Desirable ways for improvement of research and implementation of applied mechanics in Japanese civil engineering are identified and discussed. Current Japanese society is demanding rivision of styles of conventional disciplines toward enhancement of productivity.We first review the sustainability condition of researchers and engineers. Next we generalize the basic concepts of applied mechanics so as to treat a wider range of problems. We finally apply our method to safety problems and, specifically, to earthquake engineering and nuclear cycle management.

抄録全体を表示

Using the X-FEM (eXtended Finite Elememt Method), we analyze the energy release rate rate at the onset of crack kinking for an isotropic linear elastic material in an infinite plate with a centered crack under an uniaxial tension load. In the X-FEM, we propose a simple method in which only the four-node quadrilateral isoparametric elements in the fractured element are employed without the enrichment by the near-tip asymptotic solution. In the trial example for a stationary crack, we find that this method gives the rather good approximation for the stress distributionear the crack tip.The result of the energy release rate for any kinking angle using the path independent E-Integral shows a very high accuracy in comparison with a known theoretical result.

抄録全体を表示

This paper is concerned with the effect of the T-stress in microcrack shielding problems in anisotropic materials. The interaction between a macrocrack and associated microcracks in anisotropic elastic material under the T-stress effect is analyzed and comparedwith the results in the isotropic material. Three cases where a microcrack is in front of the main crack tip, a microcrack is on the side of the main crack tip, and two microcracks are on the both side of the main crack tip are investigated. The analysis is performed by an extended Pseudo-Traction Method. The stress intencity factor of the main crack tip and its incremental value due to the existence of microcracks and the T-stressare computed. The T-stress determines the magnitude of shielding and/or amplification due to the inclined near-tip microcracks. The effect depends on the positon of the microcrack, the inclination angle of the microcrack, the sign of the T-stress, the magnitude of the T-stress and the material anisotropy.

抄録全体を表示

This paper presents an application of Genetic Algorithm combined in Neural Network System for decision maker's retrofit priority. Sometimes it has difficulty to decide and standardize an evaluation function of retrofit priority, because large variety of demands exists in the problem depending on local culture or decision maker's personality. In order to reflect these demands, the proposed method makes use of Neural Network System which studies the knowledge of engineer's judgment on the paired comparison. The proposed method evaluates very well the retrofit priority following the real priority sense of value.

抄録全体を表示

We dealt with simple unsteady flows in a pipe with a vertical shaft at the inlet. A non-linear analytical solution of fundamental equations for unsteady pipe flows was derived firstly under a few hydraulic conditions with and without oscillatory inflow discharge at the inlet. We investigated the non-linear behavior such as the amplification of discharge variation near the resonance relation, which indicates the time period of free oscillation in the pipe flow system. The non-linear solution was compared with the results of numerical analysis to verify the solution. The hydraulic experiments were also carried out in this research to verify the applicability of the unsteady pipe flow model. It was pointed out through these considerations that the theoretical findings are in good agreement with the numerical and experimental results.

抄録全体を表示

In order to develop quantitative stress measurement in the material, stress inversion technique combined with laser speckle interferometry method is proposed. Three different shapes of aluminum specimen are used, and stress increment fields under the monotonic loading are studied. The result shows that the spontaneous stress field can be obtained by means of the sum of stress increments with the continuous displacement measurement.

抄録全体を表示

A new numerical algorithm for nondestructive identification of the boundary conditions applied on a 3D body based on photoelasticity is developed. This new method for boundary conditions identification is based on the load incremental approach, which linearizes the nonlinear governing equation of photoelasticity by considering small increments of applied load. In an earlier attempt, a method for stress identification in a 3D body was developed based on load incremental approach by the authors. In this new attempt, the earlier method is strengthened by applying the equilibrium constraint and instead of directly identifying the state of stress, identification of boundary conditions is considered. The use of equilibrium constraint not only reduces the computational cost but also stabilizes the numerical scheme in the load incremental approach

抄録全体を表示

A modified three-dimensional inverse scattering method is applied to reconstruct the shape of three-dimensional flaw in a cylindrical specimen. In the modified method, a measurement plane is restricted to the plane perpendicular to the axis of the cylindrical specimen. Thus the crosssectional image of the flaw can be obtained in this modified method. Moving the measurement plane along the axis of the cylindrical specimen, the cross-sectional image is obtained for each measurement plane. The three-dimensional flaw image is reconstructed by piling up the obtained cross-sectional images. Cylindrical specimens with a cylindrical cavity model and with a hemispheroidal cavity model are prepared. The performance of the modified method to reconstruct the three-dimensional flaw is confirmed by using the experimentally measured waveforms.

抄録全体を表示

An inverse method is developed to estimate advective surface velocities from sequential sea surface temperature (SST) data. The present method employ an advective diffusion equation of SST and the functionals in tems of advective diffusion equation and spatial curvatures of velocity components of u and v to estimate 2-D surface velocities from sequential SST data. The functionals with the spatial curvatures play an important role as a stabilizer and can detect small scale SST variations so that inverted velocity fields include the velocity components of the same scale. The numerical simulations show that inverted velocity components agree well with the target velocities when the SST data contains small scale variations.

抄録全体を表示

We construct a high order finite difference method in which the quadrature points can be set at arbitrary locations. It is known that the Cauchy problem of the Laplace equation is unstable with respect to L²-norm. We apply our method to Cauchy problems of the Laplace equation. The finite difference approximation in this research is formulated as follows: the derivative of a function is replaced with a linear combination of values of the function for each quadrature points which are located arbitrarily. We propose an algorithm in order to determine weights in this linear combination. In numerical experiments we obtain a highly precise numerical solution. These results imply effectiveness on solving the Cauchy problem of the Laplace equation by our method.

抄録全体を表示

Rigid plastic analysis is widely used for the analyses of metal forming process or bearing capacity problems. Theory and numerical formulation of rigid plastic analysis is investigated from the point of nonlinear optimization, in this paper. At first, duality of the upper bound analysis and the lower bound anlaysis is expounded from the theory of Lagrangian duality. Then, the interior point method is applied for the duality and the complementality of the rigid plastic analysis to formulate the hybrid rigid plastic finite element method (HRPFEM). A numerical code of the HRPFEM is developed to examine its performance. Three different meshes for the same surface loading problem are used to check the validity and the convergence of the numerical analysis.

抄録全体を表示

An analytic averaging method of moduli of composites is built into a standard constant-straintriangular element in plane strain. Such an analytic method is useful from a designing point of view. Furthermore it may have an advantage in suggesting some bounds of behaviors of composites. The element is used to show a relation between moduli and randomness of distribution of inclusions. And two boundary value problems are examined to demonstrate feasibility of the elements, and the results are compared with those by the standard homogenization method

抄録全体を表示

In this paper, X-FEM (eXtended FEM) proposed by Belytschko et al, Mose et al is modified by means of a relative discontinuous displacement on the crack surface. In comparison with the relative discontinuous displacement, the enriched nodal degree freedoms in the X-FEM have no direct physical meaning. Also, the number of enriched freedom decreases in this modified X-FEM. For the problem of infinite plate with acrack, the result of the modified X-FEM shows better approximation than that of the original X-FEM.

抄録全体を表示

By virtue of the development of metallurgy and the theory of dislocation, one can qualitatively anticipate mechanical properties of a material from its microstructure. Moreover, not qualitative but quantitative and precise anticipation is more beneficial for efficient development of new materials. In this context, we utilize a multi-scale method with atomic model, which is a particle system that has potential, and discuss the possibility of quantitative anticipation of material properties including plastic deformation through several representative numerical examples.

抄録全体を表示

For the simulation of rock fall, the robust rigid body dynamic simulation algorithm based on the technique used in the computer graphics field was developed. The rotation of rigid body was represented using quaternion was used and voxel based collision detection algorithm was developed. The strategy for determining time step was described. Several examples are shown to demonstrate theefficiency, robustness and applicability to the non-convex objects of the method, and comparedwith polygon based method. It was shown that by changing the level of voxel representation, it was possible to control the accuracy and computational efficiency.

抄録全体を表示

We propose a global-local computational method for simulating the phenomena of global objects with local heterogeneous structures. This method is based on a nonlocal voxel FEM combined with interface elements. The voxel FEM is used to achieve high accuracy of the numerical result for local structures, whereas the interface elements are introduced to glue independently modeled finite element subdomains since the nodes of voxel mesh and the standard finite element mesh do not coincide along their common interface. The performance of the proposed method is well demonstrated by representative numerical examples.

抄録全体を表示

An explicit dynamic method has recently been used in finite element structural and solid analyses especially for industry. Very little literature on the explicit dynamic method, however, exists.In addition, treatment of an arbitrary Lagrangian Eulerian (ALE) and Eulerian mesh controls for the explicit dynamic method is different from a case of a implicit method. This paper describes the explicit dynamic method in various mesh controls.

抄録全体を表示

An optimization method for design of the reinforcement of rock mass structures by rockbolts is presented. The method utilizes the topology optimization scheme that has been developed on the basis of the mathematical homogenization theory. The installation spacing among rockbolts is employed as a single design parameter. The mechanical reinforcement by rockbolts is quantitatively evaluated through the homogenization analysis and represented by two characteristic functions of the design parameter, such as the homogenized elastic tensor and the quadratic safety factor which are defined over the reinforced rock mass. By use of these two functions, three functionals representing a sort of stiffness or a safety index of the reinforced rock mass structure are defined. Then, by employing these functionals three kinds of optimization problems are formulated in the form of mathematical constrained maximization or minimization schemes and solved by the optimal criteria method. Optimal distribution of spacing of rockbolts is determined for a model rock slope through these optimization schemes, and their validity is examined and discussed.

抄録全体を表示

The paperless design is becoming popular in the manufacture with the advance of computer technology. However, the design based on the freehand drawing and the beam theory is still common in the bridge design field.
This paper proposes the method of designing the steel truss structure with the help of the optimization technique and the finite element method. The grid computing technology and the replacement with a Bezier curve are adopted in this approach for the further improvement in the computing speed.

抄録全体を表示

We study the performance of the mortar method incorporated with the dual Schur method for elastodynamic problems for structures modeled with incompatible finite element meshes.Although this method ensures kinematic continuity throughout the specified interfaces, such as material interfaces and interfaces between different mesh topologies, we are concerned with the performance of the method especially for the case that two mutually incompatible meshes have different sizes of finite elements and different material properties.Several numerical experiments are conducted and the results provide the excellent performance of the method which is equivalent to that of the standard FEM with compatible meshes.

抄録全体を表示

A new computational method for incompressible fluid has been developed on the collocated grid system.This method consists of an implicit algorithm (C-ISMAC) for the prediction stage in a MAC method, a fifth-order QSI scheme for non-linear terms and an improved pressure computation method (C-HSMAC).The developed method, including the grid generation, was completely parallelized on the basis of a 3D domain decomposition method.As a result of the computation of cubic cavity flows, the improvements of the computational efficiencydue to the C-HSMAC and 3D domain decomposition methods were clearly confirmed.In addition, the velocity profiles in the 3D cavity were reasonably predicted with the present method.

抄録全体を表示

We develop an analysis method for discontinuous deformation without re-meshing. The proposed method is based on the Finite Cover Method (FCM), which is actually an alias for the Manifold Method. By virtue of the mesh-free nature of the FCM, propagating discontinuous boundaries can be traced regardless of initial mesh discretization. Furthermore, the nonlocal material modeling is introduced to avoid the dependency on mesh size. Representative numerical examples demonstrate the performance of the proposed method and well simulate the propagation of discontinuities with independently generated FCM mesh.

抄録全体を表示

In this paper, we attempt numerical simulations of ruptures of bubble at water surface by using the Level Set method. The Level Set method proposed by Sussman et. al. is one of the Gas-Liquid two phase analyzing method. Firstly, we try to investigate the effect of the numerical scheme for advection term on the performances of the Level Set method by solving a dam break problem. Especially, volume conservations and mass conservations are checked. The third order TVD-MUSCL scheme was used for advection term of a movement equation, and the CIP method was used for advection term of a distance function. Secondly, the numerical simulations of ruptures of bubbles are conducted.

抄録全体を表示

This paper applies an Eulerian hydrocode to fracture analyses of solid material. The hydrocode treats the solid like fluid and includes not only momentum but also mass and energy equations for conservation. The Eulerian description fixes meshes on space and materials flow through the fixed meshes. Hence, the Eulerian code allows arbitrarily large deformations and new free surfaces without a special algorithm. Fracture phenomena, which are difficult to compute by conventional structural codes in the Lagrangian description, can be simulated by the Eulerian hydrocode.

抄録全体を表示

This paper presents an accurate interface-capturing method based on CIVA/VOF method for the analysis of free surface flows. The incompressible Navier-Stokes equation and the advection equation is employed for the basic equations. The stabilized FEM based on the SUPG/PSPG method is employed in order to solve the velocity and the pressure. The CIVA method is employed to solve the position of the interface accurately. A new volume correction method is also presented. The present method is applied to the two dimensional broken-dam problem and three dimensional sloshing problem. The computed results show the validity and efficiency of the present method.

抄録全体を表示

This paper presents a stabilized finite element method for the non-linear free surface flows with an arbitrary shape of wall and an outflow boundary.The incompressible Navier-Stokes equation based on arbitrary Lagrangian-Eulerian description is used as the governing equation. The SUPG/PSPG formulation is employed for the finite element discretization.The outflow boundary condition and the parallel computing method are also introduced to compute the large-scale free surface flow with an outflow boundary.As numerical examples, the present method is applied to the sloshing problem in a rectangular tank, a trapezoidal tank and the open channel flow of the meandering waterway.The efficiency of the present method is shown by numerical results.

抄録全体を表示

In the analysis of the mechanical behavior of deformable body, FEM is reliably andextensively used since FEM is based on well-established continuum modeling and numerical analysis method for BVP. However, when it comes to the analysis on failure behavior, interpolationby overlapping smooth shape function used in FEM makes the situation difficult. Tedious techniques such as remeshing or joint elements are needed. On the other hand, DEM is known to be an efficient numerical computation method for failure-related problems. However, lack of rigorous method to determine spring constants prevents DEM to be a popular method for analysis on failure behavior of deformable body.
To cover the above mentioned problems on FEM and DEM, a new method for numerical simulation for failure behavior of continuum has been developed. The new method proposed here, namely, FEM-β, provides a rigorous block-spring model for deformable body with easy treatment of failure. This model can be regarded as i reformulation of DEM in the framework of FEM, or ii introduction of non-overlapping discontinuous shape function to FEM.

抄録全体を表示

By incorporating the constitutive framework of crystal plasticity into the multi-scale modeling based on the mathematical homogenization theory, we introduce the multi-scale analysis method for polycrystalline metals. The macroscopic material behavior of polycrystalline metals is characterized by the microscopic mechanical behavior affected by the shape and the chain axis of crystal grain, whereas the constitutive model for a single crystal is formulated by means of the multiplicative decomposition o the deformation gradient. Several numerical analyses are conducted to evaluate the macroscopic yield surfaceof a polycrystalline metal and examine the micro-scale inelastic behavior, which is very complex and inhomogeneous in the microstructure composed of several crystals.

抄録全体を表示

The finite cover method (FCM) is applied to elastic-plastic problems and its performance is evaluated in comparison with the standard finite element method (FEM). First, we examine the influence of the location of integration points for stress evaluation in the FCM. Next, using the reference solution available in articles, we demonstrate the performance of the FCM to reproduce the mechanical behavior that is typical in elasticplastic plate with a hole. Then, the performance is also examined for the approximation at the material interfaces of multi-phase materials. Finally, after proposing a new modeling method for FCM with level set function, we realize the capabilities of the FCM to characterize the elastic-plastic mechanical behavior of meso-structures of materials.

抄録全体を表示

The element-free Galerkin method.(EFGM) is known as an alternative numerical approach that does not require the finite element sub-division. The authors have been already presented an EFGM, which is based on the Lagrange polynomial, and applied to finite displacement analyses of plates and beams. In this paper, we reformulate our EFGM for the elastic analysis of 3-D bodies. As a numerical example we adopt a bending problem of thick plates. Especially, we investigate the affection of numerical parameters to the 3-D analysis. In order to evaluate the efficiency of the present EFGM, FEM and Fourier analyses are also performed. It follows from numerical results that the present EFGM gives us excellent approximation for the linear elastic analysis of 3-D bodies.

抄録全体を表示

Numerical solutions for shape identification of flow past a circular cylinder are treated in this paper.The Sakawa-Shindo method is employed as a minimization algorithm.Thus, unified computational approach to the simulation of flow and the shape identification is presented.As a numerical approach to spatial discretization, mixed interpolations by the stabilized bubble function and linear elements are applied for velocity and pressure.

抄録全体を表示

The generalized minimal residual method (GMRES) is a well-known iterative method for solving large non-Hermitian linear systems of equations.Since GMRES becomes increasingly expensive and requires more storage as the iteration proceeds, it generally uses restarting, which slows the convergence.However, if the new starting vector is chosen appropriately at the time of the restart, this can improve the convergence.By the implicitly restarting GMRES (GMRES-IR) method, approximate eigenvectors determined from the previous subspace are included in the new subspace and this deflates the smallest eigenvalues.We apply ths GMRES-IR method for the analysis of the boundary value problem related to the diffraction wave field around a Very Large Floating Structure (VLFS) and compare it with the usual GMRES method.

抄録全体を表示

In the fast multipole method for Helmholtz' equation, one usually uses the conventional FMM for low frequency problems and the diagonal forms for high frequency problems. This isbecause the conventional FMM is inefficient in high frequency problems while the diagonal formbreaks down in low frequency problems. It is, however, desirable to have one single code which covers the whole frequency range from the point of view of the efficiency. In this paper, we develop a fast method of analysis for Helmholtz' equation in 2D which can handle any frequency by switching between the diagonal form and the convention multipole method in the process of the calculation. We also investigate preconditioners which reduce the number of iterations required to solve linear equations for BIEM.

抄録全体を表示

The scattering analysis of torsional wave in a rod with a notch is carried out by using the mode exciting method. In mode exciting method, a set of torsional wave modes are excited by appropriate boundary conditions given on the virtual edges of the rod. The excitedtorsional wave modes constitute a system of equations, which can be used to determine the scattering coefficient for each torsional mode. To investigate the ability of torsional wave in detecting and characterizing notch in a rod, 8 types of notches, which differ in the depth and the shape, are analyzed. The scattering coefficients of Oth mode and 1st torsional wave mode incidence are shown for the frequency below the cut-off frequency of 3rd mode. The results show that Oth mode is more sensitive to notch than 1st mode, which is understandable considering the distribution of power as a function of radial position in the rod. The Oth mode isalso suitable to the characterization of notch.

抄録全体を表示

In the present paper, we investigate theoretically and experimentally the number of non -zero matrix entries generated by the wavelet BEM with a Beylkin-type compression algorithm. The Beylkin-type algorithm, which is based on a prescribed level-independent threshold, retains the asymptotical convergence rate of BE solution, like widely-used level-dependentcompression schemes. The coefficient matrix compressed by the Beylkin-type scheme has O (N1+r)(0<r<1, N: degree of freedom (DOF)) non-zero entries; the level-dependent schemes enable us to reduce the matrix entries up to O (N (log N) a)(a&ge;1). However, for matrix compression usingthe Beylkin-type scheme the compression rate is greater than or comparable to thatunder the Schneider's level-dependent scheme, in the moderate DOF range.

抄録全体を表示

This paper presents the degenerate modes and modal coupling in vibrating rectangular plates with the same boundary conditions along four edges.The spline element method is applied to analyze frequencies and mode shapes of orthotropic rectangular plates with tapered thickness.Some of degenerate mode pairs are split in frequency in vibration of isotropic square plates with the four edges clamped, free or simply supported, respectively. And nondegenerate modes such as X mode or ring mode can occur in tapered square plates and orthotropic rectangular plates having certain width/thickness ratio and aspect ratio.

抄録全体を表示

This paper presents an analysis of the vibration and the loss tangent of viscoelasticity orthotropic thick rectangular plates using the three-dimensional spline strip method.To demonstrate the convergence and accuracy of the present method.several examples are solved.and results are compared with those obtained by other numerical methods based on the theory of elasticity and Mindlin plate theory Good accuracy is obtained.The effects of thickness/width ratio and variation of elastic modulus on frequency parameters of wooden plates are investigated.
A loss tangent of wooden plates is calculated by the present method on the basis of Maelntyer's procedure, and the influence of thickness/width ratio and boundary conditions on the loss tangent is examined.

抄録全体を表示

A discrete method is proposed for analyzing the bending and free vibration problems of isotropic and orthotropic plates with uniform or non-uniform thickness.The fundamental differential equations involving Dirac's delta functions are established and satisfied exactly throughout the whole plate.By transforming these fundamental equations into integral equations and using numerical integration, the solutions are obtained at the discrete points. The Green function, which is the solution for deflection at the discrete points, is used to obtain the characteristic equation of the free vibration.Some numerical analyses are carried out for the bending problem of isotropic plate with uniform thickness, the free vibration problem of isotropic with uniform thickness and orthotropic plates with variable thickness in one or two directions.The efficiency and accuracy of the numerical solutions by the present method are investigated.

抄録全体を表示

Although many experimental researches have been carried out to examine bearing capacity of spread foundations, numerical analyses result in poor predictions of localized deformation patterns and their developments, except a few studies with softening effects or a double-slip theory. We here use the double-slip theory to simulate numerically developments of the deformation localization under spread foundations.A main focus is put on effects from the boundary conditions under the footing as well as a nearby slope.Some of the numerical results show patterns of localized deformations in good agreement with the experimental observations

抄録全体を表示

Incremental elasto-plastic average behaviors of composites with many different inhomogeneities are simulated on the basis of the Mori-Tanaka theory. An emphasis is put on the initial yielding and flow rule of such composites, and the average behaviors are compared with the classical plasticity models. It is shown that an approximate model of yielding of compositeswhich have random slip systems can be described by the von Mises yield condition and theordinary associated flow rule.

抄録全体を表示

A simple slip-weakening friction law, which is presented by a reduction of friction coefficient with the increasing fault slip, is adopted to simulate the fault rupture process at the initial stage of earthquake generation. It is implemented into GeoFEM with Newmark method for dynamic analysis. The fault is modeled by master-slave contact element. A large-scale 3-D simulation of dynamic fault rupture of a part of Northeast Japan area with over 2 million finite element nodes is performed on the Earth Simulator. The attempt of such a large-scale FE simulation has been achieved and the dynamic fault rupture process is presented.

抄録全体を表示

The paper presents a bond slip analysis between concrete and steel using the rigid body spring model (RBSM) and the finite element method (FEM) combined analysis. A new RBSM model proposed here in has 3 couples of spring normal to and parallel to the failure plane located at the center of the element with an arbitrary angle, and 4 nodes at the out side corner of the element so that this element can be installed in the FEM platform without any interface. This model can take into account the bilinear elastic plastic gap behavior and Mohr-Coulomb slip behavior including softening in elastic plastic behavior. The proposed combined model is validated by simulating Pull out test of rebar under 3 kinds of load conditions; i. e., static, high speed, impact.

抄録全体を表示

It is necessary to construct the analytical method that can reproduce real phenomena accurately so as to discuss the safety of rock structures withheterogeneity, and to design them economically andrationally. Thus, in this study, the design flow of rock structures with heterogeneity is proposed and the effects of heterogeneity on the behaviors of rock structures are di scussed. In the design flow, at frrst, the constitutive equation of the rock is determined by the comparison between a analysis and a laboratory experiment with rock core sample. Next, the numerical examination with the determined constitutive equation is carried to investigate the safety and durabihty of rock structures. Trough the comparison to experiments, the small region of heterogeneity has little influence on the macroscopic behaviors of rock structure

抄録全体を表示

Time dependent constitutive equation for soft rock based on the chemical reaction rate theory is presented. The present equation is qualitatively evaluated using experimental result obtained from either monotonic loading or creep test for artifitial soft rock. Stability and localization condition for the constitutive equation is, further, obtained from stability of irreversible problem. The localization condition is qualitatively compared with the experimentally obtained results.

抄録全体を表示

The eminent effects of anisotropy on mechanical properties are frequently reported, and the corresponding constitutive models are proposed. In some cases the resultant constitutive models failed to satisfy the fundamental requirements and the effects of anisotropy are formulated in an arbitrary and inappropriate manner. In this paper the consistent formulation of the anisotropic behabiors is discussed based on the modified stress method, the features of the method are simple, consistent, and mathematically well defined. Some examples of the modified stress method are discussed with an emphasis on frictional geomaterials.

抄録全体を表示

To examine the effect of the initial soil structure on slaking susceptibility of Tertiary mudstone, one-dimensional compression tests are carried out for three kinds of tertiary mudstone with a different resistance to slaking. These one-dimensional compression properties are simulated by the modified Cam-clay model with Super/subloading yield surface and rotational hardening. The new findings are as follows.(1) The initial stress state of the mudstone with a high resistance to slaking in e-σv space is above the normal consolidation line (NCL) of the clay that was remolded from the mudstone, while the state of the mudstone with a low resistance to slaking is below the NCL.(2) From the model, the mudstone with a high resistance to slaking can be regarded as the high structured and overconsolidated clay, while the mudstone with a low resistance to slaking can be regarded as the heavily overconsolidated and less structured clay.

抄録全体を表示

It is well known that viscosity of water-saturated clay is affected by temperature variation as well as strain rate. In addition, the conservation of energy during deformation should be considered since the plastic stress power changes the temperature of soil. In this paper, thermohydro-mechanically coupled finite element method using an elasto-thermo-viscoplastic constitutive model for clay is newly developed. And then, a series of strain localization analysis under undrained plane strain conditions and thermal consolidation mainly due to heatgenerated pore water pressure are simulated by thermo-hydro-mechanically coupled finite element method.

抄録全体を表示

The knowledge of material properties of solidifying concrete is of major importance during construction process when many subsequent damages can be prevented. In this paper an evolutionary material model which can be used during the whole period when solidifying concrete undergoes the transition from a liquid to a solid state is proposed. The evolution of the material parameters is governed by a function of water/cement ratio and elapsed time after mixing which corresponds to the degree of hydration of concrete. The model can be used for modeling uniaxial behavoir of solidifying concrete including creep. The parameters for the evolutionary function were determined from three experimental methods for uniaxial loading.

抄録全体を表示

The object of this research is to develop an accurate particle detection method for image analysis. Image analysis with in-situ storage image sensor (ISIS) camera is expected to be a powerful tool for finite deformation problem of continuum medium under high-strain rate. Accurate particle detection is of importance in this application, since the accuracy of detected strain strongly depends on that of particle detection. Tracking of Gaussian particle is combined with the size and configuration of photo diodes on ISIS. Then, the problem of particle detection is successfully reduced to the optimization problem with three variables. The proposed method results in the accuracy of 0.011 pixel according to the numerical simulation.
Finally, the image analysis with proposed method is applied to the crack propagation problem under the impact loading. Strain concentration preceding the crack propagation is obtained through the image analysis.

抄録全体を表示

To investigate the development of shear band, numerical analyses were carried out.These analyses were performed by finite element method with infinitesimal and finite deformation theory using subloading tij model.This model can take into consideration the influence of density and confining pressure in three-dimensional stresses particularly. The results of these analyses show that the analyses using this model can describe the development of the shear bands in normally and over-consolidated clay.

抄録全体を表示