Journal of Structural Engineering

A typical elevated silo consists of a cylinder and a discharge hopper. The hopper in a silo supports the majority of the material stored within the silo, and is subject to biaxial tension. This paper presents the first known study of the collapse behavior of welded steel silo hoppers under the bulk‐solid pressure distributions given by the theory developed by Walker in 1966. The aims of this paper are to identify the possible failure modes, to examine the effect of different pressure distributions on the collapse strengths of welded hoppers and transition junctions, and to discuss current criteria for the design of welded‐steel hoppers. The two most important messages from this study are: (1) For most practical welded‐steel silo hoppers, the failure strength of the hopper is independent of the pressure distribution in the hopper; and (2) where the actual pressure distribution does matter, the filling pressures are more critical than the flow pressures.

A methodology for complete automation of the conventional highway bridge design for minimum cost is presented. To make such an automated process practical and cost‐effective, efficient interface of three tasks—finite‐element analysis to generate influence surfaces, critical design force computation, and optimization is suggested. Experiments with the program developed based on the suggested methodology indicate that the cost of conventionally designed continuous steel plate girder highway bridges can be reduced by as much as 25%. Approximately 80% of the savings is due to the use of three‐dimensional finite‐element analysis and influence surfaces. About 20% of the savings accrues from the use of optimization. The methodology is general and can be applied to the design of any type of continuous multispan highway bridge.

Ten experiments concerning the carrying behavior of square knee joints are performed. It is shown that, analogous to transversely stiffened girders with thin webs, tension fields that are anchored on flanges also develop in knee joints. The load capacity of the knee joint is limited by the formation of plastic hinges in the flanges. Comparison with the buckling load shows that a stiffened knee joint with a thin web plate exhibits considerable overcritical load reserves. The failure mechanism of stiffened, square knee joints is described using a simple tension field model. This model introduces two load‐carrying parts: one carried by a shear field, the other by a tension field. The calculated load capacities are approximately 10% greater than the experimentally determined capacities. Further tests are proposed to verify the model and to derive a design concept.

The limit of serviceability and the residual safety against structural collapse of a concrete cooling tower, which was built 25 years ago, is evaluated numerically by the finite element method. The present condition of the shell includes a relatively large number of long meridional cracks. The process of damage is simulated by means of applying different thermal load histories. Each of them is followed by the incremental application of the wind load. For purposes of comparison, results from ultimate (wind) load analysis of the originally uncracked shell are presented. From the analyses it is concluded that the degree of corrosion, for which several assumptions are made, controls the safety of the cooling tower. With respect to the sensitivity of the structural response concerning the value of the tensile strength, the effectiveness of a repair by attaching stiffening rings to the shell and the influence of the location of a single ring on the structural response also are investigated.

The nonlinear cyclic behavior of reinforced concrete beam‐column members is complex and governed by a variety of interrelated physical phenomena. The ability of a refined uniaxial finite‐element model to simulate many of the factors influencing the planar behavior of reinforced concrete structures is examined in this paper. Following a brief description of the analytical model and its implementation in a nonlinear dynamic analysis computer program, analytical results are compared with experimental data. Structures considered include individual reinforced concrete columns under static and dynamic loads and a complete frame structure subjected to dynamic loads. Numerical simulations are also presented to demonstrate the ability of the refined model to predict the response of structural systems that experience axial load variations and severe local damage.

The nonlinear behavior of reinforced concrete columns under biaxial excitation is complicated by a variety of factors. A finite‐element model is presented for the analysis of such members under generalized biaxial loading. The formulation accounts for most aspects of axial‐flexural behavior and is capable of handling numerically ill‐conditioned problems associated with softening behavior. The element formulation is initially outlined and the ability of the element to predict experimental data is demonstrated. Numerical simulations are presented of cantilever columns under static load and of an entire frame in bidirectional seismic excitation. The results demonstrate the complexity of the biaxial resistance mechanism and its influence on structural response, particularly related to the concentration of local damage in columns.

This is the second part of an investigation on probabilistic finite element methods for nonlinear concrete structures. The study is concentrated on two reinforced concrete application examples including a simply supported beam and a portal frame. Nonlinearity in material and geometry, and randomness in loading, material, and geometry are considered. Extensive computations using a probabilistic finite element method imbedded in the computer code PFRAME developed at the University of Colorado, Boulder, are carried out to verify analytical results. The effect of Taylor series expansions about different values of various random variables on structural response is demonstrated in both application examples. Furthermore, using the simply supported beam example, it is shown that the probabilistic finite element method proposed is applicable to the assessment of structural safety of material and geometric nonlinear concrete structures. Further investigations are necessary in order to develop this probabilistic finite element approach into a mature structural safety assessment method.

A fractional‐derivative Maxwell model is proposed for viscous dampers, which are used for vibration isolation of piping systems, forging hammers, and other industrial equipment, as well as for vibration and seismic isolation of building structures. The development and calibration of the model is based on experimentally observed dynamic characteristics. The proposed model is validated by dynamic testing and very good agreement between predicted and experimental results is obtained. Numerical algorithms for the solution of the constitutive relation in either the frequency or the time domain are presented. Some analytical results for a single‐degree‐of‐freedom viscodamper system are presented. These results are useful to the design of vibration‐isolation systems. Furthermore, an equivalent viscous oscillator is defined whose response is essentially the same as that of the viscodamper isolator. Finally, the model is employed in the analysis of a base‐isolated model structure that has been tested on a shake table.

The paper is based on the results of a detailed underwater survey conducted in 1989 on the United States Navy timber pier in upper San Francisco Bay. The survey was performed within the framework of the large multiyear program set up by the U.S. Navy for the inspection and maintenance of its facilities worldwide. The paper describes causes of damage, its type and location, methods of quantitative structural evaluation, and repair strategy. Inspection findings indicated a very substantial upsurge in marine borer activity, which effectively put the 45‐yr old pier out of service during the last decade. Analysis of the dynamics of pile deterioration became possible due to continuous follow‐up observations. It clearly demonstrated the advantages of the multiyear inspection program. Structural evaluation revealed that partial replacement of timber piles infested by marine borers should be made using regular pattern, regardless of the observed local condition of timber piles at the time of inspection.

Nine sites across the United States are examined to obtain pertinent macro time variability statistics of wind loads. The parameters developed are those used in many of the existing probabilistic‐based load combination methods. It is found that a clustering model is appropriate for the macro variations in wind. The parameters are verified by comparing extreme annual and 50‐year winds obtained from the developed model with those statistics previously obtained by other researchers. Estimates of composite statistics for the contiguous United States are made. These statistics include the mean rate of occurrence, mean duration of the cluster, mean duration of individual winds, and mean number of winds per cluster. Parameters for an extreme type I distribution of wind speed intensity are developed in terms of the 50‐year design wind speed.

Live load is an important load component for highway bridges. The paper summarizes the available data base and formulates the approach to calculating maximum moments and shears for various time periods. The live‐load data base consists of truck‐survey results. Moments and shear forces are calculated for surveyed trucks and various spans. The maximum load effects for time periods from one day to 75 years are derived by extrapolations and simulations. Calculations are performed for single‐lane and two‐lane bridges. For single lanes, the maximum moment and shear are caused by a single truck for spans up to about 100 ft (30 m). For longer spans, two trucks following behind each other govern. For two lanes, the maximum effect is obtained for two trucks side‐by‐side, with fully correlated weights. For the maximum 75‐year moment or shear, each side‐by‐side truck is represented by the maximum 1.5‐month truck. The effect of a 1.5‐month truck is about 15% smaller than that of a 75‐year truck.

This is the first of a two‐part paper involving a fundamental study of bond in fiber‐reinforced cement composites. The concept of a relationship between the bond shear stress and the local slip at the interface between the reinforcement and the matrix for fiber‐reinforced, cement‐based composites is relatively new. Moreover, the difficulties involved in conducting strain measurements on short discontinuous fibers have made it impossible so far to develop experimental bond shear stress versus slip curves for fiber‐reinforced concrete. This paper involves a fundamental study of the bond‐stress‐slip relationship between steel fibers and concrete. The analysis consists of a primal problem, whereby a complete pullout curve is predicted from an assumed bond‐slip relationship, and a dual problem in which the bond‐slip relationship is obtained from an experimental pullout curve. The frictional behavior of the interface is also related to and explained by the shrink‐fit and fiber‐matrix misfit theory. The theory derived in this paper is validated in a second paper using experimental pullout load versus slip curves.

This is the second of a two‐part paper involving a fundamental study of the bond‐stress‐slip relationship between steel fibers and cement‐based matrices. It presents a validation of the analytical model developed in the first part by providing comparisons between experimentally and analytically predicted pullout load curves of steel fibers embedded in cementitious matrices. Very good agreement is observed. The model is also used to derive, for the first time, bond shear stress versus slip curves typical of fiber‐reinforced concrete and considered a constitutive property of the fiber‐concrete interface. This has good practical significance since, while pullout tests of fibers are easy to conduct, there is currently no experimental method that directly yields their bond shear stress versus slip relationship.