In 2008, the Nobel Prize in Physics were awarded for Dr. Yoichiro Nambu, Dr. Makoto Kobayashi, and Dr. Toshihide Maskawa. Dr. Kobayashi and Dr. Maskawa are graduates of Nagoya University. In this report, we will review on the particle physics, which they studied. It will be claried what we know about the particle physics and what we do not know. Recently we obtain several clues about the particle physics from the observations of the universe. It will be mentioned why and how the cosmology is related with the particle physics.

View full abstract

The tuned cradle mass damper (TCMD) relies on the motion of a swing mass on a curved surface to dissipate structural vibration energy. The objectives of this study are to obtain constant swing speed of the TCMD for a large amplitude of the swing and to verify its performance through experiments when the structure is under free and forced vibration. In order to obtain a constant speed of the device, the variable radius of the curved surface were calculated by using simple pendulum dynamics. For this study, the damper was installed in a one-story simple rigid frame model with a frequency of approximately 1 Hz.

View full abstract

In this paper, a series of physical modeling with different sample preparation methods and types of basal support was conducted to investigate vertical pressure profiles beneath a planar valley of loose sand. A linear elastic solution of stress field for semi-infinite valley loaded by self weight was derived and compared with the experimental results. Despite of different geometry, the solution obtained from an elliptic equation system with an assumption of self-similarity was found equivalent to a classical Fillunger's solution of a planar semi-infinite wedge with body force. The Fillunger-based solution is found to underestimate the measured vertical pressure. Moreover, the analytic stress field was considered as an inadmissible state because linear elasticity failed to satisfy the limit stress condition along cascading down slopes.

View full abstract

This paper develops analytical methods to predict the bending behavior of skewed thick plates under transverse load on the Winkler foundation, which has not been re ported in the literature. The thick plate solution is obtained by using a framework of an oblique coordinate system. To include the effect of transverse shear deformation, the Mindlin ' s theory is employed to analyze the plates. First, the governing differential equation in that system is derived, and the solution is obtained using deflection and rotation as derivatives of the potential function developed here. This method is applicable for arbitrary loading conditions, boundary conditions, and materials. The solution technique is applied to three illustrative application examples, and the results are compared with literature and numerical solutions derived by the commercial finite element package ANSYS 11. All approaches yielde d results in reasonable agreement.

View full abstract

Tensegrity structures give us so much motivation to find a rational form by shape analyses. A tensegrity structure with virtual stiffness has so many equilibrium shapes and it may be hard to obtain a target solution by designation of a set of primary conditions. This study shows the results of some numerical experiments whose aim is to find rational forms of tensegrity structures. Form finding analyses for tensegrity towers in a gravitational field often bring unexpected equilibrium shapes, in such a case; the incremental analysis by compulsory displacement is effective. Moreover, when we consider that a tensegrity structures constructed by virtual elements behaves in an equilibrium system, some solutions are connected each other on equilibrium path. In this paper, load-displacement curves include bifurcation path are also shown and the properties of the equilibrium system are discussed.

View full abstract

Ambient vibration tests have attracted increasing attention over the last few decades because they can be performed economically with the structure under working condition without artificial loading. Ambient modal identification techniques do not require knowledge of the loading but they assume that it is statistically random. A Bayesian approach provides a fundamental means for extracting the information in the data to yield information about the modal parameters consistent with modeling assumptions. Issues do exist in the implementation and interpretation of results. This paper presents an overview of a Bayesian frequency-domain approach for ambient modal identification. Field data from a tall building is used to illustrate the method. Amplitude dependence of modal parameters can be investigated.

View full abstract

This study deals with a damage detection method using a time reversal technique which does not require reference data at an intact state but only data at a current state to detect damages. The method assumes linear damages such as notches or holes, and is applicable to damped structures. A tone burst force is firstly input to a structure at one point A and the acceleration response is measured at the other point B. Then the time-reversed acceleration response is input at point B as a force and the acceleration response is measured at point A. This response is then time-reversed and is defined as “reconstructed wave”. Damage existence and location can be detected simply from the waveform of the reconstructed wave. Numerical analysis was carried out on a plate structure with and without a rib, and the effectiveness of the method is verified.

View full abstract

For detecting defects inside a square billet, pulse echo method is generally used. This method has a problem that it is difficult to detect a tiny defect because the received signal power obtained by the reflected wave is too small. Then, we has proposed a detection method by ultrasonic CT method using time-of-flight of ultrasonic signal instead of the pulse echo method. However, the ability for detecting a defect by this method is unrevealed. In this paper, to clarify the ability for detecting a defect by this method, we numerically analyzed this ability by varying scan pitch and signal frequency. The result by changing the scan pitch showed that the necessary scan pitch depends on the signal wavelength. The result by changing the signal frequency showed that it was efficient to use the lower frequency signal for detecting the tiny defect. This was caused by that the measured receiving wave contained some effects of phase delay which was generated by the interference among the direct wave, the diffracted wave by the defect and the scattered wave from the defect. In this result, it was found that high frequency was not necessarily useful for the practical purpose of detecting defects. This method is more efficient for detecting the tiny defect than pulse echo method, by which it cannot be detected. In addition, it was found that the scan pitch can be set larger in this method.

View full abstract

This study has proposed a novel method for identifying degraded areas in geotechnical structures such as piping holes and cavities due to seepage flow using data assimilation. An existing method for identifying degraded areas using the particle filter, which was proposed by the authors, cannot identify the degraded areas with high accuracy in case of simultaneous identification of several degraded areas. To overcome the above technical issue, the more advanced method with a step-by-step identification procedure is newly proposed. In this paper, numerical tests of a geotechnical structure loaded by a road-roller are conducted to study whether the proposed procedure can identify several degraded areas. The results of numerical tests show that the proposed procedure can identify several degraded areas in geotechnical structures with high accuracy.

View full abstract

In the performance-based design concept, the chances in which the limit state of the displacements is speciied, will increase. Under the situation, the more accurate prediction of displacements will be required, and the observational approach have been developed for the soft ground engineering. In this research, as an observational method, the particle ilter (PF) is employed. The PF can make the identiication of the elasto-viscoplastic parameters possible. Particularly, several sampling methods to generate the particles, which mean the displacements and the pore water pressures calculated from the Monte Carlo method, are examined, and the effective method is discussed. Furthermore, not only primary consolidation parameters, but also the secondary consolidation parameters are identiied in this study.

View full abstract

Data assimilation, using the particle filter (PF), and incorporating the soil-water coupled finite element method, is applied herein to choose the elasto-plastic constitutive model and to identify its parameters based on the sequential measurements of hypothetical soil tests and an actual construction sequence. An appropriate constitutive model can be selected by identifying some parameters within the Exponential/Logarithmic Contractancy model (EC/LC model) proposed by Ohno et al., which covers a wide variety of constitutive models, including the Cam-clay and the modified Cam-clay models, and intermediate models with different yield curves. The hypothetical observed displacements of a soil specimen for CD tests were obtained through synthetic FEM computation, and actual measured data were used for the settlement behavior of Kobe Airport constructed on reclaimed land.

View full abstract

This paper focuses on approximation properties for strong/weak discontinuities in the PU-based finite element methods. The extended finite element method (X-FEM), which is one of the PU-based FEM, enables us to approximate the discontinuous deformation due to crack/interface with the enrichment functions. In contrast, the finite cover method (FCM), which is also one of the PU-based FEM, is capable of representing discontinuous behavior by defining the multiple sets of finite covers (elements) instead of using the enrichment functions. We examine the properties and the effects of these different PU-based approximations for strong/weak discontinuities in this paper. Section 2 shows two types of approximations of strong/weak discontinuities in the PU-based FEM by taking X-FEM and FCM for instance, and explains the equivalence of them. Several numerical examples are presented to examine the analysis accuracy and effi ciency for structures involving a line crack, a branched crack and a material interface in Section 3.

View full abstract

Time integral algorithm is one of the major factors for numerical stability and precision. Many methods have been proposed to calculate time integration of an equation of motion. The Newmark family of algorithm, such as Newmark's beta method, is popular in practical calculations. In this study, however, some characteristics of FETD (finite element time domain) method with Bubnov-Galerkin formulation are investigated. Period and phase of numerical solutions calcualted by a GSSSS (Generalized Single Step Single Solve) algorithm with a cubic interpolation shape function are calculated and summarized. According to the calulated results, numerical damping can be observed, if a time step is around the third of a natural period, eigen values of a trainsient matrix are complex values, i.e., numerical damping is observed. If a time step exceed a natural period, numerical solutions will be unstable. On the other hand, a time step is less than one third of a natural period, numerical results show good agreement to the analytical solutions.

View full abstract

We propose a characteristic Galerkin scheme using B-spline basis functions based on a mixed formulation for incompressible flows which is capable of being second-order accurate in time. One of significant features of the B-spline basis functions is that the B-spline basis functions of p degrees have at most C^p−1 continuous derivatives although classical finite elements based on Lagrange polynomials have only C⁰ continuous derivatives. We use the mixed formulation to velocity and pressure fields, which satisfies the inf-sup condition. The velocity element is obtained by a subgrid of the pressure element. The pair of the pressure and velocity elements is referred to as the subgrid (SG) element, and it allows for both velocity and pressure at the highest regularity. In order to confirm effectivity of the present scheme, we perform some numerical experiments.

View full abstract

Some kinds of element test are generally conducted in order to evaluate mechanical characteristics of structural members that have complex microstructures such as steel and concrete composite beams. However, the element test with a part of the structural member can not faithfully reconstruct deformation state in actual structures. To that end, we formulate a beam with averaged mechanical properties in order to evaluate mechanical properties of three dimensional Timoshenko beams with microstructures.

View full abstract

A parallel computation method is investigated to predict the motions of complicated-shaped solid objects in 3D free-surface flows, taking account of their collisions and fluid-solid interactions. Since the applicability of the prediction method has been confirmed through the comparisons with experimental results in our previous studies, an emphasis is place on the parallelization of the computational method in this paper. The present parallelization is based on the 3D domain decomposition method using flat MPI. In particular, an effective collision-detection technique using two scales of grids is newly implemented to find the neighboring objects and contact-detection spheres. Numerical experiments have been conducted to understand the computational efficiencies when the deviations of object numbers arise among the subdomains. In addition, it was demonstrated that this computational method is applicable to the sloshing including about 100,000 spheroid-shaped objects and the dam-breaking flow with about 5,000 tetrapod-shaped objects.

View full abstract

In this paper, we present a technique for simulating free surface flow of non-Newtonian uid. The method is based on the min-max problem which automatically enforces the governing equation and the boundary conditions. The method can avoid complicated calculation of free surfaces. We apply the method to the Particle-in-Cell method and demonstrate examples of our technique simulating three problems; dam break, water drop, and concrete slump test, and the numerical simulations show plausible results.

View full abstract

This paper presents a convolution quadrature time-domain boundary element method for a fluid-saturated porous medium. Large scale wave analyses for a fluid-saturated porous media using the conventional time-domain boundary element method (BEM) are not so popular for the following reasons: 1) No time-domain fundamental solutions are known for the problem, 2)The conventional time-domain BEM sometimes causes numerical instability, and 3) needs much computational time and memory. The formulation presented herewith improves these disadvantages using a convolution quadrature method (CQM) and MPI parallelization. The scattering problems of an incident plane wave by cavities in the poroelastic media are solved to validate the accuracy, computational efficiency and formulation of the present method, and investigate the effect of interaction parameters between uid and solid on elastic wave scattering.

View full abstract

This paper describes computational method for large deformation elasto-plasticity with nodal-integration stabilized finite element method in Eulerian formulation. The employed stabilized method is conventional SUPG (Stream Upwind Petrov Gelerkin) method. The SUPG method has been widely used to capture interface for uid dynamics in the Eulerian framework, however, a special computational treatment to apply the SUPG method to path-dependent solid deformation is indispensable. In the Eulerian formulation for the path-dependent solid deformation, stress and any path-dependent variables (for example, equivalent plastic strain in elasto-plasticity) must be advected. Although the path-dependent variables are defined on integration point in mesh, the SUPG method can advect variables in only node. In the present paper, variables are approximated on node using information in mesh, and advects the variables on node with the SUPG method. We test the present approach in representative computational examples.

View full abstract

This paper proposes an algorithm of interface capturing using convergent calculation. Three dimensional piecewise linear interface calculation (3DPLIC) has been shown its preciseness not only in two dimensional problem but also three dimensional problems. However, some algorithms of three dimensional PLIC is known to be complicated and prohibitive in three dimensional problem. This paper proposes an new algorithm to determine the interface plane based on the fraction of cell volume and local normal vector. The proposed algorithm defines a function which returns volume of the fraction when the position of the plane is given. Utilizing the function in an iterative algorithm, the position of the interface plane is decided. A series of examinations of the computational time and precision of 3DPLIC with proposed algorithm are presented. Computational time and precision are discussed in the context of comparison with an existing advection method, THINC/WLIC. The results show that the proposed algorithm is superior to the existing method in precision.

View full abstract

This paper presents an interactive mesh modification system using virtual reality (VR) technology. The mesh quality is evaluated and the quality is improved by the node relocation. The user can change the position of nodes of finite element mesh in VR space by using a handy controller. The software is developed by VR programming languages; Open GL and CAVE library. The present system is applied to the mesh modification for the simulation of 3D solid analysis and is shown to be a useful tool to assist the high quality computing.

View full abstract

For nondestructive testing, we developed a simulation tool of elastic wave propagation in a material with complex outer surfaces or various inclusions. The tool was based on the elastodynamic finite integration technique (EFIT) with an image-based modeling. In this study, the image-based EFIT code is accelerated using graphics processing units (GPUs) with the CUDA Fortran programming language. The methodology on the speed-up using the CUDA Fortran is described, and checked by numerical experiments with multiple GPU boards. The calculation speed with a GPU can be dramatically improved in comparison with the speed obtained by running the same simulation on a classical CPU. In addition, the validation of the image-based EFIT is performed by ultrasonic measurements of SH Lamb wave. The results of the simulation show good agreements with the measurement.

View full abstract

A coupling method of FEM and wavelet BEM is developed for 2-D steady-state scalar wave equation. The wavelet BEM is formulated using the non-orthogonal spline wavelets. These wavelets have the vanishing moment property independent of the order of piesewise polynomials. The coefficient submatrices of the reduced equations on the wavelet-based BEM are also compressed with the truncation of small matrix entries. The linear algebraic equations with a sparse coefficient matrix are solved using iterative solver, e.g. Bi-CGStab, GPBi-CG and GMRES. In particular, the use of GMRES shows rapid and robust convergence in the present numerical experiments.

View full abstract

The authors have performed nonlinear plate bending analyses using meshfree/particle method. Reproducing kernel approximation is used to approximate the in-plane deformation and out-of-plane deformation. Stabilized conforming nodal integration is adopted to integrate the tangent stiffness matrix to impose so-called integration constraint condition. Total Lagrangian formulation is adopted for solving plate bending problems with geometrical non-linearity. In this paper, an enforcement technique of essential boundary conditions is introduced using Multiple Point Constraint (MPC) method to solve thin-plate buckling problems. As numerical examples, nonlinear plate bending problem and thin-plate bucking problem are analyzed to validate the proposed approach.

View full abstract

This paper presents an interactive visualization system for 3D flow simulation based on unstructured grid for the CAVE system. The present system is available to the visualization of scalar and vector fields, and is developed by VR programming (Open GL and CAVE library). User can select a visualization method by operating the controller in VR space. The improvement of the computational time and the accuracy of visualization are investigated. The present system is applied to several numerical examples and is shown to be a useful visualization tool to investigate the three-dimensional flow problems with complicated geometry.

View full abstract

This paper presents a convolution quadrature time-domain boundary element method for 2-D elastic wave propagation in general anisotropy. Boundary element method (BEM) is well known as an effective numerical approach for wave propagation problems. However, the conventional time-domain BEM has a critical disadvantage: it produces unstable numerical solutions for small time step size. To overcome the disadvantage, we develop a new time-domain BEM based on the convolution quadrature method (CQ-BEM) for 2-D general anisotropy. As numerical examples, the problems of elastic wave scattering by a cavity are solved to validate the present method.

View full abstract

The present study proposes topology optimization of a microstructure for composites considering the macroscopic structural response applying a decoupling multiscale analysis based on a homogenization approach. The stiffness of the macrostructure is maximized with a prescribed material volume of constituents under linear elastic regime. A gradient-based optimization strategy is applied and a semi-analytical sensitivity approach is introduced. It was verified from a series of numerical examples that the proposed method has great possibility for microscopic advanced material designs.

View full abstract

Compacted soil is widely used to build earth structure. Although construction management of the soil progresses day by day, it is said that mechanism of compaction effect isn't fully made clear. This is due to the complication of compaction mechanism of unsaturated soil. Recently, some constitutive models for unsaturated soil and SWCC models were proposed and advanced the study on unsaturated soil. In this study we regarded compaction as the consecutive cycle of compression and expansion of unsaturated soil under undrained condition, and simulated it with unsaturated soil/water/air coupled finite element analysis. In this study, we expressed the characteristics of the mechanical behavior of compacted soil, which has not been explained for a long period of geotechnical engineering. And we considered the difference of compaction procedure affected the quality of earth structures. The achievement can be managed reasonably the compaction of earth structure.

View full abstract

Compacted soil is widely used for earth structures. However, the mechanism of compaction has not been explained, so it is difficult to grasp distributions of stress, void ration, and soil moisture within compacted earth structure. Therefore, estimation of safety and stability of compacted earth structures exposed to natural disaster is complicated. Compaction is decreasing void ratio by applying stress under certain water content and pushing air out of soil mass. Thus, the mechanics of unsaturated soil, which includes air within void, is needed for understanding of compaction. In this study, we regarded compaction as compression and expansion phenomenon of unsaturated soil under drained air and undrained water conditions and formulated it in initial-boundary-value problem. Here, the influences of permeability and soil water retention characteristics on compaction were considered and static compaction tests were simulated. Moreover, multi-layered compaction was examined.

View full abstract

Coal fly-ash is a product of the coal-burning process in thermal power plants. Coal fly-ash has self-hardening property and increases its stiffness with time. Therefore, it has been reused as a concrete aggregate in the production of concrete up to now. Recently, coal-fly ash is mixed with dredged soil from construction waste and reused for reclamation. In order to reuse coal fly-ash in the geotechnical engineering field, the self-hardening property in the framework of the constitutive model needs to be expressed. In this study, the self-hardening property is assumed as increases of the frictional angle and the yield stress. Our model is formulated for the initial-boundary problems with finite element method. Triaxial tests and the self-weight consolidation test are simulated with soil/water coupled F. E. code.

View full abstract

Recently, the failure of embankments, such as levees and small embankment dams for irrigation reservoirs, has occurred more frequently because of a greater chance of severe typhoons and localized heavy rains. Overflow is known as a primary cause of embankment breaks. The purpose of this study is to develop a numerical method which can predict the breach process of an embankment caused by overflow. This paper presents two-dimensional and three dimensional numerical simulation of the embankment breaching. The finite volume method with a Riemann solver is applied to numerically solve shallow water equations for computing the overflow on the embankments and the configuration change of the embankment profiles is successively calculated in accordance with the erosion rates of the embankment materials as a function of the bed shear stress exerted onto the embankment surface. In order to achieve the stable computation, the surface gradient method was incorporated into the finite volume discretization. The comparison between the numerical results and experimental investigations has revealed that this numerical method stably computes the embankment profiles subjected to overflow breaching and the time up to embankment breaks can be well predicted.

View full abstract

Ultimate bearing capacity of ground is an important subject in practical engineering, but the accuracy of Terzaghi's formula, which is widely employed in design, has not been clarified sufficiently. The problem of the formula is to employ a simple linear strength model into soils. The size effect in ultimate bearing capacity can't be evaluated in the formula properly. This paper develops a rigid plastic constitutive equation for soils the strength of which is non-linear against confining pressure. Higher order non-linear yield function is introduced into to express stress dependent soil strength adequately. The dilatancy property of soil is introduced into the constitutive equation by a penalty method to solve the boundary value problem explicitly with kinematical property. The applicability of proposed method has been examined through some case studies on ultimate bearing capacity analysis against vertical and inclined loads.

View full abstract

A huge number of boulders rest on slopes along the railway. In an attempt to evaluate the degree of rockfall risk of these boulders effectively and quantitatively, basic experiments, together with eigenvalue analyses by 3-D FEM, were performed for revealing the vibration characteristics of the rigid bodies simulating boulders. When the natural frequency of the rigid body, together with its shape exposed above the ground level and the ground strength, is known, the in-soil penetration depth of the rigid body may reasonably be estimated.

View full abstract

The coastal area of eastern Japan was hit by Tsunami due to the gigantic earthquake on March 11th 2011. Tsunami gave serious damage to Rikuzen Takada and drowned out famous pine forest, called “Takadanomatsubara”. However, only one of 70,000 pines could survive miraculously without falling down. People decided to preserve this pine as the monumental pine for the symbol of recovery. However, there are some problems, groundwater rise due to land sinking and salt concentration from seawater. These factors can kill root system. In this study, the current situation around the monumental pine was simulated with soil/water/air coupled F. E. code applying the mass transfer equation first. Next, the alternative preservation methods were examined. Consequently, it was found that pumping up from deep inside of sheet pile is effective for inhibiting salt diffusion and flashing is effective for washing salt away from ground.

View full abstract

In recent years, the flood disaster caused by ‘guerrilla’ heavy rain has been increasing in many regions. Once the flood disaster occurs in an urban area, the economic damage becomes high denomination because of the highly and widely developed underground spaces including subway and underground mall. In this paper, we focus on a water-stop structure made of fabric reinforced rubber and steel members, and we propose a rational design tool by employing a nonlinear Finite Element (FE) Analysis. The water-stop structure includes several kinds of nonlinearity in the mechanical response analysis. One is material nonlinearity mainly for the fabric reinforced rubber, and the other is boundary nonlinearity related to contact behavior. The second nonlinearity problem can be solved by the commercial FE software, but the first material nonlinearity cannot support with pre-set material library. Therefore, original material constitutive model, which is based on the anisotropic hyperelasticity, is applied to the nonlinear FEM by using user defined subroutine. FE analysis of the water-stop structure with our proposed constitutive mode is validated with comparisons of experimental tests, in which the hydrostatic pressure acts on the water-stop structure.

View full abstract

Case study was carried out into the causes and countermeasures against excessively deformed concrete culverts during and immediately after the construction in the embankment. Two cases dealing with the issue of typical backwards problem in geotechnical engineering were examined. In this paper, first of all, the outline of the damaged culvert as well as the surrounding embankment is in detail described. The background, together with the cause of damage, is discussed based on the results of site investigation. Secondly, it was attempted to elucidate the deformation mechanism of the embankment by means of numerical analysis, and the countermeasures are proposed. Finally, the stability of the embankment with the countermeasures was evaluated.

View full abstract

The aim of this paper is to propose the simple Plane-Cap (PC) model as well as to classify an accuracy of analysis for the response of the reinforced mortar beam subjected to low velocity impact loading. The present study deals with the Smoothed Particle Hydrodynamics (SPH) method by employing Drucker-Prager (DP) with a new PC yield surface and Von-Mises (VM) yield surface for mortar material. The influence of the different material model of DP and VM that used numerically for mortar beams are studied in this paper. Besides, the new simple cap model is employed on compression, while, orthotropic constitutive equation due to the damage effect is utilized on the tension side. Finally, the results of numerical analyses are discussed by comparing them with experimental observations.

View full abstract

Uniaxial compression experiments were carried out to study the strengths and failures of gypsum samples with some pre-existing flaws in non-overlapping geometry. These prepared samples were expected to have same strength by the stress analysis14) which is based on the simplified linear mechanism. The experimental results demonstrate the effect of the sample's geometries on the compressive strengths, and the failure mechanisms governing the strength characteristics. And then the stress analyses were improved by using the extracted failure mechanisms. The improved analyses show that the predicted compressive strength agrees fairly well with that of experiments.

View full abstract

In this paper, the bi-axial cyclic loading tests of RC and SFRC columns under the different loading histories have been performed. The influence of loading-path on the deformational performance, such as the flexural strength and ductility of RC and SFRC columns, and on the buckling behavior of re-bar has been examined in details in the post-peak loading area. Furthermore, the energy absorption capacities have been discussed in relation to the buckling behavior of re-bar and the deterioration of core concrete during cyclic loading. It has been found that SFC may be effective to make the plastic hinged zone of column more ductile, particularly for the column with rough interval of hoop tie, i.e., the interval of 120 mm in the present study.

View full abstract

The mechanism of sediment disaster induced by heavy rain should be elucidated and then the prediction method is applied to prevent the sediment disaster. It is qualitatively known that the sediment disaster induced by heavy rain is caused by the change in the mechanical properties of unsaturated soil, for example the decrease in suction. The soil-water characteristic curve (SWCC) which has hysteresis is a key relation to simulate the seepage of rain water into unsaturated soil. In this paper a numerical & mechanical model is proposed to express the main drying curve (MDC), main wetting curve (MWC), drying scanning curve (DSC) and wetting scanning curve (WSC), where the maximum and minimum void ratios are used to determine the residual and quasi-saturated degrees of saturation, S_rr and S_rs.

View full abstract

It is known that woody debris flowing down rivers accumulates at bridge piers, intake facilities and other structures. The phenomenon can cause significant damage by flooding as well as destroy of buildings. To tackle with those problems, it is important to understand the mechanism behind woody debris flow and accumulation. In this study, an experiment was conducted focusing on woody debris in a meandering compound channel in order to analyze the mechanism of woody debris accumulation. In the experiment, particle tracking velocimetry (PTV) based on binary correlation analysis was applied both to tracer particles and to woody debris models to clarify the flow field and woody debris transport, and to enable study of the relationship between them. Also, the Lagrangian numerical simulation of drift wood using a simplified momentum equation was developed in order to reproduce the experimental result. The numerical study results confirmed that woody debris concentrated avoiding high shear regions in the channel, and this behavior seems to correspond with the characteristic motion of particles in fluid, which is known as "preferential motion".

View full abstract

This paper reports on the results of a medium-scale experiment regarding the impact applied by ice floes with a length of 0.15 to 1.2m and with an impact velocity of 0.14 to 5.4m/s against a pile structure and numerical simulation using the 3-D discrete element method (DEM). Ice caused brittle failure/splitting in most cases. In those cases, we found that the increase rate of maximum impact load due to impact velocity decreased in comparison with that in non-fracture cases, and that the length of ice floe in the impact direction did not affect the impact load. We also developed a fundamental numerical method to simulate the impact of ice. The simulation results showed a close correlation with the experimental outcome in terms of the time-series change of impact loads, the dynamic failure processes of the ice floe.

View full abstract

An accurate velocity model for porous flow plays an important role in the prediction of the ground water pollution. However,there is little information about the pore fluid velocity distribution. We here focus on the pore fluid velocity distribution observed in mesoscale of a porous media composed of many irregular shaped grains. The reason is that complicated ground water pollution found in a field scale is affected by the pore fluid velocity distribution measured in the mesoscale. The Lattice Boltzmann simulator, which works on a graphics processing unit(GPU), was employed to evaluate the pore fluid velocity distribution in an accurate three dimensional digital model involving Toyoura sand and then we performed a series of porous flow simulation. The pore fluid velocity distributions obtained from the simulations are in good agreement with those measured by visualization experiments. These velocity distributions converge into a unique non-gaussian distribution under various Reynolds numbers ranging from 2 to 10.

View full abstract

Cushioning materials, such as granular mats and sand mats placed on rock sheds, are attracting attention as construction devices that can effectively disperse and reduce rock fall energy before rocks collide with protection works. To support performance-based designs for rock fall countermeasures, the present study estimated rock fall behaviors and impact forces using the discrete element method (DEM), and investigated a parameter determination method for the rational use: especially, damping factor was focused. The effects of loading conditions were also examined, such as shape of falling mass and its velocity, by considering the accumulated impulse and rate dependency of load-displacement of granular mat.

View full abstract

Recently, on-site recycling method used mobile crusher has been receiving considerable attention. As the function of the mobile crusher is to crush the concrete blocks and to use the fragments as the aggregates at the construction sites, the machine needs to have an ability to crush the concrete blocks to the target particle sizes. However, many parameters of the machine such as the discharge width and rotation number affect the crushing performance. So the effect of these parameters on the crushing performance is not cleared. In this paper, the simulation model by using Distinct Element Method (DEM) to predict the crushing performance of the mobile crusher was developed, and then the effect of several parameters on the crushing performance was analyzed by using Principal Component Analysis (PCA), which is one of multivariate analysis technique.

View full abstract

In this study, we propose a method to consider deformability of elements for a failure analysis of masonry structures using the DEM. In the original DEM, the deformability of the structure can be modeled by overlapping between rigid elements, but Poisson's effect cannot be modeled. In the proposed method, an element is divided into two parts, an inner part to consider deformation of the element itself, and an outer part to deal with contact between elements. The stiffness of the inner part is modeled using the stiffness matrix of the finite element method. When two elements are continuous or in contact, springs are set between the elements and the spring constants are estimated based on the original DEM and the length of the outer part. The validity of the method is confirmed through the comparison of elastic deformation with the FEM. It is found that the original DEM and the proposed method show different failure patterns, and considering the deformability of element is found to be necessary.

View full abstract

In this paper, the coupled thermal-mechanical processes in the Äspö pillar stability experiments (APSE) were simulated using Distinct Element Method (DEM). By considering pre-existing cracks in the rock model, mechanical response of the rock during excavation phase and heating phase were successfully represented by DEM. Simulation results agree qualitatively well with the experimental results. However, the microcracks in the simulation were widely distributed around the heater and exfoliation of rock surfaces observed in the in-situ experiment was not formed accurately. To simulate more realistically the experimental results by the DEM models, the calibration of the microscopic parameters considering the model scale should be done. Moreover, more detailed discussion on the excavation damaged zone around the borehole and the distribution of pre-existing cracks are required.

View full abstract

The mechanical & numerical model for deformation behavior of unsaturated soil is proposed, where the probability theory and statistics are applied to estimate the mechanical behavior at the contact points in unsaturated soil. Adding the consideration on the change in contact angle, i.e., probabilistic motion of soil particles, the energy surface which has been a key physical factor is found not to be needed in the revised model. The numerical experiments are carried out to compare the stress∼strain relations with those obtained from the triaxial compression tests on unsaturated sandy soil under drained condition. It is found that the revised new model can express the deformation behavior of unsaturated soil and is promising.

View full abstract

In this paper, a hysteretic restoring force model is proposed to predict the inelastic response of steel piers under strong ground motions. Instead of multiple lines used by former researchers, a series of curves are adopted to approximate the complicated force-displacement hysteretic relationship of steel piers. In this model, hysteretic rules are proposed using six parameters, which can be obtained by static cyclic loading tests, to understand the hysteretic character of piers including deterioration. To verify the accuracy of the proposed model, six static cyclic tests and eleven hybrid tests using three kinds of steel piers are conducted. By comparing results due to the simulation and the tests, the validity of the proposed method is clarified accurately in the seismic response simulation.

View full abstract

Forces that a tsunami exerts on a concrete girder bridge have been evaluated by a Computational Fluid Dynamics (CFD) method and a possible mechanism is examined that may have been responsible for damages of typical small road bridges observed after the tsunami resulted from the Great East Japan Earthquake of 2011. The CFD method employed is a fully three-dimensional Large Eddy Simulation (LES) method that allowed a detailed examination of time-resolved changes of magnitudes of the horizontal and vertical forces and the moment forces including the buoyancy effects. A typical 8m wide four-girder bridge is found to experience a moment force that is large enough to push the seaward edge of a bridge deck upwards when the tsunami comes with a bore of one to two meters. It is consistent with the indication of survey results of several damaged bridges.

View full abstract