Journal of Structural Engineering

An analytical study was conducted on the cyclic loading behavior of beams in steel moment frames. Lateral–torsional buckling and local buckling were explicitly considered in the analysis. Beam cross-sectional dimensions were varied to represent a wide range of rolled wide flange shapes. The unbraced length of the beams corresponded to slenderness ratios about the weak axis ranging between 60 and 100. Based on the analyses, flange and web width-thickness limits were established that would permit the beam to achieve various target rotation capacities. These limits are presented in terms of limit curves which plot the flange width-thickness ratio against the web width-thickness ratio. The limit curves clearly show strong flange–web interaction. Different limit curves were developed for a range of target rotation angles, weak axis slenderness ratios, and residual strength levels. Postbuckling behavior and strength degradation mechanism were studied. The results of the analyses are compared against current building code requirements for beam stability in seismic steel moment frames, and the adequacy of the current code requirements are evaluated.

Results from the tests of six cantilever double-channel built-up members under reversed cyclic bending are presented. The main parameters investigated were stitch spacing and lateral bracing requirements in order to sustain large inelastic rotations, such as those induced by severe earthquake events. Experimental results indicate that current lateral bracing requirements specified in the AISC-LRFD provisions are not adequate to prevent lateral–torsional buckling of individual elements, as well as of the entire built-up member. Based on the results from this research, a new equation is proposed for stitch spacing for interconnection of individual channels. In addition, lateral bracing for the overall member is required near the plastic hinge region to ensure total plastic hinge rotation capacities on the order of $0.06\mathrm{rad}$ .

To compare the hysteretic behavior, stiffness, and strength of stiffened and unstiffened extended end-plate connections of beam–column joints, six specimens with eight high strength bolts were tested under cyclic load. During the test, the axial force of the column was kept constant. The end plates of three specimens and column webs of five specimens were reinforced with rib stiffeners, and the thickness of end plates was varied with the same configuration of beams and columns of all the specimens. Although the plastic rotation of all connections exceeded $0.03\mathrm{rad}$ , they showed different hysteretic behavior, the stiffened ones exhibited higher load carrying and energy dissipation capacities, and the end-plates’ stiffness had a crucial influence on their cyclic behaviors. For thinner end plates, stiffeners can increase the stiffness of the connection greatly, and delay the premature fracture of welds. Stiffeners of the column web can change the stress dissipation panel zone and reduce the possibility of strength failure of column flange.

The paper shows the shear-lag effect in slabs of twin-girder steel–concrete composite decks due to the main prestressing techniques such as support settlements, bonded cables inside the concrete slab, and external slipping tendons. The analysis is performed by means of a beam model taking into account the loss of planarity for the slab cross section, the flexibility of the shear connection, and the time-dependent behavior of the concrete. A linear elastic behavior is considered for the steel beams and shear connection while viscoelastic behavior is assumed for the concrete slab. A parametrical analysis is carried out on realistic bridge decks for each kind of prestressing technique, by varying the span length and the girder spacing. The distribution of longitudinal normal stresses in the slab is shown for each prestressing technique and the effective slab widths are then evaluated and compared with those suggested by ENV 1994-2 for different span lengths and for different spacing of the beams.

As a result of stress concentration, welding transverse stiffeners in flat-web plate girders reduces their fatigue life. The focus of this study is to investigate the behavior of trapezoidally corrugated-web plate girders (CWPG) under fatigue loading by means of evaluating their fatigue life under different stress ranges. The test girders failed by fatigue cracks, which started at the weld between the web and the tension flange near the web fold line. A relationship between the stress range “ $F_r$ ” and the number of cycles to failure “ $N$ ” was developed for CWPG. Fatigue lives of the tested CWPG were compared to other girder types and the test results were related to the current AASHTO specifications.

This paper presents finite-element and fracture mechanics studies of plate girders with corrugated webs. The failure mode of this type of girder under fatigue loading is distinguished by its location in the tension flange and is discussed in detail. A comprehensive nonlinear finite-element analysis was conducted to study the effects of the different geometrical parameters of this type of girder on the stress concentration at different locations, and to explain findings of an earlier experimental study. Different parameters, such as the radius of curvature between the web folds, the inclined fold angle of inclination, and the use of sinusoidal corrugation were investigated. Two sources of stress concentration leading to fatigue cracking were isolated; one is dependent on the radius of curvature between the web folds and the other depends on the inclination angle of the inclined fold. An analytical technique based on fracture mechanics is proposed to determine the fatigue life of the girders. The technique enables designers to evaluate the fatigue life of structures built using corrugated web steel girders and subjected to fatigue-type loading, such as bridges or industrial facilities subjected to dynamic loading.

A strain rate dependent material model for concrete has been proposed for analysis of three-dimensional reinforced concrete structures under transient dynamic loading. The model takes into account the pressure sensitivity in three-dimensional loading situations and strain rate sensitivity in dynamic loading situations. The cracking in tension has been modeled using the smeared cracking approach. The strain rate sensitivity has been considered at two levels: the failure surface is made strain rate dependent as is also the fluidity parameter. The tension stiffening effect observed in reinforced concrete structures has also been considered.

This paper presents a plain strain analytical model, based on the elasticity theory, to determine the confining pressures of transverse reinforcements on the concrete core of a reinforced concrete member. The analytical evaluation of the confining pressures was first carried out on reinforced sections with square and circular stirrups, and subsequently on reinforcement configurations of greater complexity with square and rectangular stirrups and supplementary cross ties. Finally, the model has been used to evaluate the confining pressures applied by external wrapping in any material [fiber-reinforced polymer (FRP), S-glass, steel, etc.] and to design better combinations of techniques and confinement materials. In order to obtain the stress–strain curves due to passive confinement, an analogy between square and circular sections has been introduced. In this way, any active confinement model derived by triaxial tests on cylindrical specimens can be used. The model has been validated by comparing its predictions with results from existing models and experimental tests.

This paper introduces a new concept in shear wall design. One ply of sheathing material is placed at the center of the wall between a series of pairs of studs and plates oriented in $90°$ rotated position relative to those in standard shear walls. The new wall design is called “midply” in reference to the location of the sheathing panel. Test results for midply walls under static and cyclic loading conditions are presented and compared with those presented for standard shear walls under similar loading conditions. In midply walls, nails connecting framing members to sheathing work in double shear in contrast to connections in standard shear walls, which work in single shear. This results in substantial improvements in the shear performance of midply walls. As verified by the test results presented herein, load-carrying capacity of midply walls is over three times that of comparable standard shear walls.

Twenty one $2.4\mathrm{m}$ by $2.4\mathrm{m}$ ( $8\mathrm{ft}$ by $8\mathrm{ft}$ ) wood framed shear walls were tested to study the effect of vertical load and hold-down anchors on the cyclic behavior of the wall. Seven different configurations were investigated including walls without hold-down anchors and uniform vertical loads of 0, 6, 12, and $25\mathrm{kN}∕\mathrm{m}$ (0, 425, 850, and $\mathrm{1,700}\mathrm{lb}∕\mathrm{ft}$ ) and walls with hold-down anchors and uniform vertical loads of 0, 12, and $25\mathrm{kN}∕\mathrm{m}$ (0, 850, and $\mathrm{1,700}\mathrm{lb}∕\mathrm{ft}$ ). The walls were loaded cyclically according to the Consortium of Universities for Research in Earthquake Engineering test protocol. Results show that the effects of vertical load on the cyclic response of wood frame shear walls are to: (1) increase the lateral stiffness and (2) increase the energy dissipation capacity. The lateral stiffness increased by up to 80% for walls with vertical loads of $25\mathrm{kN}∕\mathrm{m}$ $(\mathrm{1,700}\mathrm{lb}∕\mathrm{ft})$ , over the full range of displacement amplitudes; energy dissipation increased by up to 100% for walls with vertical loads of $25\mathrm{kN}∕\mathrm{m}$ $(\mathrm{1,700}\mathrm{lb}∕\mathrm{ft})$ , over the full range of displacement amplitudes. Results also show that hold-down anchors have minimal effect on the stiffness and energy dissipation capacity of shear walls when loaded to the design allowable load of $25\mathrm{kN}∕\mathrm{m}$ $(\mathrm{1,700}\mathrm{lb}∕\mathrm{ft})$ .

The potential applications of tensegrity structures are not only increasing in civil engineering but also in fields like biomechanics. The key step in designing tensegrity, the form-finding problem, has been investigated by many researchers but until now they have tended to focus on methods for regular shapes. Since there is an increasing need for design tools devoted to more various and complex systems, the objective of this paper is to present the form-finding of nonregular tensegrity structures with a numerical approach. It is based on the dynamic relaxation method with kinetic damping, and new tensegrity configurations in more intricate and creative forms can be obtained this way. During the form-finding process, either the force or length of some elements can be fixed by an appropriate choice of related stiffnesses. The application of the process is illustrated by several numerical examples. It can be concluded that an improvement in tensegrity form-finding has been achieved extending research from regular shapes toward “freer” shapes.

A new method via inverse problem formulation including frame stability ( $P\text{-}\delta$ and $P\text{-}\Delta$ effects) is presented for state-specified design of an arbitrary planar elastic frame. The well-known governing equations for second-order analysis with respect to unknown displacements for a set of prescribed member stiffnesses of a frame are here regarded as those with respect to unknown member stiffnesses for a set of prescribed member-end strains and displacements. A method of sequential mapping for such a “direct design”problem is generalized for a hybrid inverse problem of an arbitrary plane frame such that a combined set of unknown member stiffnesses and unknown member-end strains (or displacements) is to be found for a set of prescribed member-end strains (or displacements) and prescribed member stiffnesses. The concept of “state-design vector” is shown to be useful for such a general problem of hybrid-inverse mapping. A general design program has been developed on the basis of the proposed algorithm on the first-order equations. Four design examples not only demonstrate the program performance but also disclose $P\text{-}\delta$ and $P\text{-}\Delta$ effects on the design. A guideline is described for practical application in compliance with an allowable stress design code.

Recent studies have successfully introduced genetic algorithms (GA) to identify the important failure modes of complex structures and quantify their contributions to the reduction of the reliability of structural systems. In these studies, the efficiency of traditional GA techniques was substantially improved by incorporating linkage-learning operators that were used to explore relations among the random variables controlling the safety of a structural system. However, the currently used linkage learning methods were found to be either too complex for easy implementation in routine reliability analyses or too narrowly focused on finding the global optimal solution and thus reduced the capacity of GA to identify local optima. Other techniques especially suitable for exploring relations and linkages among data sets are known as data mining (DM) algorithms. One of the most popular and successful DM tools available in the computer science literature is the a priori algorithm. The purpose of this paper is to propose a hybrid algorithm that would combine the benefits of the pattern identification ability of the a priori DM algorithm to the capability of GA operators to explore new significant search domains. The implementation of the shredding operator and its ability to reduce the computational effort through its self-learning process will further lead to the development of an efficient and robust structural reliability analysis algorithm. This paper will demonstrate that the proposed algorithm will significantly reduce the computational effort associated with determining the probabilistically dominant failure modes of structural systems. Examples are provided to verify the high efficiency and accuracy of the proposed hybrid data mining/genetic shredding algorithm for failure mode exploration, identification, and exploitation.

Current seismic design procedures, which permit the estimation of inelastic deformation capacity of lateral force resisting systems, have been questioned since no rationality exists for determining the values of $R$ tabulated in seismic design codes. This study focuses on the seismic behavior of three-, nine-, and 20-story steel moment resisting frame (MRF) structures designed using different $R$ factors (8, 9, 10, 11, and 12). A total of 30 different structures were subjected to 20 ground motions representing the hazard level equal to a 2% probability of being exceeded in 50 years (2% in 50 years) for a detailed case study. Then, a reliability-based performance evaluation procedure, which provides a degree of confidence for satisfying the collapse prevention (CP) performance objective, was applied to model buildings to assess the effect of the $R$ factor variations. The results show that the current $R$ factors provide conservative designs for the three- and nine-story buildings for achieving the CP performance objective. However, the 20-story buildings designed with different $R$ factors without the minimum spectral acceleration bound on $C_S$ prescribed in the 2000 International Building Code show a low level of confidence for achieving CP performance.

The viability of effective force testing (EFT), a real-time dynamic testing procedure for large-scale structures, has been validated by comparing results of tests on idealized structural models to analytical results. In this paper, the results of tests on a shake table are compared to those conducted using the EFT method. The test structure was a one-story steel structure consisting of typical structural components. The paper also presents an overview of critical issues required for conducting the EFT test, and highlights several problems and solutions that may affect the performance of EFT tests. The comparison of the test results indicates that with proper velocity feedback compensation, the EFT method can be used to apply real-time seismic simulation to structures undergoing nonlinear deformation.

Motivated by the observation that existing damage indices do not take into account the bidirectional and torsional response effects, a new three dimensional damage index is proposed in this paper. The derivation is based on decomposing the 3D structure into planar frames. A local damage index that is sensitive to out-of-plane responses is described and a method to combine local damage indices is presented. The proposed index is verified by comparison with conventional damage indices.

This paper presents a nondestructive global damage detection and assessment methodology based on the changes in frequencies and mode shapes of vibration of a structural system. The method is applied at an element level using a finite-element model. According to continuum damage mechanics, damage is represented by a reduction factor of the element bending stiffness. A nonclassical optimization approach involving the use of genetic algorithms (GAs) is proposed to localize damaged areas of the structure. The method has been verified by a number of damage scenarios for simulated beams and by using directly experimental data from the vibration tests of a beam. It is shown that the proposed GA yields a suitable damage location and severity detection while introducing numerous advantages compared to classical methods. The influence of noise in the modal data has also been considered.

The flexural deformation of structural members is due to their curvature. For reinforced concrete (RC) members, this curvature depends on the tensile strain of the reinforcement and the compressive strain of the concrete. As concrete is a brittle material with little ductility, RC members achieve ductility and adequate deformation capacity mainly through the tensile straining or yielding of the reinforcement. When the tensile straining of the reinforcement is limited, such as in the case of over-reinforced RC beams and RC columns with large axial loads, whereby the tensile reinforcement does not yield and the member fails due to concrete crushing, the ductility of the member is limited. The application of high-strength steel or fiber reinforced polymer reinforcement significantly reduces the ductility of the tensile reinforcement, which results in a substantial reduction in the ductility of the structural member. Existing methods for improving the ductility of concrete structures with nonductile reinforcement are not satisfactory. A new scheme of providing ductility to RC members through compressive yielding instead of tensile yielding is developed and introduced in this paper. The effectiveness of the new scheme is illustrated by experimental testing of RC beams. The successful application of this novel technique may pave the way for a new field of research and application of concrete structures.