Axial creep tests of a 10vol% boron-aluminum hot-pressed monolayer composite were carried out under several constant loads at 300°C in air. The composite exhibited slight primary creep, but did not show appreciable secondary creep. Several specimens experienced a momentary increase of strain during the creep test which separated the creep curve into two regions, because of individual fiber breakages in the composite. Then, almost all the specimens abruptly fractured without a tertiary creep. The creep-rupture surfaces consisted of relatively complicated paths with a few flat regions perpendicular to the fiber axis linked by longitudinal matrix shear fracture. From the viewpoint of materials reliability engineering, the statistical properties of the creep-rupture lifetime were investigated. The average lifetime decreased with an increase in the applied load, and the considerably large coefficients of variation were estimated in every case, being around 1000%. However, these scatters were estimated to be smaller than the scatter of the creep-rupture lifetime of the boron fiber itself. That is, such statistical relations between lifetime properties of the fiber and those of the composite are almost the same as those for tensile strength. Additionally, a stochastic model which predicts theoretically composite strength from fiber strength was developed for predicting the probability distribution curves of the creep-rupture lifetime. The predicted distribution curves agreed well with the experimental results under high applied load conditions.

View full abstract

Since carbon fiber reinforced carbon composite (C/C composite) materials have good mechanical properties at high temperature and low radioactivity under irradiated conditions, they have been used as the plasma-facing components in fusion facilities. Efficient utilization of these mechanical properties will lead to the use of c/c composite materials, these materials are expected for use as control rods for gas-cooled rectors and irradiating capsules. In the case of using c/c composite materials as the structural components, it is very important to evaluate the mechnical and thermal properties as well as fracture behavior. For structural components used in a reactor core, it is necessary to evaluate neutron-irradiation effects on these materials. However evaluation of these effects has been difficult, thus data on irradiated c/c composite materials are lacking. In this study, three kinds of c/c composite (CC-312, CX2002U and MCI) made of different fiber materials were irradiated under the condition of maximum neutron fluence of 2×10²⁵ n/m² and maximum irradiating temperature of 1200°C, and the relationship between change of mechanical properties due to irradiation and the constitution of c/c composite materials was evaluated. In the case of CC-312 made of PAN fiber, Young's modulus changed slightly and bending strength after irradiation. In the case of CX2002U made of rayon fiber, Young's modulus and bending strength increased after irradiation. In the case of MCI made of pitch fiber, Young's modulus increased after irradiation, but bending strength decreased slightly after irradiation. These results enable the elucidation of the relationship between properties of fiber meterials and irradiation effect on c/c composite materials.

View full abstract

Interlaminar crack extension in continuous fiber-reinforced laminates due to accidental transverse loading is one of the most serious problems in actual service applications of this kind of composite. Here we attempt to numerically obtain the relation between the in-plane two-dimensional extent of interlaminar cracks and the transverse load applied to the laminate by quasi-static indentation. Our attention is focused on the local energy release rate along the interlaminar crack front. The two-dimensional multi-layered finite-element method is developed, which employs two-dimensional bending elements of each lamina stacked to represent the laminate. This FEM enables us to easily estimate the energy release rate at any point on the interlaminar crack front. Under the condition of a constant critical energy release rate, we carry out the computational simulation of interlaminar crack extension for the case of a three-layered cross-ply laminate. Simulated extension behavior of interlaminar cracks under increasing indentation load appears to be in good agreement with the experimental observations.

View full abstract

An analysis of a crack with bridging fibers in an orthotropic elastic infinite plate is carried out. Based on Lekhnitskii's anisotropic elastic theory, the distributions of the stress and the displacement of the orthotropic elastic infinite plate with a crack are analyzed under a uniform remote tension stress. As well as utilizing the model of fiber pull-out, the relationship between the traction force at a bridging fiber and the crack opening displacement is established. The stress intensity factor K_I and the strain energy release rate G_I of the crack are obtained by linear elastic fracture mechanics, and the influence factor B_K of the bridging for the stress intensity is defined. Finally, in order to discuss the effectiveness of the present method, an example problem, the finite element method solution of which was given, is numerically solved.

View full abstract

The fracture behavior and crack propagation process of carbon-fiber-reinforced laminate composites are examined by in situ SEM, and acoustic emission (AE) signals are measured simultaneously. As a result, the fractures of matrix, fiber-matrix interface and bridge fiber give rise to AE signals, and the patterns of AE event count rate are found to depend strongly on the fiber orientation in a laminate. The direction of crack propagation is along the fiber orientation even for cross laminates and crack propagation is terminated by the reinforced fibers. The amplitude of the AE signal is low at the period of initial crack occurrence, but becomes high and the number of AE events increases rapidly as the crack propagates. This means that microfractures in the material increase during crack propagation. This result can also be confirmed by the results of an AE power spectrum analysis. Furthermore, by matching AE signals to the fractographs, it is found that AE activity is weak and has a low amplitude in the region of a flat fracture surface, while it is strong and has a high amplitude in the rough surface region, especially for a unidirectional material A.

View full abstract

When a joint between materials with different properties is subjected to cooling and heating, thermal stresses develop near the apex of the joint due to thermal expansions mismatch. The thermal stresses are represented by Kγ^p-1, where p-1 is the order of the stress singularity and K indicates the intensity of the stress field. The order p-1, which is calculated from an eigenequation, depends on the wedge angles of the materials and their mechanical properties (Young's modulus and Poisson's radio). The investigation under conditions where the stress singularity disappears has so far been performed by examining whether or not a root for the eigenequation exists in the range of 0<p<1. The stress singularity disappears in the case of combinations of materials yielding K=0, even if the root p exists in the range of 0<Re(p)<1. In this paper, the relationship between K and various combinations of materials is studied in detail. For the conditions yielding K=0, the relation between the wedge angle φ₁ of material 1 and the sitffness ratio k₁₂, and the one between φ₁ and the total angle φ₁+φ₂ of the bonded wedge are examined, and a method for determining conditions yielding approximately K=0 is presented. Also, the distribution of the stress intensity is classified into two modes, which exchange under conditions satisfying K=0, p→1 (p=1 is a double root of the eigenequation) and p=ξ±iη (p is a complex root) depending on K=0. Moreover, when the materials and the wedge angles of both side regions in a three-phase bonded structure are identical, respectively, the distribution of the stress intensity against angle coordinate θ is either zero or an even function for components (γγ, θθ) and an odd function for component (γθ).

View full abstract

The plane elastic problem of a crack in an infinite plate which contains an elliptic inclusion is analyzed by the body force method. In numerical analysis, a fundamental solution for the force acting at a point in the infinite plate containing the elliptic inclusion is used. Based on numerical results, the effect of the inclusion on the stress intensity factors of the crack is investigated.

View full abstract

A method is developed for nondestructive evaluation of multiple cracks by means of measurement of magnetic field produced by d.c. current flow. Two-dimensional multiple cracks are treated. The distance between cracks is assumed to be known in advance and the depth of cracks is to be found. First the distribution of the change in magnetic flux density is measured over each crack by input and output of d.c. current on both sides of each crack. Next the distribution of the change in magnetic flux density is calculated numerically by assuming the depth of every crack. Then by comparing the measured and calculated changes in magnetic flux density, a correction factor is obtained for the assumed depth of the respective cracks. Modifying the crack depth by the correction factor is repeated until the difference between the measured and calculated distributions of the change in magnetic flux density is minimized. It is shown that multiple cracks are sized accurately by the present method.

View full abstract

The object of the present study is to develop a method for enhancing the sensitivity of the DCPD technique for evaluation of length and inclination of an inclined crack on the surface of materials. A method using extremely adjacent probes is proposed for this purpose. The potential drop technique uses four probes which are a set of two probes for current input and output and another set of two probes for measuring potential drop. In the present method, potential drop measuring probes are located near the crack and current input and output probes are set close to them. Some experiments conducted on laboratory specimens validate the use of this method for nondestructive evaluation of cracks based on the utilization of a small sensor composed of the probes. It is shown that the sensitivity of the present method is higher than that of the method locating current input and output probes far from the crack.

View full abstract

Fatigue reliability of HIP-treated aluminum alloy casting was evaluated with a compact test machine. the main results obtained were as follows. Fatigue limit of aluminum alloy casting could be recognized in two types of fatigue test. Difference of fatigue life between axial and plane bending fatigue tests was resulted in size effect of specimen. Fatigue cracking of HIP-treated aluminum alloy casting initiated in grain of aluminum alloy was induced by slip. Fatigue crack growth characteristics of aluminum alloy casting were evaluated using ΔK. The da/dN-ΔK curves was controlled by Paris'law, and ΔK_th was not found in the fatigue crack propagation diagram of this material. Fatigue life of aluminum alloy casting was evaluated with attention to reliability analysis, and was found to be affected by aluminum grain size in the bulk of the specimen.

View full abstract

Slurry-wear resistance was evaluated on pump materials of castings, rollings, thermal sprayings, overlayings and a plating using a jet apparatus. The wear resistance (a reciprocal of volume loss rate) as a function of HV²/E (HV: Vickers hardness and E: Young's modulus) exhibited two curves. The wear resistance of brittle materials is lower than that of ductile materials. However, the resistance to penetration (a reciprocal of maximum wear depth rate) increases with HV²/E irrespective of material brittleness. This means that the wear spreads widely for brittle materials, while it concentrates in a small region for ductile materials. Therefore, the relation between maximum wear depth rate and volume loss rate was divided into two groups. For brittle materials, maximum wear depth rate is relatively low, although volume loss rate is high. It was confirmed that the above conclusions hold for the prototype tests using an actual pump.

View full abstract

Rotating bending fatigue tests of Ti added extra low carbon steel with no strain ageing were carried out. The results obtained are as follows; 1) At the stress of the fatigue limit, fatigue cracks did not propagate when N (number of repetitions) reaches 10⁷. This was confirmed to the extent of N=8×10⁷ in this experiment. That the strain ageing plays an important role in fatigue is well-known, but the above result shows that it is not always necessary for the non-propagation of fatigue cracks. 2) Apparent coaxing effect could not be observed in this experiment. Thus it is clear that strain ageing is an inevitable factor in the coaxing effect. Although propagation of crack tips become slightly harder during fatigue stressing, other regions are not reinforced to the crack initiation by strain ageing. Therefore new fatigue cracks initiate and propagate, resulting in failure, when the stress amplitude is increased.

View full abstract

The fracture process and microfracture behavior of SFC materials with different fiber surface treatments were examined under a polarized microscope in situ. We observed that the microfracture forms, such as matrix crack, void, and debonding, at fiber breakage greatly varied as the fibers or the conditions of fiber surface treatments were changed. By tracing one break in the fiber we found that microfracture forms propagated and the fracture scale became larger at each fiber breakage as the load increased. By observing fiber breakage occurring at different loads it was found that the microfracture forms and fracture scale also varied. Microfracture forms and fracture scale at fiber breakage depended not only on the interfacial strength of the fiber/matrix but also on the load. A model to describe the interfacial shear stress was proposed according to microfracture forms, and the interfacial strength for each material can be evaluated qualitatively using this model.

View full abstract

The tensile test of SiC_CVD/Al composites and their component materials was carried out by detection of acoustic emission (AE) during deformation. The characteristic of AE in the fracture process was investigated in connection with the microfractures. The results obtained were as follows: when the fibers of the composite were broken, high-amplitude AE signals were generated and the frequency was about 0.2 MFHz. Both duration and oscillation count of the signal increased with increase of the amplitude. Therefore, both parameters are effective for evaluation of microfracture of the composites. The number of microfractures in composite decreased with increase of the fiber volume fraction to the maximum stress, that is, the fracture became noncumulative with increase of the fiber volume fraction.

View full abstract

Microfracture processes of microcracking and crack propagation are simulated for 2-dimensional alumina polycrystals which have thermal anisotropy within a grain. Microcracks are generated by the thermally induced residual stresses at a grain boundary. The stress concentration near the microcrack is calculated numerically by the body force method, and superposed on the pre-existing residual stress. Stress intensity factors at the microcrack tip are also obtained by the method, and the location at which the next microfracture occurs is determined by the competition between the microcracking and the crack propagation in the new stress state. The microfracture stress increases with the progress of the fracture and decreases after maximum indicating a fracture strength. In many cases, the propagation of microcracks induces an unstable fracture. The number of microfractures prior to the unstable state decreases with decreasing grain size and increasing grain boundary toughness. For the alumina of the grain size of 17.5μm, when the fracture toughness of the grain boundary is 0.6 times that of the grain or greater, the unstable fracture occurs prior to microcracking process.

View full abstract

Thermal stress during rapid solidification is analyzed based on the motion of particles by means of a two-dimensional molecular dynamics method. The Lennard-Jones-type potential is assumed as a two-body potential. The liquid is solidified by reducing the volocities of the particles in the cooled region at a constant rate. The numerical results show that the concept of the thermal stress obtained by molecular dynamics simulation agrees with that used in continuum mechanics if the number of particles is sufficient in the region defining the thermal stress. Thermal stress decreases when lattice defects appear.

View full abstract

For a nonhomogeneous medium, both the heat conduction equation and the governing equations of an associated thermoelastic held are nonlinear forms in general. Therefore, the theoretical treatment of these nonlinear equations is very difficult and an exact solution is almost impossible to obtain. By introducing a laminated composite model, we derived a one-dimensional temperature solution for a nonhomogeneous plate in a tarasient state in our previous paper. In this paper, making use of this temperature solution, the optimization problem of material composition to minimize the transient thermal stress distribution is described. As a numerical example, two kinds of nonhomogeneous plates composed of zirconium oxide/titanium alloy and alumina/aluminum alloy are considered. Then the optimum material composition is determined from calculations. Furthermore, the temperature dependence of material properties is discussed.

View full abstract

This paper is concerned with thermal ratchetting of a cylinder subjected to a temperature front moving cyclically in the axial direction. The thermal ratchetting strain yielded before shaking down is analyzed by approximating the deformation profile of the cylinder trilinearly and by assuming Masing's rule of cyclic plasticity. An analytical expression for the thermal ratchetting strain is thus derived in the case of a power-law type stress versus strain relation with no temperature dependence of material parameters. The resulting expression is verified by comparing its prediction with results of finite element analysis and experiments on 316 FR steel.

View full abstract

Nonhomogeneous materials such as FGM (Functionally Gradient Materials) have arbitrarily distributed and continuously varying material properties. For such nonhomogeneous materials, the heat conduction equation takes a nonlinear form, which makes theoretical treatment of the temperature change difficult, and it is almost impossible to obtain the exact solution. In this paper, taking into account the effect of the heat transfer at the heated surface and cooled end, a one-dimensional transient heat conduction problem of a plate made of such nonhomogeneous materials is treated theoretically. Introducing the analytical technique for a laminated plate model, and considering the number of lamina to be sufficiently large, the temperature solution for such a nonhomogeneous plate is approximately derived. The associated thermal stress distribution for an infinitely large nonhomogeneous plate is formulated under the traction free-condition. Numerical calculations are carried out for a nonhomogeneous plate, made of zirconium oxide and titanium alloy. Numerical results such as temperature change and the thermal stress distribution are shown graphically, and the influences of a change of relative heat transfer coefficient at the heated and cooled surfaces and of a change of nonhomogeneity are briefly examined.

View full abstract

Two-dimensional asymmetric dynamic thermal stresses in a hollow circular cylinder subjected to a sudden time-dependent temperature field are studied numerically. The method employs the explicit finite difference approximations with second-order accuracy based on the integration of the governing equations along the bicharacteristics. Numerical calculations are carried out to analyze the propagation and reflection of thermal stress waves in a hollow circular cylinder subjected to sudden internal heat generation caused by the absorption of gamma-ray or electromagnetic radiation. The rate of heat generation is assumed to be constant for the duration of the pulse and to diminish exponentially with distance from the surface of cylinder and is an arbitary function of the angle around the axis of the cylinder. Heating time is accounted for ; however, heat conduction is ignored. Several numerical results are presented that illustrate the significance of the dynamic effects and their dependence on the coordinates and heating time.

View full abstract

The reflection of a bundle of laser light impinging on the warped cross section of a circular bar with an arc groove subjected to torsional moment forms a characteristic caustic on a screen. In this study, theoretical equations describing this caustic were established, a method measuring this caustic was proposed, and experiments were performed. It was shown that the theoretical curve is in good agreement with the experimentally obtained caustics. As a result, it was shown that the reflected caustics method is also useful for accurate measurement of the twisting angle and the torsional moment of the circular bar with an arc groove.

View full abstract

Effects of plastic and creep prestrain on cyclic plastic deformation are experimentally investigated. Thin walled tubular specimens of stainless steel SUS304 are subjected to combined axial and torsional loads. Prestraining is conducted at room temperature and 600°C, and subsequent cyclic loading under constant strain amplitude is performed at room temperature and 600°C. The cyclic stress-strain curves become saturated with increasing number of cycles. The properties of cyclic loading are examined based on the maximum stress values of the saturated cyclic stress-strain curves. The saturated cyclic stress subsequent to creep straining is influenced by testing temperature as well as strain amplitude. Small strain amplitudes result in large saturated stress values comparing with cyclic loading without creep prestraining. The effects of cyclic loading under combined stress state are also examined using the equi-plastic strain curves. Tensile creep straining expands the equiplastic strain curves in the directions of tensile and torsional stresses. Cyclic loading subsequent to tensile creep straining result in the expansion of the equi-plastic strain curves in the directions of tensile and compressive stresses.

View full abstract

The low-pressure plasma spray coating process has been established in gas turbines and is used for some parts, such as turbine blades and duct segments, which are exposed to corrosive gases at high temperatures. Overlay coatings based on the MCrAlY alloy system (M is Ni, Fe or Co) are commonly employed as oxidation-and corrosion-resistant coatings. Mechanical properties, such as short-time tensile strength and creep-rupture lives, of CoCrAlY-and CoNiCrAlY-coated systems were investigated at high temperature and compared with the uncoated substrate IN738LC. Tensile properties of the MCrAlY-coated IN738LC at room temperature (293K) were affected by the high strength and low ductility of MCrAlY coatings. Fowever, the tensile properties at high temperature (1123K) were not affected by the coatings for an increase in ductility of MCrAlY coatings. Also, there was no difference of creep lives between the MCrAlY-coated systems and the uncoated substrate IN738LC. This was because creep cracks initiated and grew inside the substrate in all cases and the creep crack initiation due to coatings could not be observed. It appears to be possible to select MCrAlY coatings without causing marked effects on the creep lives of the substrate at high temperature.

View full abstract

A new viscoelastic material model was developed to describe deformations under various loads at the high-temperature creep regime. In the model the inelastic strain rate is written in the form of a flow equation of the Norton-Bailey type and the material hardening during deformation under various loads is induced by an internal stress which is subdivided into back stress and friction stress. The back stress represents a conservative part of the creep resistance while the friction stress includes all the dissipative parts in the internal structure such as grain boundary sliding, diffusion and interaction between dislocation and precipitation. An inelastic constitutive equation is established based on the new viscoelastic material model. A case study was done for Hastelloy XR, a Ni-based superalloy. Deformation analyses show a good agreement between calculations and experimental results for creep, relaxation, tensile and stress dip tests.

View full abstract

Stress distributions and deformation of adhesive butt joints are analyzed by the elastoplastic finite-element method when the joints of dissimilar shafts are subjected to external bending moments. The effects of the ratio of Young's modulus of adherends to that of adhesive and the adhesive thickness on the interface stress distribution are investigated. Joint strength is predicted from using the elastoplastic interface stress distributions. As a result, it is found that the singular stress at the interface edges increases with an increase of the ratio of Young's modulus. Measurement of strains in joints and experiments on joint strength were conducted. The calculated results are in fairly good agreement with the experimental results. It is found that the joint strength of dissimilar shafts is smaller than that of similar shafts. A fracture of dissimilar adhesive shafts occurs at interface between an adhesive and an adherend which has a smaller Young's modulus. It is observed that joint strength increases as the adhesive thickness increases.

View full abstract

Propagation of stress wave and concentration of dynamic stress in tapered lap adhesive joints under impact loading were investigated analytically and experimentally. Stress distribution and time variation of stress and strain in the joints under tensile impact loading were calculated using F.E.M. considering viscoelastic properties of adhesive resin. Impact tests on the adhesive joints of aluminum alloy were conducted. Calculated results agreed with experimental results. Effects of taper length on the concentration of stress and strain were examined by comparing the calculated results with the experimental results. An increase in taper length is effective to reduce the concentrations of stress and strain in adhesive lap joints under impact loading.

View full abstract

Electron-beam welding (EBW) is judged to be a suitable welding method of alloy A286, which is an iron-base superalloy with difficult weldability. The strength of the weld of this alloy changes to be reduced by the weld heat input. However, in case of the joints with narrow weld bead, increase of the strength may be expected by the effect of plastic constraint, such as those produced by EBW. In order to clarify the welded joint geometries which are affected by plastic constraint, tensile tests with varing relative thicknesses were conducted at 293K, 77K and 4K. As a result, it was clarified that tensile strength of the joints with relative thickness less than 0.8 increased with the effect of plastic constraint, and the order of degree depending on temperature was such that 4K<77K<293K.

View full abstract

The big end of a connecting rod of medium-speed diesel engines is frequently diagonally split to reduce its width to enable easy assembly and disassembly. Accordingly, shearing forces occur along the interface between the shank and cap. To support these shearing forces, serrations which have shapes similar to screw threads are machined at both contact surfaces. It is reported that failure is sometimes observed around serrations, which seems to be caused by stress concentrations at the root of serrations and the relative displacement between mating surfaces. In this study, the strength of serrations subjected to basic loading, such as compression, shear and bending, is analyzed by FEM which deals with elastic contact problems, where the effects of the configuration of serration, pitch errors of the serration occurring during machining and coefficients of friction at contact surfaces are estimated. In addition, the effects of the distance between the serration and the blind hole in which a crank pin bolt is tapped, are also discussed by employing the axisymmetric model.

View full abstract

To study the connections of main columns of lattice steel tube transmission towers, a flange design method which ensures the strength of both the flange itself and the bolt is proposed, based on the results of experiments and FEM analyses. A stress calculation method is developed using the axisymmetric finite element method, in which the Young's moduli of ring elements of bolt and bolt hole are reduced according to the ratio of total cross-sectional area to ring element area. The applicability of the calculation method was confirmed by stress measurements of both flange and bolt in full-scale models, in which the initial induced load of the bolt was varied. Using the calculation method, the contour of flange which provides sufficient strength for use in large power-transmission towers can be determined.

View full abstract

Fatigue tests on adhesive-bonded box section beams under torsion, which are typical members of an automobile body structure, were carried out. Using the finite element three-dimensional elastostatic analysis, the stress intensity factor for interface crack was analyzed and the results of fatigue tests were tried to be arranged by the stress intensity factor range ΔK_III. Judging from the above, we conclude that it is possible unifomly evaluate the torsional fatigue strength of this beam by using ΔK_III.

View full abstract

In a horizontal well or an extended reach well, which is promising for future well construction, a well casing is subjected to local weae of wall thickness on its inside surface due to rotational contact with a tool joint or drill pipe during drilling. Burst strength of worn casing is analyzed as an unstable problem by the elastic-plastic finite-element method (FEM) and the decrease ratio of burst strength due to casing wear is formulated by nonlinear programming using parameters of the wear shape. The elastic-plastic FEM is highly applicable, since values calculated by the FEM program agree well with those of experiments. It was found that the decrease ratio of burst strength of a worn casing is linearly proportional to the decrease ratio of wall thickness, and that the apical angle of the wear shape has little influence on burst strength.

View full abstract

In a directional well, which is promising for future well construction, a well casing is subjected to bending load due to the deviated well trajectory. Collapse strength of commercial casing under bending has rarely been evaluated, making it impossible to design casing strings for a directional well. The authors performed collapse tests under four-point bending, analyzed the collapse strength as an unstable problem by the elastic-plastic finite-element method (FEM), and established an empirical formula based on Mises' yield criterion. the elastic-plastic FEM is highly applicable, since values calculated by the FEM program agree well with those of experiments. It was found that the bending load has little influence on collapse strength compared with the conventional estimation, and that the proposed empirical formula can well estimate the collapse strength under bending.

View full abstract

Automobile body structures are constructed from spot-welded thin-walled box beams. Joint stiffness of body structure is influenced by the plate thickness, service hole, spot welding around the joint, and the shape of the cross section of box beams. In this paper, an application of hierarchical neural networks to joint stiffness identification of body structure is described. First, sample data of effective parameters vs joint stiffness are calculated by the finite-element method. Second, the error-back-propagation neural network is trained using the sample data. It is found that the value of joint stiffness for effective parameters can be obtained by the trained neural networks.

View full abstract

This paper describes an inverse method to identify arbitrary boundary conditions of a beam subjected to impulsive load using the boundary element method (BEM) Generally the boundary conditions are expressed as the linear combinations with the boundary values. In present formula, the boundary conditions are simplified with two compliances about deflection and rotation, which can be obtained using a Laplace-transformed BEM with additional data from measurement of two arbitrary points on the beam. In order to confirm the validity of this procedure, deflection and bending moment are predicted with boundary conditions identified by inverse calculation, and compared with those measured directly, and it is found that the boundary conditions of beams can be identified well.

View full abstract

This report describes the transition phenomena which were found in the previous studies of a thin cylindrical vessel under vertical excitation. These phenomena had the following different processes of three frequency regions in the bifurcation diagrams on the wall surface displacement of vessel. These processes were classified into shifting from periodic motion to quasiperiodic motion, the periodic resonance of parametric excitation and the characteristics of mixing the above two motions. For a detailed investigation of trajectories and spectra, critical acceleration α_p was defined by exciting acceleration immediately before the occurrence of this transition, and α_p was a new standard that estimated the transition phenomena, On the other hand, four stringers were used to induce stiffening of the vessel. The stiffening, which increased the rigidity of the vessel, restrained the quasiperiodic motion, and therefore α_p was increased.

View full abstract

A three-dimensional microscopic model of powder compacts for viscoplastic finite element analysis is developed to investigate the deformation behavior of powder particles during sintering. The unit cells containing three types of pore: a spherical closed pore, a spherical open pore and an open pore surrounded by eight spherical particles, are analyzed. Pore shrinkage due to surface tension during the intermediate stage of sintering is simulated. The volumetric strain rate for each of the unit cell is compared with that of a spherical pore surrounded by a spherical shell and the effect of the pore shape on the sintering rate is discussed. Sintering rate is expressed as a function of relative density, viscosity and sintering stress. The expression for the shrinkage rate gives a good ht to the results obtained by finite element analysis.

View full abstract

A constitutive equation for sintering is proposed. The sintering is described as the deformation process of viscous porous materials under the action of sintering stress, the virtual hydrostatic stress produced by the surface tension. The parameters for the constitutive equation are determined from the experimental results on the sintering of glass powder. The proposed constitutive equation is applied to the prediction of the deformation behavior of glass powder compacts during sintering using the viscoplastic finite element method. The effect of powder particle size on the shape of the sintered body is examined. The sintering rate of the glass powder compact composed of small particles is high and the shape of the sintered body is not significantly affected by its own weight. The predicted shapes of the sintered bodies show good agreement with those obtained by experiment.

View full abstract

Computer simulations are increasingly replacing experiments in various fields, and the required scale of the models to be analyzed has become extremely large. To solve such large-scale problems, various parallel FEM algorithms have been investigated to date, Among them, the domain decompo-sition method (DDM) is suitable for parallel computing because of small memory requirement, arbitrary decomposition of the domain, and easy implementation on parallel systems. In this study, we applied a DDM algorithm to dynamic FEM analyses, and implemented it on a network which consists of several engineering workstations. It is well demonstrated through numerical examples that the present DDM-based dynamic FEM code shows very high parallel efficiency and fault tolerance.

View full abstract

This paper presents an efficient inverse analysis technique for the identification of an unknown boundary condition in a finite element method (FEM) on a small memory capacity machine such as a personal computer. The author has previously presented the FEM of inverse analysis introducing an inverse compliance matrix that uses an influence function into the stiffness matrix for the unknown boundary. Since the stiffness matrix of the FEM becomes singular, following two calculation processes are adopted: (1) a calculation process which obtains an approximate solution by decreasing the effect of the inverse compliance in the stiffness, and (2) an iterative calculation process computing a correct solution from the approximation using (1). First, a penalty function method is used to decrease the rank of a singular stiffness matrix in (1). Then, the number for computing LDU decomposed matrix becomes twice in comparison with fourth times in the previous paper. Next, an efficient calculation method is presented using an attack technique based on both accelerative moving lower and upper bounds in (2). Moreover, a zooming identification method is developed, owing to an improvement of node displacements on and near the edges. Therefore, this technique makes it possible to establish an efficient inverse analysis for the identification of the unknown boundary. Finally, the validity of the inverse analysis technique is confirmed from the numerical solution of a three-dimensional inverse finite element analysis model, as an example of an elastic structure with reinforced members under tension, carried out on a small memory capacity machine.

View full abstract

An optimization algorithm of a frame structure subjected to a plastic deformation is presented in this paper. This method is based on the generalized layout optimization method proposed by Bendsφe and Kikuchi in 1988, where a solid-cavity composite material is distributed in an admissible domain and cavity size is determined so that it becomes large in the area where strain energy is small. In the present algorithm, a material is distributed inside frame members so that the external work done by prescribed loads may become minimum. As a result, some regions inside frame members may be hollow while others may be solid.

View full abstract

An optimization algorithm of a frame structure subjected to a plastic deformation was presented in the previous report. It is based on the generalized layout optimization method proposed by BendsΦe and Kikuchi in 1988, where a solid-cavity cmposite material is distributed in an admissible domain and cavity size is determined so that it becomes large in the area where the strain energy is small. In this report, the present algorithm is verified by several numerical examples in which it is shown that the object function is highly improved by the optimization and converges rapidly. The optimization algorithm including the effect of shear deformations is also discussed for frame structures with fairly short members and a numerical example is shown.

View full abstract

This study deals with the methodology of structural design with a constraint on system reliability. A structure with high redundancy usually possesses a number of failure modes, which results in difficulty in evaluation of system reliability. Two optimum design problems are formulated with reliability constraints on elements and failure modes, respectively. The constraint on system reliability is not directly introduced into the design space, but it is satisfied through control of the constraints on element reliabilities or on the formation probabilities of failure modes in optimization. Sensitivity analysis on element reliabilities is performed to reduce the computation effort in the design process. It is shown from numerical results that the proposed method is effective for designing structures at specified system reliability levels.

View full abstract

In recent years, genetic algorithms have been used as an optimization technique in various fields. In a genetic algorithm, practical problems are described in terms of chromosomes. Chromosomes in living organisms are made up of double-helix DNA and previous genetic algorithms idealized them as one-dimensional in general. If a problem is subject to complicated restrictions or is originally a matrix type, a multidimensional chromosome is more effective than a one-dimensional chromosome. In this paper, a genetic algorithm using multidimensional chromosomes is proposed and applied to the layout problems of members in a truss structure and ribs in a thin plate structure.

View full abstract

The movements of the condyle with maximum mouth opening were measured by image processing of tomograms. The reference point of the condyle was determined and the distance between the reference points at closed and open positions was measured. The trajectory of the condyle during mouth opening was also analyzed by considering the shape of the temporomandibular joint. The calculated results of the condylar movements agreed with ones obtained by the image analysis. It is thought that the method presented in this paper is useful in diagnosis of temporomandibular joint dysfunction, and the condylar movements are evaluated directly from the tomograms.

View full abstract