Journal of Structural Engineering

A method for the discrete minimum weight design of welded steel columns is developed. The design problem is formulated as a discrete nonlinear programming problem. The problem is solved using an interior penalty function method combined with a variable metric discrete search technique. Starting with an initial feasible solution, the first unconstrained problem is solved by searching along the direction of the integer gradient of the modified objective function. For the next unconstrained problems the direction of search is modified by the inverse of the Hessian matrix of the objective function. This way the slow convergence of the gradient method near the optimum is avoided. An example is given to show the efficiency and rapid convergence of the method. The results obtained show that more efficient use of materials can be achieved by using welded steel sections. The method presented can also be used to find the minimum cost design of welded steel sections if the cost factors are known.

The response of a typical steel‐lined reinforced concrete nuclear reactor containment to postulated internal hydrogen detonations is investigated by axisymmetric nonlinear dynamic finite element analyses. Internal wall pressure histories used as input to the analysis are first generated by numerical solution of the hydrogen detonation problem with a technique that reproduces the sharp discontinuity at the shock front. In place variability of the mechanical properties of reinforcing bars and of the strengths of mechanical splices is included in the containment model through effective stress‐strain laws of the elements modeling the reinforcement. Dynamic analyses performed for atmospheric initial pressure show that the response is more sensitive to the point of initiation than to the strength of the detonation. Due to the vulnerability of the containment to postulated detonations, it is important to prevent formation of a detonable mixture in the containment atmosphere.

The geometric nonlinearities in elastic sway frames are examined. Various approximate methods of second‐order analysis for sway frames are reviewed and the conditions limiting the use of these procedures are stated. A rational method for combining the nonsway and sway moments is developed. A practical method of including the effects of sway deflections due to gravity load moments and out‐of‐plumb construction in the approximate second‐order analysis is derived.

Tremendous progress has been made in the art and construction of long span bridges, of which history goes back almost to the third millennium B.C. In this evolution structural concepts and strengths of materials used have played key roles. The limiting span is defined as that span at which the structural system will just support itself. The ultimate span is the maximum span possible for a particular type of construction. Comparison is made between ultimate spans and those already built to indicate that in concrete construction the limit has almost been reached but for steel construction much larger spans are possible.

Measurements were made of the response to wind action of the 265‐m high, reinforced concrete stack at Mount Isa Mines in Queensland, Australia. Data were reduced from accelerometer, strain gage and pressure transducer records from four wind storms. Particular attention was paid to determining reference freestream wind speeds at stack height and wind structure, which involved using records from a number of anemometer sites and for which position errors were determined from topographical model studies in a boundary layer wind tunnel. In this paper, the full scale data are reduced and presented such that they can readily be used to compare with results from physical and mathematical model studies of stack response to wind action. Mean, standard deviation and peak tip displacements, and base overturning moment coefficients are given, along with estimates of structural damping, peak factors and dominant wake frequencies, at Reynolds numbers up to $2\times {10}^7\text{.}$

The guyed tower concept is designed for use in relatively deep waters [more than 1,000 ft (300 m)]. A model is described that mimics the response by a single degree of freedom (SDOF). The properties of this simple model are derived from stiffness, mass, and hydrodynamic properties of more detailed models. The ability of the simple model to represent the tower response is evaluated by comparing calculated SDOF response with that of multi‐degree of freedom (MDOF) finite element models. Calculated responses from detailed finite element mooring system analyses (of an individual line) are used to establish dynamic and static characteristics of an individual mooring line. Each mooring line is represented by an equivalent spring in the simple mooring. Overall dynamic tower responses using statically and dynamically derived mooring system properties are compared.

An evaluation of publications on reduction of structural eigenvalue problems has revealed the following: (1) Most of the published techniques are essentially equivalent to Guyan reduction; (2) published techniques to decrease the error in Guyan reduction require matrix inversions or an iterative process, or both. This paper presents a modified Guyan reduction method which decreases errors in the calculation of the eigenvectors of the system and which does not require matrix inversions whatsoever. It also shows that a simple iterative process could be used to improve further the eigenvalue solution of the system.

The raising of a continuous span reinforced concrete (RC) bridge creates the potential for subjecting the bridge to loading deformations for which it was not designed. This, coupled with the inherent weakness of concrete in tension and the associated critical positioning of the rebars, makes the possibility of rupture or excessive cracking during raising very real. Experimental testing was conducted on a one‐lane test bridge to evaluate force‐deformation responses for both abutment and bent jacking. The bridge is a continuous threespan, haunched girder type concrete bridge of standard design. Bridge support reactions and displacements were the parameters measured in the experimental testing. Results of the experimental testing were used to test several popular analytical analysis response prediction models. Giving due consideration to accuracy, simplicity, and convenience, it is recommended that a continuous prismatic member model be used in predicting bridge responses to both abutment and bent jacking. Recommended bridge modefing and idealizations are also presented.

A method of analysis is developed for determining the response of precast concrete floor decks to point loads. Torsion is accounted for, and the method can be applied to a variety of cross sections, including double tees and hollow‐core units. A computer program is described which generates influence surfaces for internal member forces.

Tests have recently been conducted at the University of New South Wales on half‐scale models of edge panels of a reinforced concrete flat plate floor. Moments were measured along lines parallel to the edge of the building: (1) Just inside the spandrel; (2) at mid‐span; and (3) at the first interior column. The measurements provided the distribution of slab moments across a panel and, thus, by integration, the total panel moment at each of these locations.

The current ACI Code prescribes two alternative methods for the determination of slab moments, i.e., the Equivalent Frame Method and the Direct Design Method. The test results indicate that for calculating the full panel moment, the first of these is unnecessarily complex and bears no relation to reality.

Modifications to the Code methods are suggested which are not only simpler but consistent with observed behavior. Tests also indicate the degree of loss of stiffness which might be expected after cracking.

A parallel structural system is one in which more than one critical section must reach a limiting state before the structure as a whole can reach a limit state. An obvious example is the plastic analysis of redundant frameworks in which individual member strengths can be maintained while the structure deforms to mobilize the strength of all those members which can add to its ultimate load capacity. In a reliability analysis, the effect of member strength correlation may be an important parameter governing the reliability of the whole system. This aspect is investigated in this paper. It is shown by means of a series of examples of relatively simple frames that the reliability of the system is sensitive mainly to correlation between the member strengths within the one member or between related members. However this sensitivity varies with the dominant collapse mode for the frame.

An accurate and efficient analytical model describing the hysteretic behavior of R/C beam‐column joints subjected to random cyclic excitations is presented. The model takes into account the cyclic bond deterioration along anchored bars as well as the hysteretic behavior of cracked R/C sections at beam‐column interfaces, thus allowing consideration of the interaction between the two end sections of the joint.

A general instantaneous or tangential stiffness matrix for a plane frame member is presented. Expressions in the form of integral equations are developed for members having axial forces and arbitrary material properties, distributed loads and moment‐axial force‐curvature relations. The expressions are used for w sections of elasto‐plastic material and several examples of plane frames are loaded to failure. Experimental results reported in the literature for plane frames tested to inelastic failure are closely matched by the analytical technique.

The response of interior reinforced concrete plate column connections in a laterally loaded structure is investigated. Nine specimens were tested, five statically and four dynamically to simulate earthquake loading. This paper describes results from statically tested specimens. Experimental variables include reinforcement ratio and amount of superimposed slab load. Nominal slab dimensions were $6\times 6\mathrm{ft}(1.8\times 1.8m)\text{.}$ Slab thickness was 6 in. (76 mm). The column was square with plan dimensions of 1.0 ft (0.3 m) by 1.0 ft (0.3 m). Reinforcement layout in the slab was isotropic. Statically tested specimens provided data on influence of reinforcement amount and slab load oh stiffness and strength. In the range of joint rotations less than 4%, strength was controlled by torsional stiffness of the slab. Slab shear failures were observed for the specimens with slab load at joint rotations over 4%. A numerical model was developed to enable calculation of the overall moment‐rotation response.

A method of two dimensional analysis of sandwich panel assemblies used as main resisting elements of building structures is provided. This method is based on the equivalent wide column frame analogy, with due consideration given to the composite nature of materials and to the nonlinear behavior of the connection system. A simplified structural idealization is adopted to predict the theoretical response of a six story two cell box structure, subject to a series of loading conditions. Theoretical and experimental results compare favorably. The influence of secondary effect of shear deformation, and the effective width of floor beams on the structural response is also taken into account.

The effect of external hydrostatic pressure on the strength and behavior of fabricated steel tubular members is studied. The axial force is considered to be produced only by the applied external force but not by the pressure as is the case for the chords of offshore fixed platform. Column strength curves for columns loaded axially at the ends and column interaction curves for columns loaded eccentrically at the ends are obtained considering the influence of hydrostatic pressure. It is found that external pressure reduces the load carrying capacity of tubular columns significantly when the out‐of‐straightness and pressure are large. Also, sectional capacity is reduced significantly by the presence of external pressure when axial load is small. Design formulas are derived for axially loaded as well as eccentrically loaded columns by a simple modification of the present American Institute of Steel Construction (AISC) design formulas.