Journal of Structural Engineering

Computationally efficient macromodels are developed for investigating the progressive collapse resistance of seismically designed steel moment frame buildings. The developed models are calibrated using detailed finite-element models of beam-column subassemblages and account for the most important physical phenomena associated with progressive collapse. The models are utilized to compare the collapse resistance of two-dimensional, ten-story steel moment frames designed for moderate and high seismic risk according to current design specifications and practices. The simulation results show that the frame designed for high seismic risk has somewhat better resistance to progressive collapse than the system designed for moderate seismic risk. The better performance is attributed to layout and system strength rather than the influence of improved ductile detailing. The alternate path method is shown to be useful for judging the ability of a system to absorb the loss of a critical member. However, it is pointed out that the method does not provide information about the reserve capacity of the system and so its results should be carefully evaluated.

The potential for progressive collapse of a typical reinforced concrete (RC) moment frame structure initiated through the loss of one or more first-story columns is numerically simulated using a macromodel-based approach. The development of the simulation model is guided by the realization that the characterization of nonlinear behavior associated with the transfer of forces through the joint is critical to predict the large deformation response associated with progressive collapse. A simplified simulation model of a beam-column joint is used to represent essential and critical actions in the floor beams and the transfer of these forces through the joint region to the vertical elements. The validity of the macromodel developed is evaluated through comparison of both overall response and element actions with those obtained from high-fidelity finite-element analyses. Two prototype buildings designed for lateral load requirements in a nonseismic and seismic region are considered in progressive collapse studies. Two-dimensional models of the frames are subjected to gravity loads and then one or more first-story columns are removed, and the resulting large displacement inelastic dynamic response of each frame is investigated. It is demonstrated that the proposed approach using a validated macromodel is a viable methodology for progressive collapse analysis. The study also finds that special RC moment frames detailed and designed in zones of high seismicity perform better and are less vulnerable to progressive collapse than RC frame structures designed for low to moderate seismic risk.

A nonlayered approach is proposed for the finite-element analysis of reinforced concrete (RC) slabs to predict their strengths including the case of combined in-plane and transverse loading. The nonlayered form is derived from the layered Mindlin-type shell element, and the material matrix is derived from the assumed simple stress-strain curves of the materials. The proposed nonlayered approach is simpler than the layered approach. However, the numerical examples show that it is capable of estimating the deflection as well as the strength of RC slabs under both in-plane and out-of-plane loads with acceptable accuracy.

This paper presents a rational and efficient analytical approach for predicting the nonlinear in-plane flexural behavior of longitudinally posttensioned hollow block masonry shear walls under the action of reversed cyclic lateral loading. The present approach is based on a modified fiber-element (microelement) model of the longitudinally (vertically) posttensioned slender hollow block concrete masonry shear walls with ungrouted cells (i.e., ungrouted longitudinal posttensioning steel) that employs realistic cyclic constitutive rules for the constituent materials. The approach also accounts for the “rocking” type of motion encountered in such walls when subjected to cyclic load reversals in the postcracking range of masonry. The present microelement modeling approach is also valid for slender hollow block masonry shear walls with unprestressed longitudinal steel reinforcement. An important objective of the study is to rationally model the effect of the absence of bond between the longitudinal reinforcing steel (posttensioned or unprestressed) and surrounding masonry, as well as the effect of posttensioning of longitudinal steel on the nonlinear in-plane force-displacement behavior of slender hollow block masonry shear walls subjected to reversed cyclic lateral loading. The proposed approach is implemented to model the in-plane hysteretic force-displacement response of slender posttensioned/reinforced hollow block concrete masonry shear wall specimens with ungrouted cells that were tested under quasi-static reversed cyclic lateral loading. The theoretical predictions are compared with available experimental results for validating the proposed microelement modeling approach.

The loads on buildings and structures caused by the buffeting action of wind have traditionally been analyzed using the gust loading factor (GLF) approach in most codes and standards. In the GLF approach, the equivalent static wind load (ESWL) for structural design is equal to the mean wind force multiplied by the GLF. However, this method fails to provide meaningful guidance in cases with zero mean response. This paper presents a new description of the ESWLs on long-span roof structures based on the load-response-correlation approach, which eliminates the shortcoming of the GLF. The ESWL for a given peak displacement response is expressed in terms of the mean and dynamic components; and the wind loading inputs are determined based on wind tunnel test results of multiple point pressure measurements on rigid structural models. It is noteworthy that in the proposed approach the total dynamic response is directly calculated by the complete quadratic combination approach, in which there is no need to separate the response into the background and resonant components and meanwhile the contributions of multimode response and modal response correlations are taken into consideration. Moreover, unlike existing approaches, it is not required to calculate the correlation of the load and the response, which is difficult to determine by conventional methods. Finally, an extra-long-span roof structure, which is the world’s longest spatial lattice structure, is considered to illustrate the determination of the ESWLs by the proposed approach and to demonstrate its effectiveness in the design and analysis of long-span roof structures. It is shown through an example that the proposed approach can be used in conjunction with wind tunnel tests in predicting the response components not directly measured from the model tests and providing the design loads accurately.

The effects of ten different opening configurations on the internal pressures in a typical two-story North American house were examined using volume-scaled, wind tunnel experiments. The configurations examined include the effects of (1) dominant opening locations and sizes; (2) wall leakage; (3) compartmentalization of the attic space from the living space; and (4) roof and soffit vents. Helmholtz resonance was observed to significantly amplify the internal pressures for open area ratios greater than 3% and full-scale volumes of approximately $700{\mathrm{m}}^3$ . The ASCE 7-05 was found to significantly underestimate the peak internal pressure coefficients for all configurations with dominant openings, regardless of whether there was significant Helmholtz resonance or not. Peak external roof pressures were observed to be highly correlated in time with the internal pressures. It is shown that sealing the attic space from the main (living) space of the house has a significant benefit in not allowing the internal pressure to act on the underside of roof sheathing. However, for just 0.4% open area between the two spaces, peak pressures of 80% of the peak in the living space are transmitted into the attic.

The NatHaz Aerodynamic Loads Database (NALD) (http://aerodata.ce.nd.edu) introduced in 2000 has served an important first step in establishing an on-line experimental archive of high-frequency base balance (HFBB) data for use in the preliminary design of high-rise buildings subjected to wind loads. As a result, NALD was recently introduced in the Commentary of ASCE 7-05 (C6.5.8) as an alternative means of assessing the dynamic wind load effects on high-rise buildings. This paper presents NALD version 2.0 (v. 2.0), integrating the latest advances in data management and mining for interactive queries of aerodynamic load data and an integrated on-line analysis framework for determining the resulting base moments, displacements, and equivalent static wind loads for survivability and accelerations for serviceability (habitability). The key feature of NALD v. 2.0 is the flexibility its analysis module offers: Users may select not only the data from the on-line NatHaz aerodynamic loads database, but also may input desired power spectral density (PSD) expression or wind tunnel-derived PSD data set obtained from a HFBB experiment for the evaluation of wind load effects on high-rise buildings. Thus, it serves as a stand-alone analysis engine. Examples illustrate the capabilities of NALD v. 2.0 and provide comparisons of response estimates to demonstrate the flexibility of the analysis engine to provide a platform that can be readily expanded and supplemented to yield a comprehensive, simplified, and efficient avenue for e-analysis of high-rise buildings.

The role of soil–structure interaction (SSI) on the response of seismically isolated bridges is studied. A generic bilinear hysteretic model is utilized to model the isolation system. The behavior of the pier is assumed to be linear and the foundation system is modeled with frequency-dependent springs and dashpots. Two bridge systems were considered, one representative of short stiff highway overpass systems and another representative of tall flexible multispan highway bridges. Nonlinear time history analyses were employed with two sets of seismic motions; one containing 20 far-field accelerograms and one with 20 near-fault accelerograms. The results from these comprehensive numerical analyses show that soil–structure interaction causes higher isolation system drifts as well as, in many cases, higher pier shears when compared to the fixed-pier bridges (no SSI).

Soil-structure interaction may play a major role in the seismic response of a bridge structure. Specifically, soil layers of low stiffness and strength may result in permanent displacement of the abutments and foundations, thus imposing important kinematic conditions to the bridge structure. A study to illustrate such phenomena is undertaken based on three-dimensional nonlinear dynamic finite-element (FE) modeling and analysis (for a specific bridge configuration under a given seismic excitation). A bridge-foundation-ground model is developed based on the structural configuration and local soil conditions of the Humboldt Bay Middle Channel Bridge. The FE model and nonlinear solution strategy are built in the open-source software platform OpenSees of the Pacific Earthquake Engineering Research Center. Based on the simulation results, the overall system seismic response behavior is examined, as well as local deformations/stresses at selected critical locations. It is shown that permanent ground deformation may induce settlement and longitudinal/transversal displacements of the abutments and deep foundations. The relatively massive approach ramps may also contribute to this simulated damage condition, which imposes large stresses on the bridge foundations, supporting piers, and superstructure.

In this paper, an analytical pulse model for velocity pulses observed in near-field ground motions has been proposed for a systematic design and assessment of seismic protective systems. The proposed pulse model utilizes pulse period, decay factor, and shape parameters to model both buildup and decaying phases observed in recorded ground motions. It is demonstrated that the proposed model can successfully simulate observed ground motion pulses. Peak displacement, velocity, and acceleration of the analytical pulse model are directly correlated with peak ground motion parameters and the pulse period. The response spectra and displacement reduction factor using the pulse model correlate well with those using recorded ground motions for structural periods close to or longer than the pulse period, although there is lesser correlation for shorter structural periods because of exclusion of high frequency components in the pulse model. For a base-isolated building with $2.5\mathrm{s}$ natural period and equipped with supplemental viscous dampers, it is demonstrated that supplemental dampers are the most effective when the structural period is close to the pulse period. Displacement reduction using supplemental passive dampers is smaller when the pulse period is longer or shorter than the structural period, although the displacement demand is quite significant.

Reinforced concrete columns with light transverse reinforcement may be susceptible to shear and subsequent axial-load failures during strong earthquake ground shaking. An experimental program examined the behavior of two half-scale, one-story frames with axial loads representative of those expected for the lower story of a multistory building. The frames were subjected to unidirectional simulated earthquake motion applied at the base. Shear failures of an interior column led to axial-load failure and redistribution of internal forces to adjacent framing components. Recorded data illustrated the details of the shear and axial failures and the redistribution to adjacent components. This paper discusses the global response of the frames and the shear and axial-load response of the shear-critical interior column.

Although approaches implementing strut-tie models are valuable tools for designing disturbed regions of structural concrete, the provisions of current design codes have to be improved because of uncertainties in terms of the selection of the strut-tie model and the structural type of strut-tie model. To improve the uncertainties, a grid strut-tie model approach is presented in this study. The presented approach, which allows for a consistent and effective design of structural concrete, employs a single type of grid strut-tie model whereby various load combinations can be considered. In addition, the approach performs an automatic selection of an optimal strut-tie model by evaluating the capacities of struts and ties using a simple optimization algorithm. The validity and effectiveness of the presented approach is verified by conducting the analyses of the eight reinforced concrete deep beams tested to failure and the design of the shear wall with two openings.

This paper presents a simple yet efficient physical theory model that can be used to simulate the inelastic cyclic axial force–axial deformation and axial force–transverse deformation relationships of steel braces. The model consists of a brace idealized as a pin ended member with a plastic hinge located at its midlength. Input parameters of the model are based only on the properties of the brace. The model combines analytical formulations based on the nonlinear behavior of the brace with some semiempirical normalized formulas developed on the basis of a study of available experimental data. The model realistically accounts for growth effect and degradation of buckling capacity due to Baushinger effects and residual kink present within the brace and it is broadly applicable to steel braces with various section types and slenderness ratios. It is observed that the analytically obtained axial force versus axial displacement as well as axial force versus transverse displacement hysteresis loops compare reasonably well with the experimental ones.

Web holes are often made in wood I-joists to allow passage of services. Design properties of wood I-joists with a web hole are commonly developed through testing which leads to empirical design information. This paper describes a mechanics-based method to predict the strengths of wood I-joists with a circular web hole that is suitable for implementation as a design tool. Curved beam theory is used to compute stresses around a hole. Peak tensile stresses obtained from curved beam analysis agree well with those predicted using the finite-element method. Failure load is attained by the application of fracture mechanics based finite area method. The method developed is used to predict failure loads of wood I-joists of different depths with different web hole sizes. Simplification of the method leads to the development of a simple expression that can be used for engineering design. It is found that the predicted strengths are in a good agreement with the corresponding test results. The calculation method developed can potentially replace the testing approach generally adopted by the wood I-joist industry.

Methods that identify structural component stiffness degradation by pre- and postevent low amplitude vibration measurements, based on a linear time-invariant (LTI) system model, are conceptually justified by examining the hysteresis loops the structural components experience in such vibrations. Two large-scale shake table experiments, one on a two-column reinforced concrete (RC) bridge bent specimen, and the other on a two-span three-bent RC bridge specimen were performed, in which specimens were subjected to earthquake ground motions with increasing amplitude and progressively damaged. In each of the damaged stages between two strong motions, low amplitude vibrations of the specimens were aroused, and the postevent component stiffness coefficients were identified by optimizing the parameters in a LTI model. The stiffness degradation identified is consistent with the experimental hysteresis, and could be quantitatively related to the capacity residual of the components.

Various semiactive systems have been proposed to control the seismic response of civil structures. These include semiactive fluid, friction, and stiffness devices. Energy dissipation in these devices occurs through a variety of mechanisms. The purpose of the present study is to propose a new semiactive stiffness system that dissipates energy based on the principle of hysteresis. The system consists of a novel variable amplification device (VAD) connected to a simple spring. When integrated with other semiactive components, the stiffness of the VAD-spring system can be adjusted based on feedback from the structure’s response. This is done by changing the amplification factor of the VAD according to a semiactive control algorithm tailored specifically to the device’s unique characteristics. The proposed system is simulated for an eight-story building subject to multiple seismic excitations. The response of the building using the VAD-spring system is compared with the maximum amplification case and that obtained using the active variable stiffness system. Results indicate that the VAD-spring system is an effective means of controlling vibrations in seismically excited buildings.

Braces are commonly used to provide lateral stiffness and strength to steel framed buildings subjected to wind or earthquake loading. Of the structural sections that can be employed as brace members, hollow structural sections represent a very good solution due to their excellent structural properties in compression. Nevertheless, their use has been hitherto compromised for seismic applications mainly due to a lack of simplified connection details that avoid brittle failures. At present, slotted hollow structural section connections are popular in seismic zones but the hollow section is typically reinforced with steel cover plates to increase the net area at the critical location to avoid premature fracture under tension loading cycles. This reinforcing practice can likely be avoided, for circular hollow sections, if both the tube and the plate are slotted in such a manner that the weld terminates at the tube gross cross section. The potential of this innovative detail is demonstrated by means of three large-scale specimen tests under pseudodynamic loading.