Journal of Structural Engineering

Based on tests and nonlinear finite element analyses, a unified plastic analysis for infilled frames with three types of interface conditions is proposed, in which the stress redistribution towards collapse and the conditions at the infill‐frame interface are taken into account. The theory is applicable to single and multistory infilled frames, and comparison with experimental results gives good agreement. Design recommendations for multistory infilled frames are given.

Experimental study is carried out to verify the grillage idealization for dynamic analysis of multi‐cellular structures. Perspex material is used to construct two models of same size, one with no web openings and the other with 25% web openings. The natural frequencies and the corresponding mode shapes are determined for two different sets of boundary conditions, namely all four sides simply supported, two opposite sides simply supported, and the remaining sides free. The experimental results are found to compare well with the theoretical results using grillage idealization and finite element methods. As the linear response due to any dynamic load is the superposition of the modal components, it is concluded that the grillage idealization could be used to predict the forced vibration response of multi‐cellular structures. Since the study is restricted to linear elastic range, the findings of this investigation are applicable to structures made of other material, e.g., steel cellular structures.

An alternative approach for estimating the strength degradation, and ultimately, the failure of glass cladding subjected to fluctuating loadings is proposed. This procedure is formulated using fracture mechanics concepts employed previously. This approach is simple computationally and does not require time integration of fluctuating stress or pressure loading time histories.

A study of nearly 150 recent major collapses and distresses of structures around the world discloses that external events and deficiencies in the areas of construction and design were the principal sources of failures. More than one‐third of the surveyed structures were bridges, and the remaining were low‐rise, multi‐story, plant‐industrial, and long‐span buildings. The major incidents in the category of external events were the lateral impact forces and other unexpected live loads on bridges, and explosion impact loads in concrete silos. Construction deficiencies included falsework and concreting faults in several concrete structures and inadequate welding operations in steel bridges. Design deficiency related events were found to vary with the type of the observed structures, but a significant number of failures were attributed to lack of knowledge in long‐term creep and shrinkage effects on prestressed concrete members. These findings suggest that attention should be directed to three areas: (1) Identification of potential deficiencies from past failure data; (2) enhancement in procedural methods during design and construction operations; and (3) incorporation of risk analyses of structures during their service life and construction phases.

The pseudodynamic method is a relatively new experimental technique for evaluating the seismic performance of structural models in a laboratory by means of on‐line computer controlled testing. During such a test, the displacement response of a structure to a specified dynamic excitation is numerically computed and quasi‐statically imposed on the structure, based on analytically prescribed inertia and viscous damping characteristics for the structure and the experimentally measured structural restoring forces. This paper presents the basic approach of the method, describing the numerical and experimental techniques. Based on current studies, the capabilities and limitations of the method are examined, and possible improvement methods are mentioned. In spite of certain numerical and experimental errors, recent verification tests show that the method can be as reliable and realistic as shaking table testing and that it can be readily implemented in many structural laboratories. The capabilities of the method can be further expanded to test specimens under various load and structural boundary conditions.

Tests to failure of two small‐scale model stub girders are described. The tests were performed to determine whether or not behavior observed in an isolated test specimen is indicative of behavior of a similar girder incorporated into a complete floor system. Results indicate a significant difference in both the longitudinal shear strength and mode of failure of the slab in an isolated specimen as compared to a similar girder within a complete floor system.

An “exact” solution is presented for the analysis of orthotropic curved bridge decks subjected to patch, uniform, line and concentrated loads. The bridge deck is idealized as an assemblage of radially supported curved plate elements or strips. The deflection of each plate strip is expressed as a Levy type Fourier series and the loads are expressed as a corresponding series. Concentrated loads and line loads in the tangential direction are incorporated in the solution as a discontinuity in the shear force along the load line. The convergence of this procedure is achieved by increasing the number of modes rather than the number of elements. The solution has the advantage over purely numerical or seminumerical techniques, using grid analogy or finite element techniques, in that the computational difficulties do not increase as the number of terms is increased to give increased accuracy. Tabular and graphical results are presented for bridge decks subjected to various loading conditions.

A new finite element for bond stress‐slip analysis is presented. A one dimensional model which is based on equilibrium and local bond stressslip law is developed. The relationship between axial force and slip at the elements nodes is expressed through a stiffness matrix. The global stiffness matrix is assembled and solution yields slip, strain and stress distributions along the steel bar. The nonlinear bond stress‐slip relationship leads to an iterative technique which is found to converge rapidly. The proposed method predictions are compared with experimental results of monotonic and push‐pull tests and very good correspondence is found.

A gable frame design is reviewed with regard to the modeling, load specification, and strength evaluation. It is found that the approximation used for the boundary conditions at the column bases should adequately account for the behavior of the foundation and more than one approximation may be necessary. Uneven loading is shown to be very significant, and it is recommended that gable frames always be designed for some uneven distribution of load. The question of the correct effective length for the rafter is examined, and a nonlinear analysis or an approximate, but conservative, approach is suggested for gables with significant axial load in the rafter. One method of utilizing a frame buckling analysis in a conventional AISC evaluation is presented. It is suggested that if there is significant axial stress in the rafter, caution should be exercised in the use of specification limitations, which are based on girder behavior where there may have been no axial stress considered in the derivation and testing of the specification limitations.

Extensive computer analysis of an existing 44 story steel frame high‐rise building is performed to study the influence of various modeling aspects on the predicted dynamic properties and computed seismic response behaviors. The predicted dynamic properties are compared to the building's true properties as previously determined from experimental testing. The seismic response behaviors are computed using the response spectrum (Newmark and ATC spectra) and equivalent static load methods (ATC and UBC). Interpretations of the analysis results are provided. Conclusions are drawn regarding general results that are relevant to the analysis of other high‐rise buildings.

A method is developed for the optimum design of trusses including constraints on tension and compression stresses, displacements and bounds on the design variables. It is intended to solve a practical problem. The method takes into account the efficiencies achieved for simpler mathematical models where the allowable values for compression stresses are constant. In this work, these values depend on the member cross section and are prescribed in the Code AISC‐78. All the peculiarities introduced by the buckling constraints are identified and solved. They are: (1) Consideration of the radius of gyration, in addition to the cross‐sectional area, as design variable; (2) introduction of more complicated constraint functions; and (3) computation of a scale factor to the constraint surface for buckling restrictions. The problem is solved by means of a sequence of approximate problems in the dual space. The approximation is basically due to the fact that the constraint functions are replaced by their first order Taylor series expansions to improve the efficiency of the solution. The effectiveness of the method is shown through some examples.

Three simple analytical methods to study the impact effect on R.C. slabs are presented. These methods are the impact factor method, the equivalent mass method, and the continuous mass method. Several aspects that affect the slab's response due to impact are investigated. These aspects include the relation between the impact load and its actual static load when acting on the structure, the duration of impact load when acting on the structure, and the dynamic deflections of the reinforced concrete slabs due to impact. The impact is simulated by dropping a steel ball from different heights on the center of simply supported reinforced concrete slabs. Dynamic deflection, impact time, maximum penetration of the ball into the slab, and the maximum dynamic load that the slab will experience are predicted using the simple analytical methods. Results from these methods are compared with similar experimental ones. In general, the analytical results agreed with the experimental ones.

The fatigue behavior of 12 different sets of partially prestressed concrete beams was experimentally investigated. All beams were rectangular, $9\times 4.5$ in. $(229\times 114\mathrm{mm})$ in cross section, simply supported on a 9 ft (2.74 m) span and loaded in 4 point bending. Each set consisted of 2 identical specimens designed with the same input parameters. One control beam was tested under static load up to ultimate. The second beam was tested in cyclic fatigue at a constant load range varying between 40% and 60% of the ultimate load capacity of the static specimen. The main input variables were the partial prestressing ratio PPR and the reinforcing index $\overline{\omega }$. Four different levels of PPR and three different levels of $\overline{\omega }$ covering both fully prestressed and fully reinforced were explored. Typical results and observed trends are described. Throughout the tests, measurements of strains in the reinforcement, deflections, crack widths, curvatures and their variation under static and cyclic fatigue loading were systematically recorded. The six partially prestressed and the three fully reinforced beams survived 5,000,000 cycles without suffering fatigue failure. The three fully prestressed beams which were loaded beyond cracking failed, respectively, at $1.21\times {10}^6\text{,}$ $2.17\times {10}^6$ and $1.94\times {10}^6$ cycles.

By considering the physical symmetry of a structure, a number of pairs of the elements in the upper (or lower) triangular part of the stiffness matrix are equal in absolute value. Such relationships are described in terms of degrees of freedom in ``intrinsic coordinates.'' For similar nodes on each side of a plane of symmetry, the intrinsic coordinates of the nodes are mirror images of each other. The stiffness coefficients for corresponding degrees of freedom in the intrinsic coordinates for similar nodes are equal. The number of independent and dependent coefficients are delineated. For one fold symmetry, the number of independent stiffness coefficients is (n+2)n/4, and for two and three fold symmetry, it is (n+4)n/8 and (n+8)n/16, respectively. The relation of the signs of the dependent and independent coefficients is developed. The information may be used to save computation efforts needed for the formulation of stiffness matrices of symmetric structures.