This paper describes a numerical shape optimization method for designing the free-boundary shapes of plate and shell structures. The boundaries are assumed to be movable in the tangential directions to the surface so as to maintain the curvatures of the initial shape. As an application of this solution, the compliance minimization problem is formulated as a distributed-parameter shape optimization problem, and the shape gradient function is derived using the material derivative method and the adjoint variable method. The optimal free-boundary shape is determined by the traction method modified for this shell design problem. With this method, the domain variation with free-boundary shapes that minimizes the objective functional is determined while maintaining the initial curvatures and the mesh regularity. The validity and practical utility of this parameter-free method are verified through fundamental and application examples involving an automotive part.

View full abstract

Because of space restrictions and the necessity of minimum mass, the connecting rod is one of the more heavily stressed parts of the car engine. The heaviest load occurs by gas pressure during firing at the top dead center position of the piston. The second is tensile load which occurs by the inertia force of the piston assembly mass and around the small end of the connecting rod mass. The third stress due to a lateral inertia, or “whip” of the connecting rod is usually negligibly small. In this paper we investigated the optimization of the shape of the normal connecting rod for the car engine, applying the growth-strain method. The analyses were done in case of compressive and tensile loads separately. The growth criterion parameters of the principal stress on the connecting rod were determined from the fatigue strength of the original design data and the actual safety factor. We thought that it is most compatible to use the fatigue strength which is the safety side, smallest value of the material constants, because the connecting rod receives the reciprocating compressive and tensile hard loads. After analysis, the optimized configurations of the connecting rod were compared with that of the original one. Then the reduction rate of the volume of the optimized connecting rod was approximately 22.8%.

View full abstract

Constant stress creep test instead of conventional constant load one should be performed to create an accurate constitutive equation of creep at designing modern high temperature machines and structures. However constant stress creep testing machines have not been widely employed yet because of their complex structures. Furthermore even if the constant stress creep machine were employed, constraint of limbs attached on a gauge part of the creep specimen for measuring creep strain would be another inevitable issue. These extensometer limbs on the uniaxial creep specimen usually prevent uniform deformation along the gauge part. In order to release the deformation of circumferential constraint by extensometer limbs, 24 slits were introduced in each limb. After these slits were machined in the limb of the uniaxial creep specimen, uniaxial tensile tests were firstly carried out to measure the strain distribution along the gauge part employing low alloy and austenitic stainless steels. Subsequently Finite Element Analyses were performed to examine the validity of introducing these slits into the extensometer limb. The analytic results showed that the uniaxial specimen with slits could release the circumferential constraint and raise the uniform deformation along the gauge part. It was also found that there was an optimum depth of the slit to deform uniformly for each material and that its depth depended on the work hardening property of the material.

View full abstract

The present study investigated ductile fracture process numerically by Gurson's constitutive equation. Two parallel cracks having different length are located in the tensile specimen. The vertical and horizontal distances between two cracks are changed systematically. The load-displacement curves were simulated. The crack having larger length effects on the load-displacement curve at the initial stage of fracture. The significant decrease is observed in the load-displacement curve when the coalescence of cracks is occurred. The minimum value of the maximum load can be estimated by a simulation using a single crack whose length is same as the sum of two crack lengths. In addition, assessment method for multi-cracks problem was studied. The study found that H/I criterion can estimate the maximum load reasonably, and also the consideration of the effect of S in H/I criterion may improve its estimation more reasonable.

View full abstract

We present an evaluation method of the crack growth rate and behavior for a semi-elliptical surface crack in a plate subjected to arbitrarily stress distribution. To calculate stress intensity factors, the influence coefficients were developed on the basis of the influence function method with the finite element method. Although the arbitrarily stress distribution such as weld residual stress on the crack surface are difficult to approximate by using a polynomial formula, the stress intensity factor could be calculated by direct mapping of arbitrary stress field obtained from doing finite element analysis of the crack surface. To determine whether the evaluation method based on the influence function method was valid, fatigue crack growth tests were conducted for flat plates with a semi-elliptical surface crack. Flat plate specimens had a distributed residual stress by the non-filler welding before the fatigue crack growth tests. The results of a comparison of estimations and the tests show that both the growth rate and behavior agreed well. From the calculated stress intensity factors based on the influence function method, the asymmetrical crack growth rates and behaviors can be calculated quickly.

View full abstract

This paper describes creep-fatigue lives of Sn-3.5Ag-0.5Cu-Ni-Ge solder at low temperatures. Push-pull creep-fatigue tests under four types of strain wave form were carried out at 253K and 273K to study an application of creep-fatigue life evaluation method for Sn-3.5Ag-0.5Cu-Ni-Ge. There were effects of temperatures on stress amplitude, and tension stress amplitude at 253K was larger than that at 273K in fast-fast wave form test. The higher stress amplitude reduced the creep-fatigue lives. The creep-fatigue lives with slow strain rate were shorter than the creep-fatigue lives with fast strain rate. A linear damage rule and a strain range partitioning method estimated the creep-fatigue lives in a large scatter. A grain boundary sliding model was a suitable parameter for creep-fatigue life estimation for Sn-3.5Ag-0.5Cu-Ni-Ge.

View full abstract

In this paper, the mixed-mode fracture problem of two parallel cracks in a functionally graded piezoelectric material (FGPM) strip is considered. It is assumed that the electroelastic properties of the strip vary continuously along the thickness of the strip, and that the strip is under a uniform electric loading. The crack faces are supposed to be insulated electrically. Fourier transform techniques are used to reduce the mixed boundary value problems to a system of singular integral equations. The singular integral equations are solved by using the Gauss-Jacobi integration formula. Numerical calculations are carried out, and the stress and electric displacement intensity factors are presented for various values of dimensionless parameters representing the crack size, the crack location, and the material nonhomogeneity.

View full abstract

Steel conveying rollers used in hot rolling mills must be exchanged frequently at great cost because hot conveyed strips induce wear and deterioration on the surface of roller in short periods. In this study, new roller structure is considered which has a ceramics sleeve connected with two short steel shafts at both ends by shrink fitting. Here, the ceramics sleeve may provide a longer life and reduces the cost for the maintenance. However, after used for a period the steel shaft has to be pulled out for exchange. Simply, heating outside surface and cooling inside surface of the shaft are necessary for separation. However, attention should be paid to the maximum thermal stress of the ceramics sleeve in the process of separation. In this paper, finite element method analysis is applied to the structure and thermal stress has been calculated with the varying dimensions of the structure. Also several effects on thermal stress have been investigated, such as the effect of shrink fitting ratio, outside diameter, the fitted length, thickness of shaft, materials and so on. Finally the most appropriate thermal conditions to reduce maximum stress and make separation easy have been discussed which is very useful for designing of new rollers.

View full abstract

A modified formulation of floating stress-point integration with the updated Lagrangian procedure for large deformation analysis is presented. The modified formulation introduces an incremental internal force to the equilibrium equation instead of the virtual external force introduced in the previous formulation. With this modification, the temporal continuity of the mechanical equilibrium can be kept without introducing the virtual external force, and thus the accumulating error due to time advancing becomes small compared to our previous formulation. Several examples of large deformation analysis including large deformation patch tests are also presented to show the validity and accuracy of the proposing method in comparison with the finite element method.

View full abstract

Our research focuses on the techniques for monitoring the current density distribution of the real-time electroplating and a novel technique to estimate electroplating current densities on plated surfaces from the data of the electric potentials in the plating cells has been developed for LSI fabrication technology. However we have applied with only two dimensional analysis in our previous researches and it limits the application range of the proposed technique. In this research, we applied the monitoring technique with three dimensional analysis. Although the inverse analysis using the three dimensional analysis tends to be large scale, it is important for the practical usage to monitor current density on target region accurately and efficiently. Therefore we have developed an efficient inverse analysis technique for monitoring of the electroplating current density on the target region. For the selection of the measurement data and the reduction of the calculation amount for the inverse analysis, the proposed technique discards low sensitive components in the observation equation. The criteria of the discarding process are based on error ratios derived by singular values of the observation equation. In order to demonstrate the effectivity of the proposed technique, numerical simulations are performed for a sample polarization curve of copper in copper sulfate. The results show that this technique estimates the current density on target region accurately and its estimation is more efficient than that by the previous technique.

View full abstract

Nuclear fuel assembly is installed in the cask with the necessary protective function during fuel transportation. The structural integrity of the fuel rods which constitutes fuel assembly has to be evaluated to ensure the safety even for the cask vertical drop accident. The buckling behavior of thin shell tubes by an impact weight have already studied by many researchers experimentally and numerically. However the experiments of irradiated fuel cladding tubes accompany many of difficulties and the number of the experiment is limited. The authors have derived the constitutive relation of the irradiated fuel cladding material, Zircaloy-2, in the previous work and have studied the influences of material properties on the buckling behavior of irradiated fuel cladding tube under the cask drop incident by using finite element simulation code, LS-DYNA. The purpose of this paper is to elucidate the effect of impact velocities, impact mass, boundary conditions and eccentricity of the impact point on the dynamic buckling behavior of the irradiated cladding tube. It is found that the larger impact velocities under the same mass and the smaller impact mass under the same impact energy induce the larger load. The boundary conditions and eccentricities of the impact point have not affected the dynamic buckling load and absorbed energy. However, the deformed shape of the cladding tube is changed largely by the boundary condition. It is also found that the Euler buckling load has possibility to underestimate the dynamic buckling load of the cladding fuel rods.

View full abstract

In this study, compression and recovery properties for metal gasket with straight sections and arc sections are studied by using the finite element method (FEM). When a metal gasket is subjected to clamp load, a contact width which is necessary for seal-up occurs due to flattening of the arcs. Then, the combination of the straight sections and the arc sections causes various deformation modes because the combination has a large effect on the characteristic of flattening. In other words, it is possible to control the flattening behaviour by adjusting the stiffness of the straight sections and the arc sections. Based on these facts, it is possible to propose a metal gasket in which the flattening occurs at the clamp load and the dent does not occur, furthermore, the clamp load does not descend very much at the tiny looseness.

View full abstract

This paper deals with the mechanical properties of paper-based friction materials under tensile loading. In-plane tensile tests are carried out for nine kinds of friction materials in order to evaluate the mechanical properties and clarify the influence of constituents on the tensile strength and fracture toughness. The constituents of the friction materials are aramid, cellulose, carbon fiber, filler, and phenolic resin. From in-plane tensile tests the following results are obtained: (1) The stress-strain curves show the large non-linear deformation on the in-plane tensile tests with smooth specimens. The aramid and cellulose fibers are effective for improvement of the failure strain and the tensile strength, respectively. The carbon fibers are also effective to improve the Young's modulus. (2) The fracture criteria based on the net-section stress and J-integral are valid from the results of in-plane tensile tests with pre-cracked specimens. The aramid fibers are effective for improvement of the fracture toughness although the cellulose and carbon fibers have a disadvantage to improve the fracture toughness.

View full abstract

Multi-phase-field (MPF) modeling with a higher-order term, which can stably analyze the triple junction (TJ) behaviors even with large difference between grain boundary (GB) energies, is proposed for the more realistic grain growth prediction. Grain growth simulations of systems including multiple junctions show that the proposed MPF model can represent the stable TJ behaviors with wider range of GB energies than the conventional model, and well perform the quadruple junction behaviors in agreement with the corresponding theory. Two kinds of GB energy distribution models are employed here, which are based on all of and only low cuspate energies of <110> symmetric tilt GB energy in pure Al by molecular dynamics simulations. Polycrystalline grain growth simulations combining with either GB energy distribution exhibited the large amount of nucleation of low-energy boundaries, which would be quantitatively compatible to the experiments. Also, the unrealistic grain structures were observed around the TJs with large GB energy difference when the former distribution model was used. Thus, this fact suggests that the GB networks should be formed by TJs with smaller GB energy difference to avoid the higher unstable energy state.

View full abstract

In this paper, the stress intensity factors (SIFs) of two edge cracks which emanate asymmetrically from a center U-shaped notch in a finite plate under tension were computed by the body force method. The crack which is governed by the elastic notch stress field was treated as a short crack; the crack which is not almost influenced by the notch stress field was treated as a long crack. The cracks were divided into the short and long ones, and the SIFs were each investigated with respect to tendency. Then, it was found that the SIFs become nearly equal when both cracks are larger than 0.5ρ and smaller than (0.5W - t) even if the cracks are asymmetric, where ρ is a notch root radius, t is a notch depth and W is a half plate width. Moreover, the approximate calculations of the SIFs were performed by using the simple formulae for evaluating the SIFs of the eccentric crack in the finite plate and the short symmetric edge cracks emanating from the elliptical hole in the infinite plate. Then, the accuracy of the approximate calculations was examined by comparing with the solutions by the body force method.

View full abstract

To safely use of hydrogen utilization machines after large earthquakes, the effect of multiple overloads and hydrogen on high-cycle fatigue strength of SUS304 and SUS316L austenitic stainless steels were evaluated. Three kinds of notched fatigue test specimens which have different notch root radius were used. The fatigue strength of both materials was significantly reduced by multiple overloads. The cause was small cracks formed by the overloads. In SUS304, the reduction of fatigue strength became more significant by hydrogen. The cause was that hydrogen accelerated propagation of the small cracks during overloading. On the other hand, fatigue strength of SUS316L was insusceptible to hydrogen. Propagation of the small cracks existing notch root was evaluated by plastic strain range at notch root and Manson-Coffin rule in order to consider application of this study for design.

View full abstract

In the steam piping system containing liquid droplets, the pipe wall is damaged by droplet impingement. One of the most important parameter affecting erosion is droplet diameter. It was reported the volume loss per droplet increases with 3.67th power of droplet diameter. But, conventional research, intended droplet diameter or nozzle diameter is 0.81-5.0mm. In this study, rotating disk method and high pressure impingement test method with fan-shape spray nozzle were used to carry out the experiments with small droplets on pure aluminum and low carbon steel S15C specimens. The effect of diameter was expressed as a parameter of volume loss per droplet. The volume loss per droplet increases with 4.67th power of droplet diameter.

View full abstract

In the present paper, the three-dimensional and two-dimensional Green's functions due to point and line forces in anisotropic-isotropic elastic bimaterials are derived by using generalized Stroh formalism and two-dimensional Fourier transforms. When the Stroh formalism is applied to isotropic materials, the eigen matrix derived from equilibrium equations yields a triple root of i (i : imaginary unit), and then an independent eigen vector corresponding to the eigen value can not be determined. In the present paper, the eigen vectors for the triple root are determined and general expressions for displacement and stress are derived. The unknown vectors involved in the expressions are determined using boundary conditions for an interface and at a loading point. Numerical examples are given to demonstrate the validity of the present formulation of three-dimensional point-force Green's functions. The Green's functions include the Kelvin solutions for isotropic materials in two- and three- dimensional infinite domains and Mindlin solutions for isotropic material in three-dimensional half-domain.

View full abstract

This paper describes a quasi ideal importance sampling simulation combined with the conditional expectation method for the simulation-based structural reliability estimation. A quasi ideal importance sampling joint probability density function is formulated on the basis of the ideal importance sampling concept and the respective marginal probability densities are constructed numerically by recursive simulations and piecewise integrations. The respective quasi ideal importance sampling marginal probability densities are renewed through recursive simulation procedures. Numerical examples to estimate the failure probability of structures with multiple failure modes show that the proposed method gives accurate estimations of the failure probability of structures efficiently.

View full abstract

The fitness for service of the facilities of operational nuclear power plants is secured by proper inspections and evaluations. Defects took place in components will be detected by periodical inservice inspection, and fitness for service of cracked components during an evaluation period is evaluated. Although variation exists in the conditions used for inspection or evaluation, it is difficult to grasp about those influences on the reliability of cracked components by the deterministic evaluation like Fitness-for-Service Codes. In this study, influences of inspection parameters (defect sizing error, percentage of defect oversight, detectable defect size, successive tests) and an evaluation parameter (crack growth rate) on failure probability were evaluated for primary loop recirculation system piping of boiling water reactors in which many stress corrosion cracks took place. From these evaluations, the dominant parameters for the reliability of piping having stress corrosion cracks were clarified, and the requirements for the inspection and evaluation for revising Fitness-for-Service Codes were proposed.

View full abstract

In a structural integrity assessment of cracked pipes according to the fitness-for-service code, the influence of mechanical interaction between multiple cracks is considered by assuming coalescence of the interacting cracks. In this study, criteria of the coalescence were examined by finite element analyses for axial surface cracks located inside a pipe. Longitudinal semi-elliptical surface cracks were assumed to be on a straight pipe which was subjected to internal pressure. The changes in the stress intensity factor, J-integral value and limit load were investigated under various conditions. The intensity of the interaction depended not only on the relative position of the cracks but also on the pipe and crack geometries. Based on the analysis results, the coalescence criteria for assessing the failure strength of the cracked pipe were discussed. It was concluded that, for the stress intensity factor and J-integral value evaluations, the cracks should be treated as a combined crack when the offset distance H is less than 0.6a, where a is the crack depth. As for the limit load analyses, the criteria of S ≦ 3a and H ≦ 3a/2 were obtained. Furthermore, for limit load analyses, it was found that the surface length of the combined cracks could be assumed to be equal to the sum of their lengths. The ligament length (distance between the cracks) was not necessary to be included in the surface length of combined crack.

View full abstract

Usually, development of automobile brake hose and power steering hose has been realized through investigating several actual prototype hoses experimentally. Recently, high durability for brake hose has been required because periodic renewing the brake hoses has not been requested anymore. In this study, three-dimensional FEM analysis has been applied to the crimped portion of hydraulic brake hose in order to investigate the effects of manufacturing errors upon the sealing performance. In order to evaluate the sealing performance, the normal stress σ_r appearing between the inner rubber and nipple is mainly considered. Then, the sealing performance is investigated when the manufacturing error is included for the crimped portion of hydraulic brake hose. It is found that maximum stress between the nipple and inner rubber decreases by less than 30% when the manufacturing errors are smaller than the upper limit of tolerance. It may be concluded that the error may causes deterioration of sealing performance of hydraulic brake hose.

View full abstract

[in Japanese]

View full abstract