A Study on the stability and the behavior at the critical points of nonlinear discrete structural system is presented utilizing the Taylor expansion of the governing algebraic equations. The results are compared with the previous works on the same subject. This method is found to be much simpler to explain the subject than the perturbation method together with the energy function which is the technique mostly used in literature.
Three examples on plane and space elastic truss structures are presented to illustrate the analytical results.

View full abstract

This paper presents a method to trace the bifurcation equilibrium paths utilizing the Newton-Raphson method. The initial guess solutions on the bifurcation paths are obtained by considering the higher order terms of the Taylor series expansion of the incremental equations at the critical points.

View full abstract

The purpose of this study is to establish a non-iterative efficient computational scheme to trace the nonlinear finite displacement behaviour of space frames, using the tangent stiffness equation of linearized fininte displacement of a thin-walled elastic straight beam element. Direct solution of the tangent stiffness equation is used, imposing adequately small increments. Local coordinates are updated at each incremental step, utilizing a vector multiplication scheme. Numerical results for a wide variety of spatial structures are given, demonstrating the versatility of the present scheme.

View full abstract

A new method is presented for analyzing the interaction between local and overall instability of thin-walled open polygonal section members. Based on beam theory approximations, non-linear expressions for displacements are first derived taking the distortion of cross section into account, and then the governing equations for instability of members are developed. Buckling loads for I-section beam and hat-section column are analyzed as examples, and the accuracy of the method is discussed by comparing some solutions with those obtained using the more complex method based on plate theory. The method makes it possible to predict the local through distortional to lateral instability of members using a small number of degrees of freedom.

View full abstract

Box section specimens of various types of corner joints were fatigue tested under axial loading condition by using a 4MN fatigue testing machine. The specimens were fabricated by commonly used and newly developed welding processes. The fatigue strengths of joint and behaviors of fatigue crack propagation from various kinds of weld defects in corner joints are described. Influences of weld defects on the fatigue strengths of joints and permissible blowhole size in actual truss chord members are examined.

View full abstract

In this paper, an approximate method for analyzing the bending problems of rectangular Mindlin plates on elastic foundations is proposed. The solutions of the partial differential equations of the bending are obtained in the discrete form, by translating the differential equations into integral equations and applying numerical integration.
In order to confirm the convergency and accuracy of numerical solutions, comparisons with numerical solutions obtained by other investigators are made. As the application, the bending behavior of rectangular plate on nonlinear elastic foundations and on non-uniform elastic foundations are calculated.

View full abstract

The region of Nagasaki, Kyushu, Japan was inundated by heavy rainfall on the 23rd of July, 1982, and water flow of the River Nakajima and other rivers increased remarkably in this area. Many rivers in Kyushu are spanned by stone arch bridges.
To study on the effect of flood flow on a stone arch bridge and to measure fluid pressure that may damage the bridge, a solid model of Fukuro bridge and a model stone masonry arch were prepared in the experimental water channel at Nagasaki University.
The experiment was performed for two cases: in the first case the model stone arch was set in the water channel and in the second case no model arch was used. For each case, the hydraulic data and the resultant drag and down-pull forces working on the arch stones, were measured. Using the external forces obtained in the both tests, the stability of the stone arch bridge was calculated for the flood flow. The calculation confirmed that the arch stones of the Fukuro bridge could withstand the flood.

View full abstract

A stiffness equation of plane frames is formulated in finite displacements by the total Lagrangian approach. The formulation is based on the physical interpretation of the polar decomposition theorem and does not use any variational principle. The corresponding differential equations are also given to show which theoretical solutions the discretized ones converge to. For the Bernoulli-Euler beam, the postbuckling behavior can be analysed accurately without the geometric stiffness matrix which helps to accelerate the convergence of solutions. On the other hand, the Timoshenko beam element must have the geometric stiffness in order to obtain the correct solutions. Finally the typical stability problems which have the bifurcation points in their equilibrium paths are solved to demonstrate the ability of the stiffness equation.

View full abstract

The TIG arc remelting procedure was applied to retrofitting fillet-welded joints with fatigue cracks at weld toes. Fatigue tests was performed on as-welded joints and on retrofitted joints in order to verify the effect of this procedure, and the growth of fatigue cracks was analyzed using a fracture mechanics approach. Experimental and analytical results show that the fatigue lives of cracked joints are greatly improved by TIG arc remelting when the cracks have been completely remelted. When the cracks were not successfully remelted, the fatigure lives of retrofitted joints were influenced by layer thickness of remelted zone and size of remaining crack.

View full abstract

This paper presents an optimization approach for the reliability analysis of large-scaled skeletal structures by using the revised PNET method. In order to evaluate the collapse probability of the structural system, it is needed to consider all possible collapse modes and the effects of their mutual correlations. However, it is considerablly difficult to find all possible collapse modes for the highly redundant systems. To overcome this problem, an optimization technique for the PNET method is initially developed to find the approximate collapse probability of the structure, which is defined as the sum of the collapse probabilities of the representative modes classified by the extent of correlation. In order to confirm the validity of this approach, the computational results are compared with the values found by the Monte Carlo simulation using the collapse load analysis. Two fairly large framed structures are numerically analyzed to illustrate the features and scopes of this approach.

View full abstract

Proposed herein are two procedures for the strength of compressed cylindrical shells on the basis of so-called nonlinear bifurcation theory: one being a direct approach making use of the pathological curve representing the stability limit for the elasto-plastic behaviors near the elasto-plastic buckling point, and another being a unified approach making use of the pathological curve associated with the plastic failure mechanism but near the equivalent bifurcation point.
Several numerical demonstrations are provided to give the imperfection sensitivity curves, and the ultimate strength curves as the reasonable lower limit of several experimental data, yet showing good correlation with the DASt and ECCS design curves.

View full abstract

To begin with, fixed-end restraint effects on the capacity of steel arches loaded into elastic or inelastic and finite deformation range is studied by a nonlinear finite element approach. Then, based on this result, a design formula which predicts the ultimate strength of fixed arches is proposed using effective length concept. By comparing with the analitically determined ultimate strength it is shown that these formulas accurately predict the strength of fixed arches for use in design.

View full abstract

In the present study, the problem of an infinitely extended elastic solid containing a doubly periodic rectangular array of cracks is considered. The analysis is made based on the method of pseudo-tractions, and a first-order approximate but explicit solution is obtained. The study reveals the fundamental difficulty of the problems of doubly periodic cracks which gives rise to the discrepancies between the results of the previous works. The stress intensity factors of mode I and mode II and the overall compliance are derived as functions of the crack density and geometry of the crack array. The overall compliance of solids with randomly distributed, unidirectional cracks is evaluated by the self-consistent method. The validity of the self-consistent method is discussed in terms of the results of the doubly periodic cracks.

View full abstract

In this study, a detailed review of the fatigue test results of riveted joints which have been performed in Japan, is conducted. A faigue test is carried out on a plate-girder bridge which has been used in a railroad about 70 years. The fatigue strength and the initiation and propagation behavior of fatigue cracks of weathered and deteriorated riveted plate girder is discussed.

View full abstract

Hybrid procedure is developed in order to analyze the behavior of buried pipelines in soil liquefaction processes. The procedure presented herein consists of a ground response evaluation using the finite element method and a pipe response analysis using the transfer matrix method. A rise of the ground water table induced by dissipation of excess pore water pressure is considered in the former evaluation. Numerical models show that the unsaturated layer around the pipelines liquefies due to the rise of the ground water table and accumulation of the excess pore water pressure. It is evident from the present study that the response characteristics of the buried pipelines change in the liquefaction process.

View full abstract

Dynamic collapse analysis of two types of structural foundation, deep and shallow, are performed by particle assemblies simulation with Cundall's model.
Nonlinear responses of the structural foundations by the horizontal force are obtained. Relationships between applied force and deformation with hysteresis and such phenomena as dilatancy and settlement of the particle assemblies ground are found not to be expected in the analysis of continuous media. A size of the ground model is of 3400 particles.

View full abstract

Due to fact the that Rayleigh waves are attributed to a kind of natural vibration of a surface layer, their nature can be understood by vibration experiments using a shaking table or by vibration mode analysis of a layer of finite length. This has been confirmed experimentally, analytically and numerically. In the limiting case, where the wave length approaches infinity, the wave motion is found to be reduced to shear vibration and longitudinal vibration of the layer.

View full abstract

The epicenter of the Ansei Edo Earthquake (M=6.9) which occurred on 11th November of 1855 was situated in the vicinity of Edo (present-day Tokyo) and much destruction was caused in the urban area of Edo and the nearby towns in Kanto District. In this paper, I have investigated the damages caused by Ansei Edo Earthquake on the basis of various old documents of the time in which the heavy damage is described and have determined the distribution of the damage and the seismic intensity (M. S. K Scale) in Kanto District.
Also, I have computed the intensity of this earthquake theoretically at many sites situated in the Kanto District, by using several fault models and taking the geological condition in the Kanto District into account.

View full abstract

This paper presents a dynamic response analysis of a nonproportional damping system. Emphasis is placed upon evaluations of seismic response analyses of nonproportional damping systems by means of the response spectrum method. The square-root-sum-of-squares method causes gross errors which are derived from neglecting the modal coupling effects of the nonproportional damping system with modes of closely spaced frequencies. Accordingly, an alternative modal combination procedure, which provides more accurate responses for the nonproportinal damping system such as situations, is developed. Each modal combination method that is being discussed is evaluated by comparing results with a direct time integration results, i. e. by a Newmak's β method.

View full abstract

The response of mass-on-rough-plane model to both horizontal and vertical excitations is studied statistically. The critical acceleration envelope is defined for a potential sliding surface of a structure as the locus of input acceleration that mobilizes limit resistance against slippage. A method to determine the model parameters with critical acceleration envelope is proposed. Slip displacements are computed with 52 sets of strong motion records and regressed against JMA magnitude, focal distance, peak values and spectral moments of accelerograms. The analytical prediction by Igarashi is compared in a good agreement with the computed slip displacements.

View full abstract

A method is presented to measure the magnitude of earthquakes from the slip displacement of the friction-controlled SDOF system. A statistical relation among JMA magnitude, slip displacement, focal distance, and the critical acceleration of the system is obtained with 52 sets of strong motion accelerograms in Japan. This relation is then used as a measurement equation of the magnitude of earthquakes from slip displacement. The estimated magnitudes at various sites are found to be consistent within each earthquake with an averaged standard deviation of 0.10. The proposed method can be used to measure the magnitude of earthqrakes from various sources such as slip displacement measured in a block on an inclined plane, earth dams, slopes, and other structures.

View full abstract

It is well known that lower modes of vibration are predominant in the dynamic response of suspension bridges to moving vehicles or wind. Therefore, it is needed to know their characterisics, in particular the corresponding natural frequencies.
In this brief paper, natural frequencies of lower vibration modes of suspension bridges are numerically examined, mainly with respect to a side to center span ratio besides nondimensional characteristic parameters introduced in the previous paper 2).
It is found from these results that the characteristics of symmetric natural frequencies of suspension bridges associated with lower modes depend significantly on the values of the side to center span ratio.

View full abstract

This paper presents the elastic post-buckling behaviour of propped-cantilever column predicted within the framework of FEM procedure, and the result is compared with that by the analytical elliptic-integral solution. Discussed in conjunction with the particular boundary conditions is a characteristic of deformation of this structure which is completely different from that of a cantilever column.

View full abstract

This paper offers a group theoretic study of the influence of imperfections on bifurcation buckling behavior of an octagonal, reticulated truss dome structure. The dome exhibited diversified bifurcation behavior in association with various imperfection modes chosen to be covariant with subgroups of a dihedral group. Such variations in bifurcation behavior nonetheless did not alter greatly the buckling capacities of the dome. In addition, the concept of symmetry preserving bifuraction is introduced as a potential cause of bifurcation buckling phenomena.

View full abstract

This technical note presents a geometric matrix for thin-walled members derived using distribution of linearly varying bending and twisting moments inside an element. The accuracy of the matrix is improved compared with similar matrices derived in the past under the uniform distribution of moments. A numerical example of a simply supported beam is presented in which the critical moments to cause lateral-torsional buckling are compared with the results utilizing the matrix available in literature and the matrix derived in this note to show the better accuracy of the latter.

View full abstract

The Dynamic FEM is frequently used for analyzing wave propagation problems. In case of infinite or semi-infinite media, the artificial boundaries reflect the waves. The present authors could solve the problem by applying Smith-Cundall's method. It superimposes two types of waves reflected from Dirichlet's and Neumann's boundaries. In this study, the radiation damping of structure models has been analyzed using this Finite Element Method.

View full abstract

This paper proposes an approximate method, so-called modified P-δ method, for calculating the ultimate strength of thin-walled steel frames with box cross-section by the first order elastic analysis. The decrease of ultimate strength of frames due to the initial imperfection and P-δ effects of columns is estimated by assuming the initial rotational angle of columns in this method. The validity of modified P-δ method is checked through the parametric studies on the critical strength of several types of frames with variable cross-sections by the Finite Element Method based on the elasto-plastic finite displacement theory.

View full abstract

The transient stress waves in an infinite viscoelastic slab subjected to plane strain are investigated. The solution based upon displacements which satisfy directly the equations of motion for a Voigt solid and the initial and boundary conditions is given by the method using Stokes theorem for the case of arbitrary shock loads. The shock loads are assumed to be applied on the face or in the interior of the slab. The displacements are represented in terms of Fourier transforms and eigenfunctions. The eigenfunctions are derived from the equations of motion for an elastic solid and satisfy the orthogonality relation induced by the corresponding boundary conditions.

View full abstract

The authors have been studying the damage of RC slabs of bridge by 1/10 scale model made of microconcrete under a wheel running load. The relation between the magnitude of the wheel load and number of cycles to failure (S-N curve) were obtained. They made clear the failure mechanism of RC slab under a wheel running load, and investigated the effect of the differences of curing condition and loading speed on the failure. Furthermore the diagnoses of residual life of RC slab models against fatigue loading were carried out by non-destructive testing method.

View full abstract

It is known from experimental studies, that a segment joint becomes a plastic hinge as load increases, and the relation between bending moment and rotational angle at the joint is given approximately by broken line diagram. The purpose of this paper is to present a method, which enables us to analyse segment rings with staggered joints in both elastic and plastic ranges. Transfer matrix method for grid on elastic foundation is used herein, because the segment rings with staggered joints can be modeled as rounded grid with main girders of no torsional rigidity, in which the effects of transverse beams on main girders are replaced by shear springs in vertical and longitudinal directions. Numerical example is solved by iteration changing alternately the half components of state vector at a boundary cross section which is caused by cutting off segment rings.

View full abstract

Material Properties of SS 41 steel in high temperature is represented in function form by assuming an independence between instantaneous plastic strain and time-dependent creep strain, in which the latter is formulated by a rheology model including the effect of non-stational creep. Effect of prior plasticity on subsequent creep is discussed. Deformation of multi-story one-span steel frame structures in fire is analyzed using finite element method. Strength of structure during fire is evaluated by a serviceability limit load.

View full abstract

Headed studs are exclusively used as shear connectors in various steel-concrete composite constructions. The directions of concrete placing in relation to stud connectors in fabrication of composite constructions influence static and fatigue strength of studs and the composite action between steel and concrete. They can be classified to four cases. In most of the past studies, however, the effect of directions of concrete placing have not been sufficiently given consideration except the one by J. W. Fisher. The present paper discusses the effects based on the results obtained by push-out type test series and proposes design formulas for static and fatigue strength of stud shear connectors.

View full abstract

It seems there are too many formulations about optimum depth of a plate girder and one may wonder which is most preferable for his case. The authors studied the problem by theoretical but rather classical way taking into account every factors having effect to the depth.
The result shows the effect of specified allowable compressive stress formulation and the number of sectional changes in a girder must be considered and these factors lower the optimum depth of it. In case of rather short span, the practical limit on the compressive flange area has large effect, even that is not explicitly specified in a code.

View full abstract

Ultimate strength behavior of elasto-plastic thin-walled H-section stub-columns are studied experimentally and theoretically. Seven specimens with various width-to-thickness ratio of the component plates of the cross section are tested under uni-axial compression. The test results are compared with the results of nonlinear analysis by a finite element method with the consideration of the in-plane flexural component of the plate. Validity and efficiency of the present method are shown. Based on the test results, an interaction formula for predicting the ultimate strength of H-section stub-column under axial compression is proposed.

View full abstract

The present study is an attempt to discuss the relationship between the magnitude of two dimensional sectional model response and three dimensional full model response. Particular attention is paid on the effect of mass parameter on it. First, a simple theory is introduced to compare two dimensional model and three dimensional model response amplitude. Second, the simple taut strip model test and sectional model test were carried out to verify the theoretical prediction of three dimensional model response based on two dimensional test result. Comparison of the present approach with the conventional ideas of taking mass parameter in section model tests follows.

View full abstract

St. Venant torque term of the yield surface equation is often neglected in the plastic analysis of the beams with thin walled open-cross sections. However, after a plastic hinge occures, the hinged section will be free from the state of restrained warping, and so, St. Venant torque terms in the flow rule may play more important role. The yield surface equation of curved I-beam is proposed herein, which is expressed in terms of bending, warping moments and St. Venant torque. Several tests have been carried out with the curved I-beams to prove the availability of the proposed yield equation. The results show that the proposed equation is reasonable and suitable for the simple plastic analysis of the curved I-beams.

View full abstract

The diagnosis of bridge serviceability has been performed by a combination of field inspection of the bridges such as a visual inspection and static and dynamic loading tests and an analytical study. Because the mechanical behavior of bridges with various damages could not be made clear, it is very difficult to estimate exactly both of the damage mode and degree of damage of existing bridges.
In this paper, the sensitivity of both static and dynamic behaviors of bridges for their damage assessment is proved through the experiments on model bridges induced some specified artificial damages. And also a method of damage assessment of bridges based on these behaviors is discussed in detail.

View full abstract

When the fuzzy set theory is applied to serviceability diagnosis of structures, the problem of the estimation of a membership function which translates a subjective information related to diagnosis into a mathematical model is often pointed out.
In this paper, an attempt is made to formulate the membership function using a function with various parameters for evaluation of the subjective uncertainty on the diagnosis. And a process of serviceability diagnosis of structures based on such a membership function in the fuzzy set theory is discussed in detail. In order to make a quantitative comparison between the results by the present method and those by the previous one, several numerical examples are made.

View full abstract

Concept of a digital-type active optimal control of steel structures is studied especially on the following two points; that is, (1) improve of calculation time using complex modal analysis and (2) evaluation of an effect of control parameters.
Since the state equation of control in a discrete-time system contains a matrix exponential function, which gives a large defect on the calculation time and on the stability of solution, orthogonal technic based on the complex modal analysis is used, and it gives a successful result. Control state of the system is strongly influenced by the weight functions, which appears in the Hamiltonian of Pontrjagin's maximum principle, the parametric study is provided on the effect of the functions. Numerical examples are given.

View full abstract

The interaction behavior of the local and overall buckling of welded thin-walled H-section columns are studied experimentally. A total of 13 specimens with various slenderness ratios and width-to-thickness ratios of the component plates are tested under axial compression. The AISC form factor method for predicting the ultimate strength of H-section columns is revised for those composed of plates with large width-to-thickness ratios. The predictions of the revised method are found to give fairly accurate results for the columns tested.

View full abstract

Bacause of the presence of various kinds of imperfections (or errors) introduced during construction processes of bridges, the geometrical shape and stress distribution do not always conform to those intended by designers when completed. Such construction errors represent important quality characteristics of the bridge. The authors present a methodology to evaluate stochastically the effects of imperfections, such as the variation of dead load and stiffness parameters and dimensional imperfections of each member, upon construction accuracy of bridges by applying stochastic finite element method. The effects of cable length adjustment, which is usually taken at the erection of cable systems, on construction accuracy are also taken into account. Through numerical examples, some useful and important informations concerning the quality control of steel bridges are obtained.

View full abstract

The fatigue behavior of gusset joints (plate with gusset welded on the surface) for steel structures under various environments are presented and discussed on the allowable fatigue strength at 2×10⁶ stress cycles. Fatigue tests were performed on the specimens of three groups: exposed to air, immersed in 3% NaCl solution and alternately exposed to both. The test results obtained from exposure to air show that the fatigue strengths of corroded gusset joints were higher than those of the uncorroded ones. The fact may be explained by the effect of the increase of toe radii of fillet welds. Corrosion fatigue strengths at 2×10⁶ stress cycles were lower than the fatigue strength of specimens exposed in air. As for the final group, the S-N_f curve fell between the two curves obtained from the two tests.

View full abstract

Geometrically nonlinear theory of rods with shear deformations is developed. Particular attention is paid to investigate the coupling of finite rotations due to bending, twist and shearing. A finite rotation vector plays an important role in a formulation of the present problem. When the equilibrium equations and the associated boundary conditions are derived from the principle of virtual work, the magnitude of displacements, rotations, and strains is treated as finite one. The stress-strain relationships proposed herein differ slightly with the existing ones. They yield, however, the well-known and widely accepted constitutive equations expressed by the stress resultants and moments and the generalized strains. The accuracy of the present equilibrium equations is confirmed through comparisons with those obtained by the equilibrium method.

View full abstract

The embedded steel plate cellular bulkhead has been developed as a new type of sea wall which is applicable even for poor sea beds and deep water sites. As the sea wall is a widely extended and huge structure, economical construction is needed. In this study, the optimization technique was applied for the minimization of the total construction cost of the embedded steel plate cellular bulkheads including the ground improvement cost. This technique was applied also for the clarification of the relationship between the design factors and the constraints. Through the numerical computation, the optimum cross sections of the embedded steel plate cellular bulkhead and the optimum area of ground improvement were obtainedd and the useful informations for the optimum design of this structure were provided.

View full abstract

The purpose of this study is to evaluate seismic performance of column and rahmen type steel piers with thin walled box cross-section during strong earthquake excitation.
We describe the method to replace the piers into SDOF systems, and the numerical response analysis of these SDOF systems by use of five earthquake motions. Based on the numerical results, we discuss how the difference of the type and height of piers, and the ratio of thickness to width of cross-sectional plate affect on the total number of times of yielding, maximum plastic displacement, hysteretic and input energy and the relationship between hysteretic energy and elastic maximum velocity. We also discuss the effect on the earthquake damage index, as defined by the authors, and found out that this index is very useful to evaluate the damage including the collapse of the piers.

View full abstract

Vibration pollusions caused by factory machines like a forging hammer are still one of the social problems in Japan. On the deseign of the foundation for such shock producing machine, it is necessary to consider the radiated waves caused by foundation vibration. This paper investigates, from the point of view mentioned above, dynamic response of machine foundation under impact force and radiated waves from a foundation. For this purpose the experiments were carried out using the model half space made of the RTV silicone rubber. Also the approximate method to analyze the wave radiated from the foundation under impact force was presented and applicability of this method was investigated.

View full abstract

A model experiment was carried out on the seismic behavior of a buried pipeline in a partially liquefied ground. A model pipeline was laid in an unliquefiable superficial soil layer, and a part of soil layer underneath the superficial layer was made liquefiable by using loose fine sand.
When liquefaction occurred, the superficial layer and the pipe showed a characteristic behavior which could be simulated with simple mathematical models with a high accuracy. Calculations of strains in buried pipelines of the actual sizes by using the above mathematical models showed that extremely great strains could be produced in the pipelines during the actual earthquakes. Thus the models of partially liquefied ground will give a most probable mechanism of pipeline failure associated with soil liquefaction.

View full abstract

This is a report on a study of parameters estimation, such as lateral pressure acting on retaining wall and elastic constants of soil, etc, from measured data of the wall displacement (degree of rotation) during excavation. EK-WGI method which was developed based on the extended Kalman filter of controll theory was applied to the parameters estimation method. A feature of this method can be estimated recursively unknown parameters in the direction of the depth of retaining wall. In the analysis, simulated data which were used for verification of formulation to EK-WGI method was applied. From numerical results, it was clearly observed that this method was effective to estimate these parameters.

View full abstract