiveness of the proposed optimization approach on the investigated building was confirmed.

Structural health monitoring (SHM) is gaining more and more interest in historical construction. It is a tool that controls, verifies, and informs about the condition or changes in the structure so that engineers can obtain reliable information for management and decision making. This article aims to present the monitoring systems developed during the propping system removal of several brickwork domes of a modernist church after strengthening work was performed on the domes. Given the need for exhaustive monitoring of the arch and dome movements, a monitoring strategy for an appropriate loading operation of these structural elements was developed. SHM was carried out using two instrumentation systems: the first one, based on temporary sensors, ensures local monitoring of displacements in domes and arches, and the second system, based on remote monitoring of several sensors, enables any unexpected structural behavior to be followed up before, during, and after the propping system removal of the domes. The systems made it possible to achieve centralized and reliable global monitoring of the structural elements of the building.

The concept of post-tensioned timber structures has been developed and implemented since 2005. The technology relies on unbonded post-tensioned timber elements to create a moment-resisting connection. After several years of extensive experimental testing, the first buildings were constructed in 2010. Although comprehensive research has been conducted on their behavior under static and dynamic loading, limited data are available about their long-term performance. In particular, there is limited understanding of how to estimate the long-term post-tensioning losses. This paper presents two years of experimental observations on a fully monitored post-tensioned timber building: the House of Natural Resources. Additionally, a numerical approach based on a moisture-dependent viscoelastic model is presented to simulate the post-tensioning trend over time.

Evaluation of the seismic performance of infill frame structural systems using equivalent strut models and pushover analysis has gained popularity due to the simplicity of the evaluation process. Various strut macromodels have been developed for both static and dynamic evaluation of the behavior of the infill frames when subjected to seismic loading. Comparative studies of existing models revealed varying degrees of accuracy, with more detailed models able to produce better results due to improved material characterization and geometric configurations. The variations in the accuracy of model predictions can be linked to the simplifications adopted and/or variances from material behavior modeling of the infill masonry and frame. This paper provides a new approach for developing equivalent strut material and geometric properties that incorporate the behavior of dominant stress zones within the infill. The proposed equivalent strut is represented by dominant stress zones, modeled as nonlinear springs with specific characteristics derived from the dominating stress, and it connects these stress zones either as series or parallel elements. A parametric evaluation of the proposed model was conducted using numerical methods. The performance of the proposed model was validated using experimental data taken from literature. The results show good correlation of the proposed model with the experimental data.

This paper presents experimental and analytical studies on the seismic behavior of beam-to-column connections with external diaphragms in the concrete-filled steel tubular (CFST) structures. Three cruciform joints between CFST column and H-shaped steel beam were tested subjected to a constant axial load on the column and a set of antisymmetrical cyclic loads at the beam’s tips to investigate the seismic behavior of the panel zone. A weak panel zone was specially considered in all three specimens to investigate the shear capacity of the panel zone. Key variables involved are the compressive strength of concrete and the axial load ratio. Three specimens all exhibited a shear failure in the panel zone due to local buckling. The strain distribution and load-displacement behaviors were discussed. In the analytical study, a computational model for the postbuckling shear capacity of panel zone was developed and checked for its applicability.

This paper illustrates the application of the database-assisted design (DAD) method to the wind design of high-rise buildings. The paper uses publicly available wind tunnel data and DAD procedures to compare responses to (1) corner winds and (2) face winds of a high-rise building of square cross-section supported by a central core column and four mid-side legs. The responses being considered consist of overturning moments, and of demand-to-capacity indexes (DCIs) of selected members, including multistory chevron braces. The analysis accounts for structural dynamics and second-order load-deformation effects. The results show that corner winds are less demanding than face winds, both globally (overturning moments) and locally (DCIs). The along-wind and across-wind overturning moments in the corner wind case are about 20% and 50% lower, respectively, than their counterparts in the face-wind case. The peak axial forces in the legs (peak refers to absolute value) and the peak DCIs in the mid-side mast columns (continuation of the legs) induced by corner winds are lower by 20%–30% than their counterparts due to face winds. The investigation confirms that the building code of the City of New York in effect in the early 1970s can be interpreted as meaning that the design for wind of structures with a square shape in plan may be performed by assuming the wind loads to act normal to a face of the building. The building analyzed in this paper is similar to the Citicorp Building (completed in 1977, later renamed Citigroup Center, now called 601 Lexington) and the results of the analyses presented herein suggest that a re-examination of the history of the Citicorp Building design and retrofit may be warranted.