In recent years, development of the wireless communication technology is able to provide useful information and active safety for the drivers. Though many researchers try to monitor a traffic congestion, weather condition, road surface condition and car maintenance by the on-board TV camera or the probe car system, sensor signals of these have non-stationary and transitory characteristics while the vehicle is in motion. In this study, we specially turn attention to the feature extraction for the strain signals while the vehicle is in motion. We examine the local analysis based on the Discrete Wavelet Transform and the Short-Time Fourier Transform applied to the feature extraction of the failure detection.

View full abstract

In this paper, the effect of height h of the hexagonal honeycomb core on the elastic modulus is studied by using numerical results of finite element method. For honeycomb subjected to in-plane loading, there exist deformations in the height direction for cell walls due to effect of Poisson's ratio. Because these deformations are different for adjacent walls if they are not joined together along the common edge, the in-plane stiffnesses should be analyzed as a 3-dimensional problem, in which the deformations in the height direction are considered. Also it is shown that there is no influence of height to the elastic moduls for the honeycomb under the plane strain condition. Based on this fact, an equation to calculate elastic modulus in the direction of apex E₂ from their values under the plane strain condition is proposed, and its validity is verified by using numerical results of FEM.

View full abstract

In this study, the method of estimating fiber orientation distribution in a short fiber reinforced composite is numerically examined. Accuracy of fiber length obtained by the simple theory is also examined. For this purpose, packing of the fibers in space is simulated by a computer. The fibers are packed on a cubic. Position and orientation are given to a candidate fiber by random numbers. The candidate fiber is arranged in the cubic. If the fiber does not overlap other fibers already existing, the candidate fiber is adopted as a regular fiber in the composite. This procedure is repeated until becoming prescribed fiber volume fraction. The short fiber reinforced composite made on the computer is sliced to prescribed thickness. The sliced composite is observed, and orientation of fibers is measured from the composite surface. Fiber orientation distribution is calculated from the data obtained by the measurement. The estimated distribution is well near actual distribution in the composite. Furthermore, the average fiber length is estimated by substituting the number of the fiber tips counted from the composite surface to the simple theory. The estimated fiber length is about 8% to 20% larger than the actual fiber length.

View full abstract

In this paper, a parallel monolithic method for shell-fluid interaction based on the consistent Pressure Poisson Equation (PPE) is developed and its parallel computational efficiency is demonstrated. The previous and present studies show that iterative solvers without preconditioning work well to solve the PPE, even though the coefficient matrix of the original linearized coupled equations becomes to be ill-conditioned. As a consequence, combining with the iterative solvers without preconditioning parallelized based on the mesh decomposition, the developed method can archive efficient parallel computation. To demonstrate the performances of the developed method, it is applied to simulate the vibration of an elastic plate situated in the wake of a rectangular cylinder and a flapping elastic wing in fluid.

View full abstract

In present research, B-spline wavelet Galerkin method is applied to solid/structural mechanics analyses. B-spline scaling function and wavelet are used as the basis functions. The basis functions have the so-called multiresolution properties. The spatial resolutions of interpolation functions can be enhanced in the regions of high stress/strain gradients by superposing finer basis functions. B-spline wavelet Galerkin method can discretize a solid by structured cells. There are no finite element meshes in a classical sense. Then, it can be classified to be a kind of meshfree method. To control the spatial resolution of the discretization, an adaptive analysis strategy is developed based on a posteriori error estimation. The solution can be refined by superposing different wavelets of finer length scales. In this paper, the developed methodology and some numerical demonstrations are presented.

View full abstract

In order to improve piezoelectric response of polycrystals in a macro scale, micro distribution of crystal orientations was optimized by using multiscale finite element method based on crystallographic homogenization theory. As a result, two characteristic optimum solutions were obtained for micro structure. One is a polycrystalline structure laminated with three specified grains for macro piezoelectric strain constant d₃₃₃, the other is a polycrystalline structure consisting of two regularly-neighboring grains with specified orientations for d₃₃₁. As the maximum micro strain occurs by electric and mechanical effects, the optimized micro structures exhibit the highest macro piezoelectricity that is beyond single crystal. Their micro structures are available for piezoelectric ceramics such as BaTiO₃ and Pb (Zr, Ti) O₃, that have the maximum piezoelectric strain constant under off-axis electric field.

View full abstract

Fracture strength of CVD diamond films deposited by the hot-filament chemical vapor deposition method on sintered SiC substrate was estimated using the indentation method and FEM analysis. The authors determined the critical indentation force to cause ring crack in the diamond film correctly utilizing both of the acoustic emission (AE) and corrosion potential fluctuation (CPF) method during indentation of Rockwell indenter of natural single crystal diamond with radius of 400 p.m. The polarity distribution of the first arrival symmetric mode of the Lamb wave AEs was utilized for the classification of the Mode-I ring crack and the Mode-II interfacial exfoliation. The CPF measurement in the acidified solution was possible due to weak electrical conductivity of CVD diamond and SiC. The critical indentation force for the micro-meter size polycrystalline diamond (MCD) and nano-meter size diamond (NCD) films with different thicknesses were accurately determined by AE and/or CPF. Fracture strength of these films were estimated by FEM using the critical indentation forces. The fracture strength of the MCD was estimated as 6.4 GPa for 21 p.m thick MCD and 4.0 GPa for 70 pm thick MCD film. Fracture strengths of these films were supposed to be determined by the transngranular cleavage crack, while that (6.1 GPa) of the NCD film was determined by weak graphite phase along grain boundaries.

View full abstract

We have so far performed many molecular dynamics simulations on various pseudo-polycrystalline and amorphous nickels under tension, suggesting that neither the system nor the atomic elastic stiffness coefficients, or the global and local stabilities, could predict the “macroscopic” yield of disordered systems. The previous simulations, however, are implemented under the displacement control so that it would not be the force balance but the geometrical requirement that dominates the yield. In order to clarify the effect of external loading condition, we have achieved the tensile simulations again with the force control condition. The effect of the constraint on the transverse Poisson' contraction is also investigated. Contrary to the previous report, the stress-strain peak exactly coincides with the point where the deviation of local stability vanishes and the determinant of both the system and atomic elastic stiffness coefficients becomes zero, under the force control condition without Poisson' contraction. The system and local stabilities also well predict the stress drop under the displacement control without Poisson' contraction. On the other hand, the dynamic loading leads an unrealistic elongation under the force control condition with Poisson' contraction ; however, the local stability definitely points out the onset of the rapid elongation while the system one merely does that of dilatation after the unrealistic elongation.

View full abstract

To elucidate the toughening mechanism in rubber particle modified semi-crystalline polymers in which the average matrix ligament thickness plays important roles, the micro-to mesoscopic deformation behavior of rubber/semi-crystalline polymer blends was modeled by using large-deformation finite element homogenization method. In this paper, the deformation behavior of three kinds of models which have different size of rubber particles was investigated. A series of computational simulation clarified that highly localized deformation is induced by the initial orientation of lamellae at interface region for the case of smaller size of rubber particles, and it produces larger strain in the matrix due to localization and propagation of the deformation zone. Furthemore, the microscopic deformation is mainly absorbed by not only a slip along chain direction in the crystalline phase, but also interlamellar shear in the amorphous phase, which must be closely related to the toughening mechanism of the rubber/semi-crystalline polymer blends.

View full abstract

A spline is a stretchable joint that transmits power. Generally, for the fracture problem in an actual construction, three-dimensional cracks exist in most cases, and in crack propagation, the stress intensity factors K_I, K_II and K_III usually act in single or mixed modes. In this study, the maximum stress and maximum stress direction are determined by the photoelastic and finite element methods. The mixed modes stress intensity factors K_I, K_II and K_III are determined experimentally by combining the caustic method and photoelastic stress freezing method. The interaction in the spline shaft having outer surface crack was also discussed, together with a comparison of the results of these experimental methods.

View full abstract

NDT techniques based on the ultrasonic phased array technology are applied in various industrial contexts. For the reliable application of phased array techniques, it is essential to have well knowledge on the characteristics of radiated beam profile from array transducer and ultrasonic wave propagation in test materials. This paper proposes a mathematical model of the phased array transducer and a simulation tool to estimate transmitted waves from the transducer and scattered waves from flaws. The simulation tool is based on the BEM, and a pulse shaped wave is used in this analysis for the transient wave propagation. Here wave fields are visualized to understand the feature of scattered waveforms measured at the array transducer.

View full abstract

In order to examine the mechanisms of delamination between coating layer and substrate under fatigue loading in steel thermally sprayed with Ni-based self-fluxing alloy, three types of specimens with different coating thickness were prepared. The fusing was performed by using an induction heating system in order to change the stress distribution. Rotating bending fatigue tests and observations of fracture surfaces were carried out. With all fracture surfaces, the delamination between the coating and the substrate occurred. However, the coating thickness strongly affected the fatigue properties ; the thicker the coating, the lower the fatigue strength. In order to clarify the mechanism of delamination, the differences of strain on the interface between the coating layer and the substrate were performed by using FEM analysis. The differences of strain strongly affected the delamination between the coating and substrate ; the thinner the coating, the smaller the differences of strain. Considering these results and observing the interface during the cyclic loading in detail, it is considered that the greater the differences of strain on the interface between the coating and the substrate, the earlier the delamination occurred and propagated, and a lower fatigue strength resulted.

View full abstract

An ultrasonic inspection system was proposed to observe a subsurface crack growth behavior continually in two-roller rolling fatigue test. A 50 MHz delay line contact probe was used in the system to detect clearly small cracks near the contact surface and to conduct the ultrasonic inspection while the test rollers were kept attached to the rolling fatigue testing machine. The ultrasonic wave pattern was analyzed by digital image processing to predict the position and shape of cracks. The proposed system was adequate to the inspection of a crack which had about 1.5 mm circumferential extent parallel to the contact surface in the subsurface region and inadequate to the continual observation of cracks which originated at the contact surface. This result suggests that cracks originating at the contact surface hardly have a large enough extent to reflect the ultrasonic wave in the crack growth process but grow to failure very rapidly only in the final stage of the process.

View full abstract

The fatigue strength at N=10⁷ for carburized steels, which indicates the fatigue strength for specimens whose fatigue origins are not at inner defects but at as-carburized surfaces in this study, is affected by a softened structure, oxidized intergranular, retained austenite and residual stress. In the early study, it was clarified that the fatigue strength at N=10⁷ of the specimens containing the retained austenite in the structure was higher than that estimated by the √area parameter model.This study aims to quantitatively evaluate these complex influences mentioned above and predict the fatigue strength at N=10⁷ with high accuracy. Therefore, rotating bending and three point bending fatigue tests with several stress ratios were conducted using carburized SCM420s. Since the fatigue crack of all the specimens, whose fatigue origin was at the as-carburized surface, emanated from oxidized intergranular at a non-martensitic structure, the oxidized intergranular was considered to be mechanically equivalent to a surface crack. The dependence of mean stress on the fatigue strength at N=10⁷ was accurately arranged by the modified Goodman diagram. Furthermore, the hardness of a thin softened structure was estimated from the half value breadth of the structure. Based on these results, the prediction of the fatigue strength at N=10⁷ for the carburized SCM420s was performed by applying the √area parameter model, and its fatigue strength was successfully estimated at an unquestionable level for practical use.

View full abstract

Slow strain rate tests (SSRT) and fatigue tests were conducted to investigate the influence of internal hydrogen state on the quasi static SSRT strength and fatigue strength for cold drawn eutectoid steel specimens which were cathodically hydrogen charged. Internal hydrogen states were changed as follows : (a) virgin sample or non-charged one, (b) the sample that contained both diffusive and non-diffusive hydrogen, and (c) the one that contained only non-diffusive hydrogen. The SSRT strength and the fracture displacement of only non-diffusive hydrogen charged samples were the same as that of non-charged sample, whereas those of the sample having both diffusive and non-diffusive hydrogen were smaller than the others. This indicates that only the diffusive hydrogen caused hydrogen embrittlement under the quasi-static loading condition. The fatigue strength of the specimen having both diffusive and non-diffusive hydrogen decreased compared with non-charged specimens. However, S-N curves had a large scatter and the influence of non-diffusive hydrogen on the fatigue strength was unclear ; because the fatigue crack was initiated at internal inclusion. Therefore, the fatigue strength was discussed based upon the stress intensity factor, ΔK_inc, calculated from stress and inclusion size. ΔK_inc giving the same N_f decreased in the order of the non-charged specimen, only non-diffusive hydrogen charged specimen and the specimen having diffusive and non-diffusive hydrogen, indicating non-diffusive hydrogen caused a decrease in fatigue strength.

View full abstract

Diamond films were synthesized on a Mo substrate using combustion flame. During the cooling process, most diamond films delaminated. In this study, to prevent the delamination of the interface, effects of the flow ratio (R_f=F_o/F_a) of oxygen flow rate (F_o) to acetylene flow rate (F_a) on the delamination were investigated by the experiment. The results show that most prevention of the delamination was obtained in the case of the flow ratio R_f=0.90 for synthesis. The abnormal growth of synthesized films occurred in the case of flow ratios R_f=0.75, 0.80. Synthesized films delaminated in the case of flow ratios R_f=0.95, 1.00. In order to specify the synthesized substance, SEM and XRD were used. The results show that the synthesized film was a good diamond in the case of the flow ratio R_f=0.90. It concludes that the delamination-free crystal growth can be realized at the flow ratio R_f=0.90.

View full abstract

This paper describes the development of a robust and efficient system for optimization of structural design of artifacts on the Grid, considering the following requirements : (1) Computation subtasks with less dependency are processed asynchronously, and (2) An optimal solution with specified accuracy is obtained in a robust and prompt manner. To satisfy Req. (1), evaluation calculations such as finite element (FE) structural analyses are asynchronously performed on the Grid. Each FE analysis is performed using an open source parallel FE system named ADVENTURE. To satisfy Req. (2), a robust and efficient optimization algorithm based on real-coded genetic algorithms is developed. In addition, robust processing mechanisms are implemented to perform a whole of computation on the Grid without interruption. In this study, we test the following two kinds of Grid environments : a group of PC clusters distributed in remote sites and one of Japanese national Grid projects named ITBL (IT-based Laboratory). Practical performances of the developed system are demonstrated through some examples.

View full abstract

When the relationship between the dimensions of face form and those of spectacle frames is poor, it may cause problems such as slipping frames and sore areas on the nose and behind the ears. In this study, mechanical characteristics in face-spectacles interface are discussed. CAD models of face form and spectacle frames were constructed from the three dimensional measurements of the human face that wore spectacles. The finite element method (FEM) model of the face was constructed by using CT images of the scalp organization. Next, the FEM model of spectacles constructed with the 3 D measurements was synthesized to the face model. Then, mechanical behavior in the interface between the face and spectacles was investigated by FEM analysis. In the result of the frame which was adjusted to the face, contact area between tip and skin is so wide that the effect of holding spectacles stably was recognized with the end of tip hold head. At the frame which was not adjusted, contact area is so small that enough pressure was not generated to hold spectacles stably.

View full abstract

In general, internal cells are required to solve elastoplastic problems using a conventional boundary element method (BEM). However, in this case, the merit of BEM, which is easiness of data preparations, is lost. Triple-reciprocity BEM can solve the two-dimensional elastoplasticity problems with a small plastic deformation. In this study, it is shown that three-dimensional elastoplastic problems can be solved without the use of internal cells, using the triple-reciprocity BEM. An initial strain formulation is adopted and the arbitrary distributions of the initial strain for elastoplastic analysis are interpolated using boundary integral equations and internal points. This interpolation corresponds to a thin plate spline. The fundamental solutions for this analysis are shown using polyharmonic function with volume distribution. The theory is expressed using a few fundamental solutions. In this method, strong singularities in the calculation of stresses at internal points become weak. A new computer program was developed and applied to solving several problems.

View full abstract

The elastoplastic stress is first defined as the stress which evolves as the actual strain rate is induced in an imaginary quasi-static process of elastoplastic deformation, while internal variables evolve with the viscoplastic strain rate calculated by the viscoplastic constitutive equation. Further, the novel variable “overstress tensor” reaching the current stress from the elastoplastic stress is defined. Then, the overstress model is extended so as to describe also the tangential viscoplastic strain rate induced by the overstress tensor component tangential to the yield surface. Furthermore, the viscoplastic strain rate due to the change of stress inside the yield surface is incorporated by adopting the concept of the subloading surface which falls within the framework of the unconventional elastoplasticity describing the smooth elastic-plastic transition fulfilling the smoothness and continuity conditions.

View full abstract

In the die casting process, a formation of porosity is unavoidable. The porosity influences the mechanical properties and air leakage resistance of the products. We have investigated optimum manner of post compression process of ADC 12 aluminum alloy die castings to reduce inside porosity size. The reduction efficiency was affected by initial size and distribution of the porosity. In this study, two-dimensional imaged-based finite element analysis was conducted to find the optimum manner of post compression process considering porosity size and porosity distribution inside die castings. The mesh for finite element analyses was generated from X-ray CT images of the ADC 12 aluminum alloy die casting specimen. Comparing the deformation of porosity of numerical result and experimental result, we proved that the two-dimensional image-based finite element method works well to estimate the effect of the post compression process. We investigated the porosity reduction efficiency of biaxial and triaxial compression by using the same two-dimensional image-based finite element mesh.

View full abstract

Authors have proposed a new nondestructive method of evaluating welding residual stress, the Bead Flush Method, and have confirmed its validity for simple butt-welded plates. In this method, eigen-strain distributions are estimated from strain changes due to removal of reinforcement of the weld using the inverse analysis. Then, residual stresses are estimated by the elastic FEM (Finite Element Method) analysis using estimated eigen-strain distributions as initial strains. In this study, further development of this method was attempted for application to welded pipes. Eigen-strain distributions in welded pipes are more complicated than those of welded plates from a viewpoint of residual stress estimation. A basic formulation by assuming axisymmetric condition is attempted for this problem and problems to be solved were made clear.

View full abstract

In this paper, in considering the generation of the inflatable parabolic reflector, we describe the reflector surface generated with gore. The gore is generated by cutting the three-dimensional parabolic surface. First, the rms surface error between the shape by pressurizing the circular membrane and the parabolic surface is formulated. The scheme of generating the gore is described. The rms surface error between three-dimensional shape by connecting the gores and the parabolic surface is formulated. Finally, the possibility that the parabolic reflector with gores has high surface accuracy is shown.

View full abstract