This paper presents a newly developed numerical model to simulate interfacial multi-phase flows with high density ratio such as the water and air mixing flows in sanitary wares. In order to adequately resolve thin water films on the curved surfaces of the sanitary ware products, the VSIAM3 (Volume-integrated average and Surface-integrated average Multi-Moment) and STAA (Surface Tracking by Artificial Anti-diffusion) methods have been extended to arbitrary unstructured grids, which enables the numerical model to simulate interfacial multiphase flows in complex geometric configurations with significantly improved accuracy. Moreover, the algorithmic simplicity of the numerical framework eases the implementation of high performance computing using Graphic Processor Unit (GPU) acceleration technique and Message Passing Interface (MPI) parallel computing. Satisfactory results have been achieved in the benchmark verifications and the real-case validation of a sanitary ware product.

View full abstract

Mindlin plate analysis is carried out by GLS (Galerkin/Least Square) FEM (Finite Element Method) in order to avoid shear locking. Variational principle including transverse shear strain is formulated based on Hu-Washizu and Hellinger-Reissner’s principle. The GLS term is obtained by the square of Eluer’s equation of this functional. The same interpolation function can be used for nodal valuable z-displacement w, rotational angle β_a and Lagrange multiplier (transverse shear stress) λ_a3. Bending of circular plate is analyzed by this method using P1 and Q1 element. It is found that GLS is effective for avoiding shear locking of Mindlin model in case plate thickness is thin.

View full abstract

The inverse analysis procedure using the modal parameter of 1st vibration mode as observation data was proposed to put the damage detection technique to practical use. Number of components of the state and observation vector must be equal to make up sensitivity matrix as square matrix. Therefore, natural frequency and ratio for displacement of each stories based on 1st vibration mode shape were specifically. The projection filter which was the simplest form was used as inverse analysis method. In this paper, characteristics of the inverse analysis method proposed are made clear through some numerical calculation from viewpoints of change caused by iterative calculations of determinant and singular value for the sensitivity matrix.

View full abstract

Improvements in bending deformation accuracy of three-node triangular element is required in the FEM analysis of natural fiber reinforced carbon ceramic composite materials. Therefore, we propose a three-node triangular interface element with drilling degrees of freedoms. The proposed element was formulated by applying the Heaviside function to the interface node of a triangular element with drilling degrees of freedoms. We call this element a “triangular interface element” because the stiffness matrix is composed of a triangular element and an interface element. To validate our proposed method, we conducted crack propagation analysis using a natural fiber reinforced carbon ceramic composite of bagasse ash and carbonized bagasse fiber. In this numerical example, we changed the interface adhesion and dispersion of the carbonized bagasse fibers. The maximum principal stress, flexural strength and crack propagation behavior of the composite material were accurately and validly analyzed by the proposed element.

View full abstract

A phase-field model describes the gas-liquid interface of two-phase flows. We solve the conservative Allen-Cahn equation combined with the continuum equation on a two-dimensional computational domain with a given velocity field. The accuracy of numerical result strongly depends on the mesh resolution around the interface. The AMR (Adaptive Mesh Refinement) method greatly reduces the computational cost, since it is possible to assign high-resolution mesh to the region around the moving interface. We have developed a code to solve the equation in a manner of the tree-based AMR with multi-moment methods, the conservative IDO and CIP-CSL schemes. To reduce the implementation difficulties of AMR method, we introduce the fractional step method and the directional splitting method. In a benchmark test of the single vortex problem, the AMR computation with 5-level refinement for the interface achieves 9.26-times speed up and 1/12.3 mesh reduction to compare with the computation on a uniform mesh.

View full abstract

Plasma-Shot Treatment is a new technology for generating functional interface by moving electrode material to workpiece surface using micro pulsed discharge. It is possible to form the hard titanium carbide (TiC) layer of about 10 μm thickness which has high wear resistance when the electrode is made from the semi-sintered powder of TiC. In order to understand this Plasma-Shot phenomena, simulation by a computing technology is required because it is difficult to directly observe or measure the phenomena occurring in a micro gap in a short time. In this study, numerical simulation technology for the phenomena was developed using the smoothed particle hydrodynamics (SPH) method that is superior in dynamic expression of the large deformation and boundary movement. Since the Plasma-Shot phenomena have been visualized and relatively good agreement is found between the analysis result and the experiment one, this simulation technique is effective for understanding the phenomena and elucidating its mechanism.

View full abstract

This study focuses on the convergence of the iterative methods for solving the complex symmetric systems arising from the electromagnetic field analysis. The COCG method has been widely used to solve such systems, however, it suffers from slow convergence rate. To improve convergence rate, multiple-precision arithmetic techniques are adopted. Especially, the double-double precision operation is investigated in complex number. The double-double precision complex number and its arithmetic operations are implemented using error-free transformation based on Knuth’s and Dekker’s algorithm. Moreover, a mixed-precision COCG method is proposed to save computational cost. Using a developed system and the QD library of commonly used software, some numerical experiments are demonstrated and its performance is evaluated.

View full abstract

This paper proposes a new method for visualizing 3D crack propagation in reinforced concrete by means of 3D-printing. Cracks in reinforced concrete are formed from not only external surfaces but also steel surfaces inside the concrete, and propagate intricately in 3D. 3D printing which is suitable for forming 3D geometries is used to visualize 3D crack propagation in reinforced concrete. The propagation of 3D cracks visualized in this study is simulated by using the finite element analysis with a damage model based on fracture mechanics for concrete. First, the formulation is presented for finite element analysis of cracks with the damage model. Then, we present several examples of 3D-printed cracks in reinforced concrete to demonstrate the availability of the proposed visualization method with 3D-printing. These results obtained in this paper offer new insights into the study of fracture behavior of reinforced concrete.

View full abstract

Recently, primal-dual interior-point method (IPM) are focused as an efficient strategy for large scale contact problems. Contact algorithm based on IPM, however, is difficult to apply finite deformation problems. These problems should be divided into some loading steps and slave nodes approach master surface to very close. Too strong contact reaction force then occurs at the beginning of Newton-Raphson (NR) loop in a loading step. This force causes instability of computation. In this paper, we present contact algorithm based on IPM with warm-start method for solving this problem. Contact constraints are relaxed by additional variables. We solve some problems, including a large deformation problem, and compare the number of NR iteration to investigate convergence of contact analysis. As a result, we find that the proposed method improves convergence of contact analysis from standard IPM and conventional active set method.

View full abstract

The effect of the flow rate and water level on free surface vortices in a suction sump was studied by using numerical simulation. Free surface flow in an intake channel was solved by using finite volume method for RANS equations with k-ω SST turbulence model. A VOF multiphase model and the open channel model were used to solve the multiphase flow in the sump. Minimum position of air-water interface, air-entrained vortex length, air volume fraction contours and iso-surfaces were used to identify visually the location and shape of the free surface as well as surface vortices. From the numerical investigation with varying flow rate and water level, it was found that when the water level decreased or the flow rate increased, more free surface vortices appeared. The predicted velocity distributions at the entrance of bell mouth and the location of the center core of air-entrained vortices on the free surface were in good agreement with experimental results.

View full abstract

The mechanism of sharp drop phenomenon of peel force on the standard peel test of multilayer films is elucidate using finite element (FE) analysis. In the ordinary peel tests of multilayer films bonded with adhesives, it is well known that the peel force depends on the peel velocity and sharply drops in the higher velocity range. First, we propose an effective two-dimensional FE modeling method to reproduce the phenomenon. Taking a two-layer film as an example, one of the films, the other film, and the adhesive connecting the two films are modeled as an elastic, elastoplastic, and viscoelastic body, respectively. The peeling interface is modeled by the cohesive zone model (CZM) to consider the damage on the interface with cohesive elements. The dynamic implicit scheme based on the backward Euler method is adopted to stabilize the highly nonlinear analysis with damage and large deformation. Second, we give an interpretation of the phenomenon based on the proposed FE results. The mechanism is elucidated from the two different points of view: stress distribution and energy consumption. It is also revealed that the peel force is negatively correlated with the film curvature radius.

View full abstract

In this paper, High Performance Design Pattern is introduced. It is a design pattern dedicated for HPC environment, to implement an abstract data type of relatively small-size as a library using C, C++ and Fortran. Each data entity of the abstract type is actually represented as a set of scalar variables, and each associated procedure is implemented as a pre-processor macro, rather than an ordinary function or a subroutine. As an example, a vector, tensor and small-sized matrix library, AutoMT, is re-implemented based on this design pattern, and the performance benchmark using this highly tuned version is demonstrated.

View full abstract

In wide box girder bridges or multi-girder bridges with wide girder spacing, the bending strain on the flange or slab is not uniform in the direction perpendicular to the bridge axis due to the shear lag effect. For the analysis of shear lag, an analytical method with the assumption of displacement distribution due to shear lag has been proposed, then it has been developed into the method in which displacement is expressed by series expansion. Authors have proposed semi-analytical approach to solve the shear lag problem with the help of the homogenized beam theory. In this paper, we propose a beam element that can take into account the distribution of shear lag displacement from numerical analysis of the homogenized beam. A simple box girder and a continuous box girder are analyzed by the present element and these results show agreement with the analytical solution and ordinary three-dimensional finite element analysis.

View full abstract

The present study proposes a numerical method to evaluate mechanical influence on tunnel structures caused by severe hygroscopic expansion of the swelling clay minerals. These swelling clay minerals are often called smectite and known to degrade the stiffness of ground due to its own damage. Assuming that this specific physical material behavior can be expressed by an isotropic continuous damage model from engineering viewpoint, we develop a numerical evaluation method for soundness degree of tunnel structures based a decoupling multi-scale analysis. In this context, tunnel structures with surroundings are assumed to be the macrostructure, while material level is to be microstructure. These hierarchically nonlinear numerical analysis needs certain computational efforts. With this background, we formulate a simplified secant solution to solve the nonlinear mechanical problem with a cheap computation cost. The performance of the proposed method is verified by a series of numerical examples.

View full abstract

The present study proposes topology optimization of microstructure considering elastoplastic deformation based on a decoupling multi-scale analysis. Energy absorption capacity of macrostructure is maximized under a prescribed material volume of microstructure. It is assumed that microstructure consists of a two-phase material based on elastoplastic von Mises model and that macrostructure is modeled with anisotropic Hill’s elastoplasticity. In this study, we extend the analytical sensitivity method used for mono-scale analysis to the decoupling multi-scale analysis and propose a framework of new multi-scale topology optimization, which can reduce the computational costs with keeping sensitivities highly accurate. It is verified by a series of numerical examples that the proposed method provides reliable optimization results and has a great potential for advanced material design.

View full abstract

Analysis of ship vibration during design conventionally uses finite element analysis, which is highly accurate but is computationally burdensome for large-scale models. To address this, an explicit method of direct time integration that considers fluid–structure interaction with an outside fluid was developed. Numerical simulations using the proposed method were carried out with cubic and rectangular structure models. To assess the accuracy, the proposed method was verified by comparison between results from the proposed method and those from conventional vibration analysis. The observed results, including values for natural frequency, vibration mode, and velocity, were in good agreement with conventional results, showing sufficient accuracy. In addition, the proposed method required less memory for computation than conventional analysis did. Therefore, the proposed method represents a feasible method of analyzing vibration in large-scale models.

View full abstract

An initial estimate of vibration phenomena is advantageous for highly accurate prediction of vibration, such as estimating ship vibration during the design stage. However, large-scale models have been difficult to handle due to the computational power required. A conventional accurate method, finite element analysis, requires a large amount of time for large-scale models. To address this, an explicit method of direct time integration using fluid–structure interaction with an outside fluid was developed. Numerical simulations using the proposed method were carried out for a large-scale ship model. The accuracy of the proposed method was verified by comparison between results from the proposed method and the conventional method. The observed results for natural frequency, vibration mode, and velocity were in good agreement with the results from conventional analysis. Therefore, the proposed method represents a feasible method of analyzing vibration analysis in large-scale models.

View full abstract

In this study, a shape simplification modeling of insect flapping wings is proposed in order to reveal the mechanical roles of the veins in their cambering, since it seems to be unclear how the elastic deformation of insect wings caused by the aerodynamic force results in their camber. In the proposed modeling, the veins are divided into the areas according to their functions, and the macroscopic constitutive relationship in each area is described using the corresponding beam. Furthermore, the wing membrane supported by the veins is described using a rectangular shell. The aerodynamic pressure dominant in the cambering is considered in the static deformation of the model wing. The finite element analysis taking into account the geometric nonlinearity is applied to the proposed model with the setup consistent with the actual insect. As the result, the mechanical roles of the veins in the cambering of insect flapping wings are revealed.

View full abstract