To reveal vibration characteristics of bender element and help improving accuracy of shear wave velocity measurement in bender element test, the author has measured the vibration of a self-monitering bender element with a laser displacement sensor. The test results have revealed that the self-monitoring element shows complicated motion derived from at least two bending deformations not only in the length direction but also in the width direction.

By applying the modal analysis, the measured motion has been fitted to a series approximation which can be described by superimposition of some dominant vibration modes with different predominant frequencies including the two bending motions in the length and width directions. The fitted motion has well reproduced the observed motion by employing at least the two dominant vibration modes. Consequently, it has been concluded that the vibration characteristics of bender element can sufficiently be formulated by estimating modal parameters for the two dominant vibration modes.

Swarm-based algorithms are a powerful family of optimization techniques inspired by forming flocks, colonies and swarms. Here we show that a superior performace in convergence and precision of cuckoo search algorithm and gley wolf optimizer, which are recently developed meta-heuristic algorithms, through benchmark functions. The numerical optimization problem in groundwater discipline was carried out to estimate the spatial distribution of hydraulic conductivity in a limestone region in subsurface dam. Identified groundwater levels through seepage analysis were good agreement with the observed data not only at monitoring wells but at pumping wells where observation data were not utilized in search computation. Subsequent solute transport simulations using random walk particle tracking linked with a spatial pollution risk by nitrate associated with seasonal variation of groundwater pumping, demonstrating a unique application derived from inverse analysis to a practical design in the field.

We have proposed a method to identify the natural frequency of a single structure from the data obtained by measuring the vibration of bridges and viaducts. In the method, the natural frequency of a single structure can be theoretically calculated by using the undamped natural frequencies and natural modes of the whole structure. The eigenvalue analysis was performed, and it was shown that the natural frequency of the single structure can be identified by the proposed method. Moreover, we evaluated the effect on the identification accuracy when the undamped natural frequencies of the whole structure contained errors, and it was found that the errors included in the high order natural modes had a large effect on the identification accuracy

Due to global warming, torrential rain disasters have been occurring in various places, raising the demand for predictive simulation technology. In order to perform numerical simulations, material parameters are often determined based on observation data or empirical laws. In this context, this work presents the application of a class of neural networks, PINNs (Physics-Informed Neural Networks), to inverse problems. The characteristic of PINNs is its prediction is guaranteed by the informed physical laws, initial, or boundary conditions, by forming the loss function as a combination of predictive and physical loss. Predictive loss is the difference between PINNs prediction and the ground truth, while physical loss is defined by how much PINNs prediction satisfies the physical laws and conditions. This paper investigates the effect of each loss and the performance improvement by introducing a weighting factor, and discusses PINNs applicability to parameter estimation with noised observation data.

In this study, we develop linearized inverse scattering analyses for 2-D viscoelastodynamics and attempt to reconstruct shape and position of domain and boundary type defects in viscoelastic media. Inverse scattering formulations to reconstruct domain and boundary type defects in viscoelastic media are derived using the Born and Kirchhoff approximations. The scattered waves required for the inverse scattering analysis are calculated by the convolution quadrature time-domain boundary element method (CQBEM) which can produce high accurate and stable numerical solutions even no time-domain fundamental solutions are known for the problem, such as viscoelastic wave propagation. After the formulations for 2-D linearized inverse scattering analysis for domain and boundary type defects in viscoelastic media are shown, some numerical results for these defects are shown to validate the proposed methods.

Since, in homogeneous beams, the displacement perpendicular to the axis by the Poisson’s effect is not constrained, elementary beam theory uses a one-dimensional elastic constitutive law. However, in composite beams consisting of multiple materials with different Poisson’s ratios, the multiple materials constrain each other’s displacement in the perpendicular direction to the axis. Since the axial stiffness and the bending stiffness are affected by this displacement constraints, accurate evaluation of the stiffness needs to consider the Poisson’s effect. In this paper, we propose a beam theory that can consider Poisson’s effect as an extension of the beam theory including the cross-sectional deformation. Comparison of the solutions of the proposed beam and continuum shows the validity of the proposed beam theory.

This paper presents a 3-D elastodynamic time-reversal method using the topological sensitivity for the defect detection indicator in order to determine the number and position of defects in an elastic solid. The time-reversal method has been developed as one of the inverse scattering techniques in the field of the ultrasonic non-destructive evaluation and earthquake seismology. However, in general, it is difficult to quantitatively evaluate the convergence positions of the time-reversal waves in time-domain. The authors et al. have been studying the reconstruction of defect shape by utilizing the topology sensitivity for the calculation of the time-reversal method for scalar wave problems and 2-D elastodynamic problems. In this research, we extend the previous researches for such problems to 3-D elastodynamic problems., The convolution quadrature time-domain boundary element method (CQBEM) for 3-D elastodynamics is utilized to calculation 3-D elastodynamic wave fields. As numerical examples, some defect detection results are demonstrated to investigate the performance of the present method.

This study investigates the propagation characteristics of P-wave in granite. Specifically, the local phase velocity is evaluated experimentally using a granite plate. In the experiment, ultrasonic P-waves transmitted in the thickness direction are measured by scanning the plate surface with a laser Doppler vibrometer. The phase velocity is evaluated based on the phase delay relative to an independently measured reference signal. The phase velocity data obtained thus are processed statistically to produce probability density functions (PDF) for each rock-forming mineral species. As a result, a highly asymmetric and broad PDF is obtained for the P-wave velocity of potassium feldspar, while the PDF for quartz is found to be symmetric and much narrower. Finally, the PDFs are used to generate 2-dimensional numerical models of granite on which plane wave propagation analyses are performed. The numerical results showed that the phase velocity field perturbed according to the experimentally obtained PDFs gives a peculiar decay profile compared to the Gaussian perturbed models.

Monte Carlo calculation is used in the seismic risk analysis of infrastructures that consider various parameter uncertainties; however, the calculation cost increases as the parameters become higher in the nonlinear time history analysis with seismic load input. In this study, we verified surrogate modeling by the Gaussian process regression for the input/output relationships of the seismic analysis of a typical seismic isolated pier. By using a constructed surrogate model, the number of structural analyzes could be effectively reduced. Then, it was shown that the Automatic Relevance Determination (ARD) kernel can appropriately and automatically extract the degree of influence of the structural parameters on the maximum seismic response for the occurrence of different nonlinearities.

The huge tsunami induced by the 2011 Tohoku Earthquake caused severe damage to the breakwaters in the coastal area in Japan. Breakwater failure, in general, has multiple causes; (I) horizontal force due to hydraulic head difference, (II) scouring behind the caisson, and (III) piping erosion induced by seepage flow beneath the caisson. This failure can be defined as a fluid-soil-solid multiphase problem of a tsunami, rubble mound, and caisson block, respectively. This paper presents a multiphase simulation of the seepage failure using a stabilized ISPH-DEM coupled method. In this tool, the stabilized ISPH seamlessly simulates free surface flow and seepage flow in the rubble mound, with DEM applied to express the rubble mound deformation and the motion of the rigid body-treated caisson block. The caisson block is subject to the hydrodynamic force calculated by ISPH and contact forces from the rubble mound obtained by DEM, and the movement should be treated as a moving boundary condition in the ISPH fluid simulation. Fluid-soil coupling is performed with a relative velocity-dependent drag force model, allowing the overlap of each phase. The numerical simulation using the proposed method qualitatively reproduces the overall failure modes of breakwaters, such as the rubble mound deformation by seepage flow, and the caisson block slide and rotation by horizontal force. In addition, a situation in which the surface particles of a rubble mound are easily moved is created and seepage collapse occurs, resulting in the representation of the behavior of sand blowing and backward erosion during seepage collapse. It is inferred from the numerical analysis that localized failure, which cannot be traced by conventional continuum-based analysis methods, may have a significant influence on the collapse of caisson breakwaters.

In recent years, open Sabo dam has been damaged due to a large-scale boulder debris flow. A new design concept of structural safety against a large load level is discussed. The new design load is largrer than the conventional design one. This study proposes a dynamic debris flow load model subjected to open Sabo dam as a new concept. First, debris flow load is measured from a view point of dynamic load distribution. The load is focused on the overturn stability limit state of Sabo structure by using the device that overturn in exceedance condition of the limit state. The resistance moment at the overturn limit was good agreement with the measured action moment. It was considered a dynamic load model that combines the deposited gravel load acting on the lower part of the structure and the debris flow load acting on the upper part. The proposed load model was examined in comparison with experimental results and the overturn stability obtained from FEM analysis. Furthermore, it was clarified that the maximum uplift of the dam model depends on the time integrated value of excess moment with respect to the resistance moment.

A computational method was proposed to calculate the viscoelastic fluid flows with free surfaces, which are the boundaries between viscoelastic fluids and surrounding Newtonian fluids. In the proposed method, phase-averaged governing equations for the fluids are solved using the finite volume method in the collocated grid system. The Giesekus model, which is a nonlinear viscoelastic model, is used as the constitutive equation of the viscoelastic fluid. The proposed method was first applied to the flows of Newtonian and viscoelastic fluids between parallel plates. As a result, it was confirmed that distributions of calculated velocity and viscoelastic stresses in steady-state are in good agreement with theoretical solutions. In addition, the die swell phenomenon was calculated. The viscoelastic fluid was injected into the area filled with the Newtonian fluid from a gap between parallel plates. It was shown that the proposed method enables us to predict the swelling of the injected viscoelastic fluid as reported in the preceding studies.

In this study, a harmonic balance-boundary element method (HB-BEM) is extended to three-dimensional scattering problems by a closed crack. The HB-BEM is a numerical method which deals with steady-state wave scattering by a crack with contact acoustic nonlinearity. In the conventional study, the two-dimensional analysis has been carried out using HB-BEM. However, three-dimensional effects should be taken into account for realistic modeling. Boundary conditions which model clapping and rubbing motions on crack faces are extended to the three-dimension. A penny-shaped crack subjected to plane longitudinal wave or transverse wave is simulated as a numerical example. The validity of the proposed method is demonstrated by comparing the present numerical results to the ones obtained by the conventional time-domain method.

The purpose of this study is to investigate the effect of notches on fatigue strength of middle (i.e., SM490) and high (i.e., HT780) strength steels. The study is based on finite element analyses on two types of notched specimens referring to an experimental work found in the literature. A novel approach is proposed to characterize the fatigue performance of the samples based on elasto-plastic analyses carried out with an unconventional plasticity model, the Fatigue SS model, proposed by the authors. The fatigue life, expressed as the sum of crack initiation life and crack propagation life, was predicted under unidirectional cyclic loading conditions (R = 0). The main advantages of the novel methodology consist in considering the effect of irreversible deformations and the effect of the local geometry on the initiation and propagation of the crack. The numerical results were validated against experimental data proving the reliability of the new assessment method.

We studied the number of Gaussian quadrature points for a curved beam in the NURBS-based isogeometric analysis. For comparison, a cantilever beam drawn by the B-spline function was also investigated. In the case of a cantilever beam, the number of quadrature points increased with the relation "degree (of B-spline) + 1". On the contrary, in the case of a curved beam, the degree did not affect clearly, but the number of elements or weight decided the number of quadrature points. Finaly, we propose a simple index that can be used to determine the number of quadrature points for NURBS solid elements.

The response of stress of hiperselastic materials such as rubber generally exhibits complex non-linear behavior. Although the conventional identification of the material constants are carried out by using results of three types biaxial material tests, the results of stress analysis using these identified material parameters are not always satisfactory in accuracy. In this study, a novel methodorogy for identification of material parameters is proposed, which incorporates the importance evaluation of results of material testings in material constants by considering the deformation state appearing in stress analysis of target problems and the effectiveness of the proposed method is evaluated by numerical experiments.

This paper presents a numerical method for sound field analysis based on fast multipole boundary element method. The impulse response analysis is usefully introduced to realize the efficient numerical simulation. The present method is applied to the sound field analysis around the sound insulation wall with complicated shape. The noise evaluation system is also developed for the use of immersive VR device and head mounted display. The system exposes to the users the computed noise level with both the auditory information using sound source signal and the visual information using CG image. In order to investigate the validity and efficiency of the method, the present method is applied to the construction noise problem and the numerical results are compared with the observed results in VR space.

Rigid plastic finite element method (RPFEM) is one of the versatile numerical tools to solve bearing capacity problems. A hybrid-type RPFEM has been developed by the authors, which directly searches both the stress and velocity field at the ultimate limit state. In this article, clay ground reinforcement with a sheet-like member is modeled and implemented into the RPFEM. The proposed model includes the effects of strength of sheet-like members and shear interaction between the members and surrounding soils. Formulation of the model is explained in details and results of numerical experiments are presented to demonstrate its performance.

In this article, rigid plastic finite element analysis for the bearing capacity of shallow foundations subject to inclined loading is investigated. A linear constraint condition on the velocity field is introduced to express a rigid motion of a foundation and a rough contact between a foundation and a ground. A series of three dimensional bearing capacity analyses of a shallow rectangular foundation subjected inclined loads is carried out to grasp the influences of foundation shape, the angle of load inclination and the loading direction in the horizontal plane on the bearing capacity characteristics. The numerical results show that the bearing capacity mode is observed when the load inclination angle is small, and the sliding failure mode appears when the load inclination angle is large. The influence of the loading direction in the horizontal plane is observed only when the cases of the bearing capacity mode of rectangular foundation. No influence is observed for the square foundation cases and sliding failure modes.

This article presents the formulation and implementation of thermal-shrinkage-induced-cracking to study a possible use of a LASER for demolishing RC structures. Convective and radiative cooling, coupled with deformation and cracking, are rigorously treated, and related modules are implemented into a higher-order particle-discretization finite element method (HO-PDS-FEM). Developed modules are verified comparing with analytical solutions, and validated by comparing with observations from a quenching experiment. Cutting of an RC block using a LASER beam is analyzed. The pattern of thermal-induced-cracking is compared with experimental observation. It is concluded that the pattern can be predicted satisfactorily, which suggests the developed numerical model can be used study cracking induced by thermal shocks.

Explicit time integrators are suitable for wave propagation phenomena. If they are applied to the transient analysis of nearly incompressible materials such as rubber-like materials, it is necessary to use a small time step size determined from the speed of wave propagation of the volumetric deformation component with high stiffness. In this paper, the Rattle method for the constrained Hamiltonian problem is extended to develop a novel splitting time integrator for nearly incompressible elasticity.

This study proposes a critical slip line search method using particle swarm optimization where the slip lines are discretized by quadratic isoparametric elements. In this method, it is possible to describe complex non-circular slip lines with fewer degrees of freedom because each part of a slip line divided by the elements is interpolated by the quadratic shape functions of the internal coordinate. It is shown that the method enables searching the critical slip line more efficiently compared with the ordinary method with slip lines in the shape of polylines.

We investigate the influence of the vibration models of the grid-type sleepers on the simulated dynamic response of a jointed railway track. The grid-type sleepers are modeled as either non-prismatic Euler beams or 3-D elastodynamic bodies, and are continuously supported by Winkler-type elastic foundation or an elastodynamic layer. Two rails are connected by a supported joint or a suspended joint. Through numerical tests, the dynamic response of the railpad force and the sleeper acceleration show a similar tendency independently of vibration model of grid type-sleeper. The peak values at the impact and their post-peak behaviour are influenced by a choice of vibration model of the grid-type sleepers.

The region-wise zig-zag theory, which combines the improved zig-zag theory and layer-wise theory, has been developed. In this theory, it is easy to apply to delamination analysis and to structures composed of composite materials and isotropic materials, and the unknown degree of freedom is considerably smaller than that of the layer-wise theory. In this study, it is proposed to use virtual laminas for this theory in order to improve the applicability and efficiency of this theory. The accuracy and applicability of the region-wise zig-zag theory using the virtual laminas have been verified by using numerical examples of bending analysis of isotropic and anisotropic laminated plates.

In this study, compaction is assumed as loading and unloading on unsaturated soil under water-undrained and air-drained conditions, and the compaction simulation is conducted with soil/water /air coupled finite element analysis. First, the static compaction on one-dimensional compression condition was simulated to verify the definition of compaction. Consequently, the shape of the compaction curve can be drawn on the coordinates of water content and dry density. Moreover, it was found that depth distribution of void ratio is not unique within compacted specimen according to water content. Next, roller compaction was simulated to investigate the effects of compaction construction used in geotechnical engineering site. Roller compaction generated heterogeneity of compaction degree within the ground due to rotation of principal stress. It was found that the quality of compaction is strongly dependent on the direction of roller compaction.

This paper investigates on the four-point flexural deformation and load bearing behaviors of partial helicoidally laminated CFRPs. Two partial helicoidal laminates are examined. The helicoidal laminate with limited fiber direction shows a brittle behavior. On the other hand, although the helicoidal laminate skin with cross ply core also shows a brittle behavior, the preceding wide skin damage leads to the mitigation of the brittle core fracture. In comparison with helicoidally laminated CFRP of full thickness and all fiber direction, the two partial helicoidal laminates show higher load capacity and initial stiffness.

The Material Point Method (MPM) is a kind of particle method, which is advantageous for large deformation analysis because it avoids mesh tangling as grid methods normally encounter. In multiphase geotechnical analysis, it is necessary to consider the advection between phases when the relative motion between solid and liquid phases is large, such as during large deformation of highly permeable ground. In this study, a new formulation and discretization of the solid-liquid-gas three-phase 2-Point MPM is proposed. In this study, we propose a new formulation and discretization of the solid-liquid-gas three-phase 2-Point MPM, which can take into account the advection between the solid and liquid phases. The formulation of the proposed method is a u-U formulation, and thus it is not necessary to use advection term between the solid and liquid phases by using Lagrangian particles for the two phases. Small and finite deformations of the one-phase and two-phase systems are verified through numerical analysis. Finally, dynamic liquefaction analysis of embankment in solid-liquid-gas three phases was conducted to confirm the applicability of the method to large deformation phenomena with significant relative motion between solid and liquid phases.

This is an extension work based on our previous one [J. Xiaoyu and T. Matsushima, J. Appl. Mech., JSCE, 2018]¹⁾ in which we mainly discussed the influence of intergranular friction and particle size distributions on granular shear flow of circular particles by comparing discrete element simulation with 𝜇(I)-rheology model. In this paper, we focus on the simple shear flow of elliptical particles under various inertial numbers I and investigated the influence of particle aspect ratio 𝛼 on the flow rheology and microstructural characteristics. It was found that (1) the 𝜇(I)-rheology model is well applicable for the flows of the elliptical particles of different aspect ratio 𝛼; (2) the deviation of macroscopic friction 𝜇(I) and solid fraction 𝜙(I) from the quasi-static state, i.e., 𝜇−𝜇₀ and 𝜙₀−𝜙, are linearly correlated, and it is independent of 𝛼; (3) the logarithm of particle orientation distribution log 𝑃(𝜃) is well described by a trigonometric function with a particle orientation anisotropy 𝑎_ρ; (4) the linear relation in (2) is explained by the microscopic observation in which 𝜇−𝜇₀ and 𝜙₀−𝜙 are uniquely related to 𝑎_ρ, respectively, showing that increasing randomness of particle orientation causes both increasing 𝜇 and decreasing 𝜙.

Effective notch stress method is one of the general methods to predict fatigue life from weld toe and root. However, there is a probability to give over the conservative results and there’s no description for angular distortion in the guideline. On the other hand, the effective stress concentration factor proposed in recent years, which is based on the stress range on the tensile side, shows a high correlation with fatigue property of some weld joint types. In this research, fatigue experiments and modified ENS method proposal are conducted for butt weld joint with backing plate. Regarding the stress range, the accuracy of the result using tensile stress range with angular distortion was the closest to the experiment results. And S-N diagrams are modified by the suggested equations for effective stress concentration factor at weld toe and root. It is found that the fatigue life can be evaluated uniformly with a high accuracy for four weld joints by proposed method.

This study aims at obtaining basic knowledge on melting and solidification behavior of a lunar soil simulant, FJS-1 by isotropic and surface heating. The effect of heating history, soil grading and initial packing density on the mechanical properties of the solidified materials was examined. It was found that (1) the higher temperature, the longer heating time, the finer material, the denser packing density deduce harder and stronger solid product, (2) heating at 1200 deg. results in very strong material as high performance concrete, (3) the material strength is uniquely determined by the density of the solid product, (4) a linear relation between the Vickers hardness and the uniaxial strength is obtained, (5) the surface heating causes a layered product whose physical properties differ in different layer, and (6) flow of surface soil due to the partial melting greatly affects the thermal conductivity and the volume of the solid product.

In order to establish a rational numerical analysis method for appropriately evaluating the impact resistant behavior of the steel posts for rockfall protection fences installed into the protection walls, 3D elasto-plastic impact response analyses were conducted taking impact velocity of the impacted weight as variables. The applicability of the proposed method was investigated by comparing with the experimental results in the case of the through-type steel posts. From this study, it is confirmed that the impact resistant behavior of the steel posts can be properly simulated by applying the proposed method.

3D-DEM code was parallelized by using Open MP to decrease computing time and to improve the quantitative accuracy of estimated residual settlement of poorly-graded crushed stone layers. Ballast grain models are also elaborate to sharpen the angularity by adapting “Asperity” elements. Then the track stiffness after ballast renewal and initial settlement of ballasted layer during 70 trains passing are simulated by using above mentioned DEM codes and models. The computing time is reduced to one tenth at the case with 32 threads of supercomputer. The residual settlement calculated by DEM simulation fit into the range of mean value plus-minus standard deviation of measured data from real railway line.

The volume change of concrete materials during drying is influenced not only by the characteris-tics and volume fraction of the paste but also by the quality of the aggregate. Various recycled mate-rials have been repurposed as concrete aggregates for waste valorization. However, conventional quality testing methods for aggregates may be insufficient for recycled materials. The effect of re-cycled materials can then be evaluated through how they behave inside the concrete. In this study, how drying affects the interior of cylindrical mortar specimens has been probed via X-ray CT image intensities and using the strains calculated by the digital volume correlation method. As a result, the relationships between the intensity reduction rate and the mass reduction rate were clarified. The re-lationship between the volumetric strain and the intensity reduction rate in the horizontal section was similar to the one between the mass reduction rate and the length change rate. It was shown that the vertical strain inside the specimen remains unchanged in the radial direction, while the area strain increases towards the surface of the sample.

In recent years, wind-borne missiles due to tornados have frequently damaged residential houses and human lives. Although steel plates have been used in the protective design and reinforcement of nuclear facilities. Details of failure behavior of steel plates were not sufficiently investigated. In this study, in order to investigate the failure behavior of steel plates subjected to a flat-nose projectile impact, impact experiments and numerical analyses were conducted for 6 mm, 9 mm and 12 mm-thick steel plates using a 6.0 kg mass projectile with flat shape tip (diameter of 50 mm) at velocities of 40-100 m/s. In addition, an energy-based design method was proposed based on the deformation and perforation failure mechanisms of the steel plate examined by the numerical analysis.

In the civil engineering field, the research on reinforcement and repair using FRP with excellent characteristics is actively carried out. In addition, recently, a study on the repair of damaged FRP with FRP bonding has also been carried out. In the adhesive bonding between FRPs, the fiber near the surface of FRP as a main member might break near the bonding end. Farther, the debonding on the adhesive layer is one of the concerns for FRPs joint. However, there is no theoretical study on the fracture of adhesive bonding between FRPs. In this study, the mechanical properties of fiber fracture and adhesive flaking, which are factors of fracture, were clarified from the theoretical stress analysis method. As a result, it was found that by reinforcing FRP with FRP, a larger strain acts on the fiber nearest the surface in the FRP members than that without reinforcement FRP members.

Recently investigation reports of collapse incident concerning of Sabo structure have been reported due to large-scale debris flow. Therefore, it is required the collapse mechanism immediately, and is simultaneously validated the numerical elucidation. This study examines an application of Distinct Element Method for a classification of failure mechanism which was reported in Nagiso sabo dam. First, collapse occurrence condition experimentally investigated by using experimental models. Second, in order to reproduce constitute law of member models, under experiment collapse tests are carried out. Then, those results are modelized to the bilinear elastic-plastic law by breakage. All experimental cases are simulated by DEM, and collapse as well as no-collapse results are reproduced by the proposed DEM. The breaking mechanism of the section is analyzed in detail by examining the connect springs, i.e., axial, bending, or share, in case of structural collapse occurrence results. The analysis results indicated that joint parts of steel under stress concentration had been broken by bending force.

The Applicability of the damage model based on the modified von-Mises criterion is studied for compressive fracture behavior of cement-based materials. First, crack propagation behavior of mortar plate with a circular hole subjected to compressive loading is simulated using the finite element method with the modified von-Mises criterion, and the results are compared to the existing experimental results. To verify the capability for simulating the compressive fracture behavior quantitatively, the numerical results are compared with our own test results. Finally, finite element models with different mesh size are used for investigating the influence of mesh size on the numerical results of compressive fracture behavior.

A surrogate model is proposed for obtaining load-displacement response instead of nonlinear finite element analysis. The model is based on sensitivity analysis of the nonlinear FEM for reinforced concrete beam using a damage model, and is applicable to Monte Carlo simulation for quantifying the uncertainty in the numerical analysis. The experimental results for invstigating the variation of the fracture behavior of RC beam are first presented. After verifying that the fracture behavior observed in the experiment can be simulated by the nonlinear FEM with the damage model, the surrogate model is developed based on the sensitivity analyses for material parameters required in the finite element analysis. Finally, we validate the proposed surrogate model for an alternative of nonlinear FEM, and demonstrate that the model enables us to perform the Monte Carlo simulation for uncertainty quantification with a low computational cost.

Flume experiments were carried out to understand the development of sand waves in response to the downstream water level and the characteristics of the longitudinal profile of the water level that may be affected by the sand waves. In addition, no-contact measurement of the riverbed below the water surface was carried out using a blue laser, and its usefulness in movable bed experiments was discussed. In the case of rise in water level, the sand waves developed in the entire channel section and the water level rose accordingly. In the case of low backwater, the riverbed waves developed and the water level rose in the same way, but there was a section downstream of the channel where the riverbed flattened without the formation of sand waves because the water level rose in the upstream section, thereby, contributing to a larger water surface gradient.

Experimental investigation on hydraulics in pipe flows connected to manhole yields that the formation of local flow in manhole is significant for the drainage faculty. The installation of a drop between inlet and outlet pipes connected to manhole and the pressure control in manhole were conducted in the connection with angle of 90, 120, 150, and 180 degrees between inflow and outlet pipes at manhole. Under a wide range of discharges, oulet pipe slopes, and pressure control in manhole, drainage function on pipe flow has been investigated experimentally. The discharge coefficient defined at the upstream end of inlet pipe has been characterized. Also, the discharge coefficient has a maximum 1.2 under the connection angle with 120 degree.

In this study, the characteristic flow structures that often observed in water accident sites were investigated. The flow structures were numerically examined by using precise under water geometries of water channels which were obtained by relatively new methods. The results show that the several features were observed in water accident sites, such as downward flows from a rapidsection to a pool, a high velocity shear flow and recirculating currents near downstream of a channel restriction. The results of frequency analysis also showed that those structures were characterized by channel geometries and mean flows.

In order to capture floating garbage, secondary flow generated by skew strip roughness on one side wall was used. The effects of interval length and skew angle on the strength of secondary flow were investigated and optimum arrangement was obtained. However, the lateral range of effective secondary flow was limited to a certain distance in this one-side arrangement of skew strip roughness. In the garbage capturing experiment, floats were transported to one side by secondary flow and high capture rate was established by using skew catching filter.

It has been observed that the bank erosion takes place not at the peak of discharge but during the decrease in discharge. We perform linear stability analysis for the formation of alternate bars including the effect of the variation of discharge. We assume that the variation of discharge is sufficiently slower than the variation of bed elevation, and introduce two different time scales: the time for the variation of bed elevation 𝑡 and that for the variation of discharge 𝑇. With the use of the multiple scale expansion method, we obtain the growth rate of perturbation as a function of 𝑇. It is found that the bed is more unstable when the discharge is smaller than the peak in the range of large wavenumbers in particular.

For a free jump in a horizontal rectangular channel, the relation between the jump length and the energy dissipation is explained with the physical meaning by using an energy equation applied to the jump zone. In the jumps below a sluice gate, the distributions of the velocity and those of the turbulence intensity are obtained experimentally for the undeveloped and fully developed inflow conditions under a given inflow Froude number and Reynolds number. The convection of the turbulent energy and the work performed by the Reynolds stresses can be calculated, showing that the jump length is interpreted as the length of the zone within which the energy dissipation is accomplished.

With the aim of reproducing the resonance phenomenon occurring in actual high speed railway PRC girders, this paper developed an analysis method that combines a nonlinear equivalent reduction model of PRC girders and a dynamic analysis method that utilizes the railway operation information big data. First, a nonlinear equivalent reduction model which takes into account the hysteresis characteristics after concrete cracking was developed. The validity was confirmed by comparing results of the reduction model with the actual measurement and the three-dimensional finite element model. Using this model, the mechanism of the resonance phenomenon was clarified, that stiffness of the PRC member gradually decreases by repeatedly experiencing simultaneous double-track loading when train passing. As a result of evaluating the trends of the dynamic response based on the train operation analysis, the stiffness sharply drops due to double-track loading after the fundamental timetable revision, and the displacement significantly increases since the resonance speed gets close to the operating speed. These simulated results are consistent to the actual observed phenomenon and measured displacement responses.

In this study, targeting a consolidation problem, we combine a data assimilation method based on time series observations with reliability analysis to update the conditional limit state probabilities. The settlement is predicted by using soil/water coupling FEM analysis, and its uncertainty is estimated by using a particle filter as a data assimilation method. Three types of hypothetical time hitories of settlement are prepared as observation data with different magnitudes of settlement based on forward analysis of the soil/water coupling FEM. The limit state probabilities at observed and unobserved locations are updated based on the increase of observation in time. It is shown that the predicted settlement and the conditional limit state probability are updated quantitatively according to the increase of observation data, which leads to resonable decision-making.

The spatial distribution of geological properties is often modeled by dividing it into a trend component and a random component. The random component is important for understanding the characteristics of local variability in the spatial distribution. Using Gaussian process regression with superposition of multiple stochastic fields, the trend and random components are separated for two types of 1D and 3D measured data in cone penetration tests, and the autocorrelation function is investigated for the random component. Various autocorrelation functions, such as Gaussian, Markovian, Binary noise, Whittle-Matérn, and Gaussian-Markovian arithmetic mean models, were compared using the information criterion AIC and BIC. As a result of model selection, Whittle-Matérn and the arithmetic mean model were selected as the autocorrelation function for random components. Examples of estimating the trend and random components at arbitrary locations using the selected models are presented.

In order to evaluate the failures inside the materials, Acoustic Emission (AE) techniques, which is attractive in the failed of the structural health monitoring and model tests, have been applied to the non-destructive tests (NDT), and it has been used to identify the elastic wave velocity distributions and the failures in recent. AE is emitted at the occurrence of the cracks and/or the frictions, and the elastic wave frequencies and velocity are related on the targets situation. Therefore, AE behaviors are potential to visualize the failures on real time if AE measurements are constantly conducted. However, NDT which based on AE techniques, requires the accurate arrival times of AE to clearly visualize the inside of the materials. To visualize it practically, the arrival times should be detected automatically since the amount of the AE measurement data is generally huge. In this paper, a method of AE arrival time detection is proposed based on the root mean square voltage in the noises. The proposed method is verified on the model tests in which pencil lead breaks conduct on the aluminum plates to detect the AE arrival times, and the accuracy of the arrival times investigated on the proposed method and the conventional method. The result of the investigation show that the proposed method detects more accurate arrival time in the low S/N signal, in comparison with the conventional method. The suggests that the proposed method improves the results of NDT which based on the arrival times of AE.

In this paper, the void volume and void shape of damaged concrete were evaluated by X-ray CT. In addition, compressive strength tests with the AE method were conducted. The results show that the AE energy generation behavior differs depending on the void structure. In the case of the specimens with small void surface area, multiple AE energy peaks were observed and the failure proceeded in stages. On the other hand, for specimens with a large surface area of voids, the AE energy increased linearly and the failure progressed rapidly. It is clear that there is a positive correlation between void volume, void surface area and AE energy in the strain range from 0 to 100×10^-6. These results suggest that void distribution can be evaluated by focusing on the AE energy at the initial stage of loading in the compressive stress field.

The maintenance of steel structures is an important research issue. Many riveted steel structures are still in service today. The detailed behavior of rivet joints has not been fully investigated. In this study, riveted test specimens were made and loading tests were carried out. In addition, these experiments were replicated through analytical studies. The results of this study show that the slope of the load/strain relationship for the riveted specimen is lower than that for the unriveted specimen. From the comparison with the analysis, it is clarified that residual stresses are generated during riveting.