In this paper, the deformation prediction caused by condition change with Finite Element Analysis is performed for SLM (Selective Laser Melting) modeling. Various methods for predicting thermal strain have been proposed. In this study, finite element analysis is applied to SLM for the purpose of investigating the features and advantages of full thermal analysis and inherent strain method respectively. We verified the modeling method in each method with a very small laminated thickness of several tens of micrometers.

When Film-cooling effectiveness of a slanted round hole in a jet-in-crossflow configuration was predicted by Reynolds-averaged Navier-Stokes (RANS) simulation with turbulence model parameters optimized based on measurement data. A computationally efficient approximate model of the flow field based on a surrogate model or a reduced-order model was employed to estimate the posterior density distribution of the RANS model parameters using the Markov-chain Monte Carlo method, which forms a probabilistic parameter estimation framework based on measurement data. In this paper, the influence of flow conditions such as density and blow ratios was then investigated to understand the effectiveness of estimated parameters in different flow conditions.

From the mechanical point of view, welding processes are modeled as a coupling problem between heat conduction and thermal elastic–plastic problems. Such a welding mechanics problem generally requires a large amount of computational time due to its nonlinearity as well as a lot of time steps with a moving heat source. To overcome this difficulty, we are developing a large-scale welding simulator based on domain decomposition method. Domain decomposition method is one of the large-scale parallel finite element methods, and has been used for various problems such as heat conduction and structural mechanics. In the presentation, the methodology of domain decomposition method that is applied to the welding problem is presented, followed by the algorithm. Then, a bead-on-plate problem, which is a popular benchmark problem in the field of computational welding mechanics, was analyzed by the developed simulator. The bead-on-plate problem was successfully analyzed within several iteration steps of Newton–Raphson method and conjugate gradient method. Key Words : Computational welding mechanics, Thermal elastic–plastic analysis, Large-scale analysis, Parallel computing, Domain decomposition method

Diamond-like-carbon (DLC) is well known for their low friction coefficient, wear resistance and good corrosion resistance. Especially in the field of tribology, DLC attracts much attention as a promising material for friction reduction in sliding parts. Experimental studies report that the hydrogenation of DLC reduces friction and improves its tribological ability. However, the friction mechanism of DLC has not been sufficiently elucidated because it is difficult to observe the surface directly during friction. Therefore, a numerical analysis approach that can directly capture the behavior of atoms is expected. In this study, we developed an interatomic potential based on Takamoto potential which is applicable for DLC-SiC sliding simulation. The molecular dynamics simulation of DLC-SiC sliding was carried out using the developed interatomic potential. The result reproduced the reduction of the friction by the addition of hydrogen. After that, the atomistic mechanism of the friction reduction effect due to hydrogen was discussed.

In this study, we develop a non-parametric shape optimization method for designing composite structures composed of dissimilar materials in buckling problem. The first buckling eigenvalue is used as the objective functional, and is maximized under volume constraint. We construct the shape design optimization system and optimize the interface shape of the two different materials according to the developed optimization method. The optimal results show that the objective functional can be enhanced significantly.

Power modules are used to control electric power, therefore it is a key product of energy saving. A structural reliability problem of wire bonding in power modules becomes important at high temperature operation. It is possible to relate the thermal fatigue strength of wire bonding to fracture mechanics parameters such as K, J, T*, etc. In the operation of power modules, wires are subjected not only to plastic deformation but also to thermal and creep deformation, so that the fracture mechanics parameter T* should be used instead of J. We employ a near-field value of J -integral as an alternative of T*integral, and we define damage parameter ΔT* based on T*-integral. In this paper, we propose a simple calculation method of ΔT*-integral that can be calculated by using commercial finite element computer code such as Marc, Ansys, etc.

By the in situ observations for solidification of carbon steel, δ-γ massive-like transformation was observed when the peritectic transformation was expected to occur. This transformation is a solid-solid transformation between δ phase and γ phase and thus the strain energy would not be relieved by the liquid phase. This is why δ-γ massive-like transformation is said to be one of the reasons behind surface cracking. It has already been revealed by the previous study that this transformation is dominated by nucleations of γ phase out of δ matrix phase, and that there are many variants that differ crystal orientation. In this study, using phase field method that take intra-nuclear variants into account, what kind of effect that intra-nuclear variants give to the nucleation. Presence of variants accompanies interfaces between each other. At the interfaces, volume expands because of deviation from perfect atomic arrangement near interfaces. In this study, γ nucleus with intra-nuclear variants showed smaller critical radius of a nucleus than γ nucleus without intra-nuclear variants. This is realized by reduced strain energy. By the existence of intra-nuclear variants, it became easy to nucleate.

Recently, Peridynamics proposed by Stewart Silling in 2000 is studied by many researchers, because it is able to treat the discontinuity of crack surface for failure or crack propagation phenomenon without special technique such as Fracture Mechanics. On the other hand, it has been thought that it is diffucult to implement into general purpose CAE software because there is no affinity for boundary condition or contact condition with conventional Finite Element Method.In this paper, I report some validation case with general purpose CAE software named LS-DYNA which is developed by LSTC that has become possible to divert a boundary condition or contact condition of Finite Element Method with discontinous galerkin method.

In this study, compression simulations are performed on Zr - Cu - Ni - Al based metallic glass model with different composition ratios by using molecular dynamics method. As a result, localization process of deformation is different due to internal structure in models even with the same composition ratio. In order to examine the origin of local deformation occurrence and growth, we evaluated the atomic elastic stiffness coefficient and atomic deviation strain during the compression simulation. It is revealed that determinant of the atomic elastic stiffness coefficient changes depending on a drastic increase of the atomic deviation strain.

A stress relaxation model for industrial rubber materials was developed. This model is a Kelvin–Voigt type viscoelastic model combining a single spring and a fractional differential viscoelastic model. In order to confirm the effectiveness of the proposed model, various stress relaxation tests of three types of SBR-CB with different volume fractions were carried out, then the proposed method was applied to them. As a result, the stress relaxation history predicted by the proposed method using the material constant identified between from 1 to 2 minutes experimental data shows good agreement with experimental ones over 3 hours. The equilibrium elastic stress-strain relationship was predicted from the multi-relaxation test by using the proposed model.

We propose a computational method of granular materials by applying the Renormalization Molecular Dynamics (RMD). In CMD 2017, we applied RMD to a screw feeder simulation, and reported that the transportation amount and other properties were agreed with theoretical values. However, in order to accurately predict the transport amount, a diameter of constituent particles of the screw must be sufficiently smaller than a diameter of powder particles, which causes extremely poor calculation efficiency. Therefore, in this paper we introduce the signed distance function (SDF) as a method to handle a rigid body model with the high shape accuracy and low calculation cost. Computational experiments of the screw feeder are carried out and it is confirmed that the same conveyance amount can be obtained by the SDF with the shorter computation time compared to the screw composed of the particles.

It has been reported so far there is size dependence of mechanical properties in metals at the microscale. However, conventional elasto-plasticity theories do not represent any size dependence predicted from dislocation theory. In this study, we analyzed three dimensional nonuniform plastic deformation problem using a phenomenological higher-order strain gradient plasticity theory in which a yield function is considered as an additional governing partial differential equation. Specifically, we compared solutions of conventional elasto-plasticity theory and higher-order strain gradient plasticity theory for thin plate uniaxial tension. It is confirmed that deformation behavior that does not depend on number of element divisions is predicted by the higher-order strain gradient plasticity theory. Besides, a new algorithm of the higher-order strain gradient plasticity computations in which violation of the yield condition is automatically corrected without iterative calculation is proposed.

In this study, we investigated origins of Bauschinger effect in polycrystalline metals using a finite element method incorporating a higher-order gradient crystal plasticity theory in which it is possible to impose microscopic (extra) boundary conditions. In a boundary value problem of polycrystalline metals, we focus on two types of material defect, i.e. grain boundaries and heterogeneous dislocation structure, as origins of Bauschinger effect. When grain boundaries are assumed to be hard interfaces being impenetrable for dislocations without taking account of heterogeneous dislocation structure, a strong Bauschinger effect appears with refinement of the grain size. When grain boundaries are assumed to have the same property as that of the bulk part of the material, only a weak Bauschinger effect is observed. On the other hand, with a heterogeneous dislocation structure, the Bauschinger effect appears without assuming any special property for grain boundaries and dependency of the grain size on the amount of Bauschinger effect is small. The effect of heterogeneous dislocation structure cannot be neglected to account for the Bauschinger effect, and the grain boundaries may not play a role of strong barriers to dislocation movement.

It is indispensable to use simulation of computer science for industrialization of origami structure. Origami structure is more complicated than other structures, that is why we have difficulty in model creation. However, we can obtain mesh relatively easily by mesh generation function provided in finite element method softwares which should be called origami modeling method. Furthermore, we show that a structure that changes the plurality of natural frequencies considerably as expected in the system can be obtained. By using this function, we will discuss the ideal transport box design method unique to the origami structure, which solves the problem of low survival rate after transport, such as blood and iPS cells.

A multi-phase field (MPF) method coupled with finite interface dissipation proposed by Steinbach et al. is applied to simulate the dendritic solidification affected by molten metal flow in a stainless steel composition system. Thermodynamic calculation using CALPHAD database is replaced by machine learning prediction procedure in this MPF program to reduce computational cost. Solidification calculations are performed in a case of quinary system, Fe–Cr–Mo– Ni–C. Molten metal flow is solved by an incompressible computational fluid dynamics method coupled with MPF equation in the same finite difference grid points. It is confirmed that calculation results are qualitatively reasonable.

The phase-field crystal (PFC) model has been used to simulate microstructure evolutions on atomistic length scales and diffusive time scales, that is an atomistic simulation method alternative to a traditional molecular dynamics. This paper presents a large-scale three-dimensional PFC (3D PFC) simulation of polycrystalline grain growth in a body-centered cubic iron. In order to reduce the high computational cost of 3D PFC simulation, we develop the C languagebased code for multiple-graphics processing unit (GPU) computing of the PFC simulation using OpenACC and OpenMPI libraries. The large-scale 3D PFC simulations of the grain growth were performed on the GPU supercomputer Reedbush. The simulation results show the PFC model captures the grain coarsening driven by the grain boundary curvature. We investigate the temporal variation of the average grain size during the grain growth and evaluate the grain growth exponent. The results show that the grain growth exponent in the early stage of the grain growth deviates from that of the parabolic law.

In this research, we report fatigue fracture test of SM400A plate specimen and crack propagation analysis using VCCM(Virtual Crack Closure-Integral Method). In fatigue fracture test, crack shape is measured by beach mark method to find the crack growth rate and identify the Paris law constants. Fatigue crack propagation analysis on surface crack was performed using the identified Paris law constants. The feature of this research is that by acquiring the crack growth data not only in the depth direction of the surface crack but also in the multidirectional with different angles. It is possible to acquire more data on the crack growth rate from one specimen. From the results of experiments and analyses, we investigate the variation of crack growth rate data and the difference between experimental results and crack propagation analysis.

In this research, a method to carry out crack propagation analysis using the MDP-FEM. In this method, physical quantities such as stresses, plastic strains, equivalent plastic strain are stored by a group of points that are independent from the finite element mesh (Mesh-independent Data Points, MDP). The physical quantities are mapped to the gauss points for the integration to establish the element stiffness matrix. Since the strain history dependent quantities such as stresses are stored by MDPs, the finite element mesh is moved relative to the positions of the MDPs to represent the crack propagation. Thus, the amount of crack propagation in one crack propagation can arbitrary set by the analyst. In this paper, large deformation elastic-plastic crack propagation problem in a large strip emulating a stable crack propagation is presented. The results of the MDP-FEM compare favorably with those of an ordinary finite element method with the nodal release technique.

The elastoplastic constitutive equation with the damage is formulated incorporating the extended subloading surface model and, based on it, the constitutive equation of concretes is formulated in this article.

In this study, deformation analysis by a molecular dynamics method (MD) of a composite model of CNT (carbon nanotube) and polymer was conducted, and the deformation behavior from the atomic level of a composite material using a carbon material as a filling material was investigated.First, modeling of epoxy resin commonly used as a base material will be studied.The epoxy resin has an internal structure called a crosslinked structure by a crosslinking reaction.Since the crosslinked structure affects the properties of the epoxy resin, deformation simulation was performed on epoxy models having different crosslink densities.It was suggested that the mechanical properties change depending on the difference in crosslinking structure.Next, a composite model in which CNTs were introduced into the epoxy model was created and deformation simulation was carried out.

In order to discuss change in deformation responses due to contained elements in silica-based glasses, we have performed molecular dynamics simulations on silica glass models contained barium oxide actually used for optical lenses. Four glass models with different contents of barium oxide are prepared by melt-quench simulation. Then, change in atomic structures, such as tetrahedral structure of SiO₄, is discussed in tensile simulation of those models. As a result, it is suggested that local deformations occur mainly around barium atoms because tetrahedral structures are maintain in models include barium during the tensile simulation.

High speed material tensile and compressive tests were carried out on ABS, and material parameters of a dynamic constitutive model for polymer were obtained in consideration of the hydrostatic pressure dependency and the strain rate dependency. Numerical simulations of the test piece subjected a tensile and compressive dynamic loading were carried out on ABS in consideration of SD effect. The validity of the constitutive model and material parameters was examined through comparison with the test results.

The supersonic flow through a Laval nozzle is studied using non-equilibrium molecular dynamics. A barostat and thermostat heatbath, a nozzle wall with the signed distance function, and axisymmetric boundary condition for reducing the computation are implemented to simulate the nozzle flow. The flow yields temperature, pressure and Mach number distributions, which agree well with predictions of the theory of isentropic expansion. Simulation under the condition such that the flow reaches the gas-liquid coexisting phase, the formation and growth of particle clusters through the nozzle are observed.

Recently solidification process of additive manufacturing (AM) is simulated by using various methods. But most of them requires temperature distribution as an input. In this study, heat transfer analysis is performed to simulate the temperature distribution for direct energy deposition (DED), a kind of AM process for metal. In DED process, powder is supplied at the same time as laser scan. Applying a field variable to switch the state of before or after deposition, we successfully simulate the powder supply. Temperature distributions obtained in this study shows good relationship with experimental results for all three cases.

The additive manufacturing technology is currently used for a wide range of processes in the manufacturing industry. The direct energy deposition is a kind of additive manufacturing method which is often applied to produce large blank shapes. In this study, solidification microstructures of titanium alloys in direct energy deposition are calculated by multi-phase field method. Temperature distributions used in multi-phase field method obtained by thermal analyses using finite element method. The results are summarized in a solidification map for direct energy deposition process conditions.

In this paper, a coarse grained particle model with heat transfer was developed. Among multiphase flows, the fluidized bed is an important topic of engineering due to its wide range of application including environmental and chemical processing plants. When including the effect of heat transfer in the model, it is important to simulate the flow behavior of gas and powder together with their temperature changes. A DEM-CFD method together with the developed coarse grained particle model for heat transfer was adopted to simulate the cooling behavior of the fluidized bed with hot particles. The simulated average temperature of the particles was compared with experimental results as well as simulated results of the reference. It was found that the simulation results for models with and without the coarse grained particle model to be similar to those of the results of the reference model. From the simulation results, it was found that this coarse grained particle model is an effective method to simulate the heat transfer behavior in fluidized beds.

To clarify the mechanisms of swallowing, we have developed swallowing simulator “Swallow Vision.” The organs and food bolus were modeled as hyperelastic material and fluid, respectively, and their coupling analysis was performed using moving particle simulation method. In the previous Swallow Vision, swallowing motion was simply modeled using forced displacements of some organ particles, which means that even soft tissues made of muscles cannot contract. In the improved Swallow Vision, soft tissues can contract just as the actual muscle motion instead of forced displacements. In this study, particles of the pharyngeal constrictor muscles were assigned to have each anatomical direction as same as the actual muscle fibers. Then, the contraction stress was applied according to the fiber direction, length and activation level. The activation level was determined empirically to obtain the similar motion to the medical image during swallowing. As a result, with improved Swallow Vision, the contraction movement of the pharynx was reproduced better than the previous one. The muscle-driven swallowing simulator is expected to enable more precise elucidation of swallowing mechanism.

In recent years, we have expected CFRP to reduce the weight of transportation equipment. However, computational reproductions of mechanical properties of thermosetting polymers, which are matrix of CFRP, have not been sufficiently conducted at present. In previous researches, the authors showed that the molecular chain plasticity model developed for thermoplastic polymers can be applied to thermosetting polymers and it reproduced tensile loading characteristics and nonlinear strain recovery during unloading for thermosetting polymers. However, these analyses assumed homogeneous materials, so that materials with heterogeneity have not been considered. Therefore, in this report, the authors investigate effects of the initial heterogeneity of thermosetting polymers on mechanical properties using the molecular chain plasticity model.

In order to investigate the deformation behavior in the surface processing of silica glass from a microscopic viewpoint, grinding simulations are performed on the surface of several silica glass models with different thickness and height by molecular dynamics method. As a result, it is observed that changes in the atomic structure at impacted regions of indenter and an increase in Si atoms at which the coordination number became 5. The deformation behavior is different due to board thickness dimensions and height dimensions. Non-invertible deformation exceeding the elastic limit occurs more frequently in a smaller model.

In a railway, one of the brake systems, called “tread brake”, is often applied to slow down the speed of the train by the brake shoe. Temperature of the wheel tread rises due to the slip friction between the brake shoe and the wheel tread at the time of braking, thus resulting in the thermal stress, and at the meanwhile the decrease of the yield stress of the wheel. On the other hand, a high pressure of over several hundred MPa occurs on the very small contact area between wheel and rail, which leads to the local deformation on the wheel tread. To investigate such complex behaviors, we developed a coupling analysis method which can consider simultaneously both the heat transfer and dynamic rolling contact between the wheel and rail. In this paper, we report the detail of heat-structure coupling analysis method and the test simulation results.

This paper presents the result of system identification of JAXA Jet Flying Test Bed ”Hisho“ using vibration data during taxi. In the taxi vibration test, the aircraft was towed by a tractor on the taxiway and apron at Nagoya Airport and operational accelerations were measured using 48 accelerometers mounted on the main wings. The proposed method which identifies a non-proportionally damped system as a generalized mode model using output data only was applied to the taxi vibration data, and undamped natural frequencies, a non-proportional modal damping matrix, and real normal modes of the main wings were extracted.

To understand the influence of solute yttrium on the deformation process in magnesium, we analyzed the cross-slip mechanism in which an <a> screw dislocation migrates from the basal plane to the prismatic plane using transition-state searching method based on interatomic potential. The result suggests that the attractive interaction between the partial dislocation and yttrium atom can assist (or inhibit) the constriction of the extended dislocation to the compact form when yttrium is dissolved inside (or outside) the extended dislocation. In addition, we found that the local attractive interaction between the yttrium atom and kink plays a significant role on the formation and migration of kink pairs, depending on the positional relationship between the partial dislocation and yttrium atom.

Soldering is used for bonding a print circuit board (PCB) and a semiconductor chip. It is known that Sn, which is the base metal of Pb-free solder, shows remarkable crystal anisotropy. To clarify the effect of anisotropy of Sn on strain distribution is important for lifetime evaluation. The strain distribution in a micro specimen under a tensile test was measured by a digital image correlation method (DICM) with a microscope. The strain distribution was also analyzed by the crystal plasticity finite element analysis (CPFEA) considering the critical resolves shear stress (CRSS) of each slip system. The deformation of the crystal structure of β-Sn depends on the size, number and orientation of crystal grains. The CRSS was noticeably different for each slip system, and the yield stress varied with the orientations of crystal grains. Although the CPFEA without considering strain hardening was effective for predicting deformation within crystal grains, it is necessary to consider the strain hardening of crystals to predict the stress-strain curve of a micro-specimen.

Power module is sealed with molding resin to protect from the environment. A large residual stress is generated along the interface between the substrate and the resin due to the difference of the Young's modulus and the coefficient of linear thermal expansion. The delamination between the substrate and molding resin due to the thermal residual stress is one of the serious reliability problems. The fracture toughness of interface cracks between a copper substrate and three types of molding resin was measured under various temperatures. Furthermore, we analyzed the stress intensity factors of an assumed crack on an interface between the substrate and molding resin in a test package, and evaluated the possibility of the fracture during thermal cycle tests. We compared the thermal cycle tests and mechanical fatigue tests of cracked specimens, and investigated the possibility to predict the life of thermal cycle test from the mechanical fatigue test.

We have proposed the data assimilation (DA) methodology based on the ensemble Kalman filter (EnKF) to estimate the unmeasurable state and the unknown parameters used in multi-phase-field (MPF) simulations. In this study, we apply the EnKF-based DA method to estimate the anisotropic grain boundary properties of Coincidence Site Lattice (CSL) grain boundaries and the driving force for the growth of recrystallized nucleus assumed in the three-dimensional MPF simulation of the static recrystallization. In order to validate the DA method, we perform the numerical experiments called twin experiments where the unknown parameters are estimated based on synthetic observational data. The results of the twin experiments demonstrate that the proposed DA method can estimate the ∑7 CSL grain boundary mobility peak, grain boundary energy cusp and the distribution of the stored energy in the deformed grains.

Sheet metal forming simulation based on the finite element method has been used to prevent the forming failure. The prediction accuracy of the sheet metal forming simulation is strongly influenced by material models, those parameters and the work hardening law used in the simulation. In the conventional parameter identification, the stress-strain curve after the necking has not been considered. However, in order to predict the large deformation behavior of the sheet metals accurately, we need to calibrate the material models and the hardening laws which reproduce the large deformation of the materials. In this study, we propose a data assimilation methodology (DA) to estimate the parameters of material models which enables us to predict the post-necking deformation behavior of the materials. This article presents the implementation of the ensemble Kalman filter to the elastoplastic finite element analysis. In order to validate the proposed DA method, we perform the numerical experiments where the parameter of the Swift hardening law is estimated.

In order to prevent the mechanical failure of lithium ion batteries (LIBs) during the discharging and charging processes, it is essential to understand the stress evolution induced by the Li diffusion in LIBs. However, it is difficult to observe the Li diffusion behavior and the stress evolution in LIBs experimentally. In this study, the stress evolution induced by the Li diffusion in a composite positive electrode of the LIB, which consists of LiCoO₂ (LCO) particles and an electrolyte, has been studied using the phase-field (PF) method. We investigate the effects of current density and the distribution of LCO particles on the Li diffusion and the stress evolution during the discharging and charging processes. The discharging curves are evaluated for various current densities and distributions of LCO particles. The simulation results show that the lithium ion concentration in the electrolyte surrounded by LCO particles decreases rapidly, which leads to the performance degradation of the lithium ion battery. Moreover, the results demonstrate that the potential of discharging curve drops significantly with the increase of the current density.

This report describes a numerical analysis of energy flow in waves using Statistical Energy Analysis (SEA) and FEM for two panels connected in an L shape with an acoustic black hole and a viscoelastic damping layer. The acoustic black hole proposed by Kryrov includes sharply tapered edge followed by high power functions of the propagation distance. On the surface of the black hole, there exists a thin viscoelastic material. Due to the black hole, flexural waves cannot reflect theoretically at the tapered edge. FEM and Modal Strain Energy method are applied to identify the internal loss factors to account for damping coupling between substructures with the black hole. Influences of the black hole on coupling loss factors which are related with the energy flow in waves are clarified.

Voids in a micro interconnect line can migrate and grow by electromigration (EM). Because the voids cause the disconnection of the interconnect line and lead to the electronic circuit failure, the prediction of the void migration caused by EM is important to improve the reliability of the interconnect line. Therefore, this study proposes a new multi-phase-field (MPF) model to simulate the void migration by EM in a polycrystalline interconnect. In this article, we present the use of the moving simulation domain method which enables numerical simulations of the void migration in a long interconnect line with a low computational resource. The simulation results demonstrate that the MPF model can simulate the migration and deformation of a void in a three-dimensional bamboo line. It is also revealed that the void migration is accelerated and decelerated when the void gets close to the grain boundary because the grain boundary attracts the void.

Glass is one of the prevalent materials used in modern architecture. Because of the brittle nature of glass, glass products tend to fracture under relatively small load, and hence it is of special importance to evaluate mechanical strength and safety of glass. Although such properties have been evaluated experimentally, quantitative evaluation by numerical analysis is also desired in various conditions that may be difficult to achieve in experiments, which would help to develop new glass components and architectural structures. Recently, Peridynamics theory (PD), a nonlocal continuum theory that describes a continuum by particle models, has been attracting attention so as to material failures since PD can compute deformation of materials that involve discontinuity. Such computation is very challenging within the framework of FEM due to its formalism based on partial differential equations. However, PD is a relatively new theory, and thus careful verification of accuracy of PD in computation of fracture phenomena is needed. In this paper, we applied PD to simulate the so-called Double Cleavage Drilled Compression (DCDC) method, which is an experimental method to investigate relation between fracture toughness and crack propagation speed. Comparing the numerical results with corresponding experiments on crack growth, we examined the accuracy of PD theory to simulate glass fracture. Numerical simulations for the DCDC method in 2D were performed by Peridigm, open source software of PD. Ordinary state-based PD, which allows us to simulate crack propagation, was employed. Our simulation results show that the crack became longer with increasing loading pressure and eventually stopped when the energy reaches the equilibrium condition. Because the result is consistent with experimental observations, we conclude that PD is applicable to investigate glass fracture.

Iterative methods for linear equations with coefficients of complex symmetric matrices appearing in finite element analysis of the high-frequency electromagnetic field have poor convergence. Besides, as the target of analysis becomes more extensive and more complicated, the convergence further deteriorates. Multiple-precision arithmetic operations such as pseudo-quadruple precision using two double precision and arbitrary precision arithmetic operation arbitrarily specifying accuracy are useful for improvement of convergence. In this research, we apply a multiple-precision arithmetic operation to an iterative method for a complex symmetric matrix, with a matrix representing the whole-body cavity resonator model of TEAM workshop problem 29 and succeed in improving convergence and calculation time.

An iterative domain decomposition method is proposed for numerical analysis of 3-Dimensional (3D) linear magnetostatic problems taking the magnetic vector potential as an unknown function. The iterative domain decomposition method is combined with the Preconditioned Conjugate Gradient (PCG) procedure and the Hierarchical Domain Decomposition Method (HDDM). Our previously employed preconditioner was the Neumann-Neumann preconditioner. Numerical results showed that the method was only effective for smaller problems. In this paper, we consider its improvement with a variant of balancing domain decomposition preconditioner (BDD-DIAG). Specially, we consider the multi-part processing without explicit construction of Schur complement matrix though only one part processing has been tested in CMD 2017.

Crystalline polymers may have some size effects that show different deformation responses depending on size of crystalline phase with respect to that of specimen. However, such effects have never been observed experimentally. In this report, we prepare a single crystal solution of polypropylene whose crystal size is adjusted by changing cooling rates. Then, the solution is mixed with amorphous polypropylene to obtain test pieces with different crystal sizes. Furthermore, by conducting a tensile test on the prepared polypropylene test piece, the effect of the difference in crystal size on deformation response is investigated. In addition, we examine the validity of the material model considering the size dependence attributed to dislocation existence in crystalline phase by comparing the numerical results with the experimental ones.

Because the light-water plants keep in appropriate working, keeping integrity of the structure should be confirmed. When bending and torsional forces simultaneously apply to piping due to an earthquake, which should be considered for an assessment of the natural disaster. Reasonable and appropriate estimation of the integrity should be improved by the latest knowledge. Therefore, it is necessary to develop a procedure of crack propagation estimation. Crack propagation is governed by the several condition, however two physical quantities are the major parameters, which are stress triaxiality and equivalent plastic strain. In the previous studies, a crack propagation condition is described by combining stress triaxiality and equivalent plastic strain under the monotonic test. Sum of two parameters becomes constant. A validity of the sum of the parameters was proved by application phase FE analysis in monotonic loading. The objective of this research is to validate the proposed crack propagation condition under the cyclic loading. A SGV410 CT specimen is analyzed in condition of low cycle fatigue by FEM. In this paper, test trial to simulate low cycle fatigue in consideration of contact to crack surfaces is presented.

Plastic 3D printer is the new technology which is able to create more free shape and a mold is not needed. There are no such characteristics in conventional methods, so plastic 3D printer is attracting from various viewpoints. Then, it will be progress to Rapid Manufacturing (RM) from Rapid prototyping (RP). For using as manufacturing products, it’s needed to improve modeling accuracy by reduce of the warpage, to evaluate the stiffness of the product. So, we studied the process simulation of plastic 3D printer that calculates warpage and residual stress and the optimization of shape. We are approaching by two way, inherent strain method and thermo-mechanical analysis. We report about these case study and the comparison of these calculation.

Glass is a brittle material that tends to fracture easily if excessively loaded. In indentation experiments, glass materials show irreversible deformations, for instance, fracture, plastic flow, and densification. For fracture simulation, numerical methods on the basis of continuum mechanics including finite element analysis may require some special techniques to avoid discontinuous deformation fields. By contrast, Peridynamic theory, a nonlocal extension of continuum mechanics, copes with cracks and fragmentations without difficulty. Peridynamic simulation can be implemented by Peridigm, an open software package, together with a variety of built-in material models. Taking advantage of the functionality of Perdgim, we aim to reproduce crack and plastic flow in indentation simulations for glasses.

Hydrogen diffusion / Elastoplastic coupling analysis in a commercial finite element method (FEM) software Abaqus was developed for fatigue crack growth (FCG) tests for low-carbon steel JIS-SM490B in gaseous hydrogen. The FE analyses were performed under various combination of hydrogen gas pressure ranging from 0.1 to 90 MPa and frequency from 0.001 to 10 Hz. Based on the experimental and numerical results, two evaluation methods were proposed: one is for prediction of hydrogen-enhanced FCG acceleration onsets by using gradient of hydrogen concentration G obtained by FE analysis and critical gradient G_c near a crack tip, the other is for prediction of hydrogen-enhanced FCG acceleration ratio by using original parameter “effective diffusion depth X_h” based on the gradient of hydrogen concentration. Then, it was confirmed that the evaluation methods were effective of hydrogen-enhanced FCG acceleration onsets and ratio, respectively.

Cartesian grid method is advantageous for aerodynamic design of aircraft because of its fast, automatic and robust grid generation. To improve the reproducibility of the wall face around geometric features, a new Cartesian grid method which implicitly includes geometric features without refining the grid size is proposed. The key idea of the proposed method is to modify the cell around the body by adding body surface coordinates information. To compute the body surface information,rays from the cells to the body surface are used. The accuracy of the proposed method is demonstrated by grid generation and flow simulation around a wing with a flat plat.