In this paper some results of the computer simulation of mixing of non-Newtonian fluid are presented. Numerical calculations were done for dimensionless form of the equations of motion for the Carreau fluid in incompressible and viscous flow in a two-dimensional semi-cylindrical cavity. The full Navier-Stokes equations for the Carreau fluid were written in velocity-vorticity, and next in finite difference form. The solutions were accomplished by finite time-step advancement. The mixing process was studied by tracking the motion of particles in the mixing region. The particles were represented by marked points. The mixing efficiency was quantified in terms of the average distance between the tracer particles and the centroid of the particle distribution.

A three-dimensional rest-to-rest attitude maneuver of a satellite with flexible solar panels equipped by on-off reaction jets is studied. Results indicate that, under an unshaped input, the maneuver induces an undesirable motion of the satellite as well as vibration of the solar panels. Fuel-efficient input shapers are then applied to reduce the residual oscillation of its attitude. By reducing vibrations at several large-amplitude natural frequencies, the expected pointing precision of the satellite can be satisfied.

This paper describes the phenomena that occur when a simplified model of train interacts with the tunnel at three locations - before, entering and leaving the tunnel. The Navier-Stokes equation is solved by introducing the artificial compressibility to change the governing equation type from the elliptic to hyperbolic. The Baldwin-Lomax turbulence modeling is employed to simulate the flow field with a Reynolds number of 10^6, and the computation domain is divided into three blocks considering the train and tunnel geometries. The grid is algebraically adapted determining the maximum solution change plane and solution weighting factors. The pressure in the adapted solution is not changed much, however, the skin friction is severely varied comparing with those of the non-adapted solutions. When the train enters into the tunnel, there are large increase in the surface pressure and skin friction distribution on the train surface.

In the paper some results of investigations of two intelligent information systems: a feedforward neural network and an adaptive fuzzy expert system, are presented. The systems can be used for example in approximation and control problems or in diagnostics. The adaptive fuzzy expert system is constructed as a hybrid in which a fuzzy inference system is combined with a neural network. In the learning process for given set of training points an optimal value of the so-called generalized weight vector is searched. The Lapunov theory is used to examine the non-sensitivity of the optimal value of a generalized weight vector to initial conditions and training data. Some necessary and sufficient conditions are formulated in terms of the Hessian matrix of the error function.

The analysis of multi-leaf structures should be performed taking friction into account. The objectivity of friction law should be preserved because large deformation generally occurs. By use of the convected coordinate system, the objectivity can be preserved naturally. Therefore, in finite element analysis, the element local coordinate system can be used. However, when a contact point slides over the element boundary, a problem arises due to the discontinuity of the local coordinates between elements. In this work, an algorithm is proposed, i.e., the formulation is essentially based on the convected coordinate system while the sliding term is redefined as a spatial vector and is calculated in the reference configuration. Thus, the finite sliding due to large deformation can be treated without paying special attention to the limit of the local coordinate system. Two numerical examples including a simplified model of a leaf spring structure used in nuclear power plants are given.

Although there is a need for sensitivity analysis for frictional contact problems in many engineering fields, research work on it has rarely been reported due to the complexity of the problems. In this paper, a sensitivity analysis procedure based on a semi-analytical method for frictional contact problems is presented. The unbalance force due to the variation of design parameters is evaluated numerically, thus the related routine of an existing FEM code can be utilized regardless of the friction law employed. The continuum-mechanics-based formulation is carried out first and then a discretized form is derived. The stick state is modeled by introducing a penalty-type constraint. A couple of numerical examples, including a realistic leaf spring structure used in nuclear power plants, are given to demonstrate the effectiveness of the proposed approach.

A numerical method based on compliance approach is presented for analyzing an isotropic homogeneous sheet enclosing a crack. The method calculates the strain energy release rate and determines the stress intensity factors K_{I} and K_{II}. This method is suitable for any load combination in pure mode I, pure mode II and mix mode loading. A simple and efficient solution approach is developed in which the strain energy release rate is calculated by combining the finite element method with the fundamental relationships in fracture mechanics. The solution technique converges to accurate results for a small crack extension of the finite element mesh. The solution approach is also shown to be suited for separating the mode I and mode II stress intensity factors for a mixed mode loading. Numerical examples are presented to demonstrate the accuracy of the proposed approach.

The main objective of this study is to design three-dimensional geometrical and mechanical finite element model of the intervertebral disc between L2-L3 vertebras in the lumbar and C5-C6 cervical spinal segment. The model was directed toward understanding the work and the role of the intervertebral disc that performs in the human spinal segment body. The three-dimensional finite element motion segment was developed and its response to different loads was performed. The model accounted for the solid component of the nucleus pulposus while anulus fibrosus was modeled as a matrix of homogeneous ground substance reinforced by anulus fibers. End plates similarly to the nucleus pulposus were simulated using volumic elements. Simultaneously the vertebral bodies have been modeled as a complex construction of a cancellous core covered by a cortical shell of the orthotropic material properties. Isotropic material has been used to model posterior elements. To simulate ligament like behavior, tension only elements have been used. Numerical studies of the lumbar segment have been consequently compared with the experimental investigation performed on the porcine model by authors and other in vitro experiments on human lumbar spine accomplished by other scientists. In cervical spinal segment numerically two surgical techniques (Cloward and Robinson-Smith) have been tested. Two types of loads have been applied to three models - to an intact C5-C6 spinal segment and then to the vertebras after performing those two surgical techniques. All numerical analysis have been undertaken using ANSYS 5.4 commercial application.

The purpose of this polyoptimization was to find optimal tubular cross section catalogues for the specified spatial trusses. Five different spans of trusses have been analyzed, starting from 24x24 m ending at 72x72 m, trusses supported at every other perimeter node, with loading conditions typical for roofs. Four decision variables were taken under consideration, i.e. the catalogue size t, its arrangement T in a metallurgical catalogue T_M, minimal and maximal diameters of tubular elements D_{min} and D_{max}. Two objective functions: truss mass and manufacturability of particular solutions were evaluated. For that purpose, some design, technological, and computational constraints were taken into account. For the specified spans L, different catalogues of cross sections were defined by TOPSIS method.

The main objective of the paper is the investigation of localization phenomena in thermo-viscoplastic flow processes under cyclic dynamic loadings. Recent experimental observations for cycle fatigue damage mechanics at high temperature and dynamic loadings of metals suggest that the intrinsic microdamage process does very much dependent on the strain rate and the wave shape effects and is mostly developed in the regions where the plastic deformation is localized. The microdamage kinetics interacts with thermal and load changes to make failure of solids a highly rate, temperature and history dependent, nonlinear process. A general constitutive model of elasto-viscoplastic damaged polycrystalline solids is developed within the thermodynamic framework of the rate type covariance structure with finite set of the internal state variables. A set of the internal state variables is assumed and interpreted such that the theory developed takes account of the effects as follows: (i) plastic non-normality; (ii) plastic strain induced anisotropy (kinematic hardening); (iii) softening generated by microdamage mechanisms (nucleation, growth and coalescence of microcracks); (iv) thermomechanical coupling (thermal plastic softening and thermal expansion); (v) rate sensitivity; (vi) plastic spin. To describe suitably the time and temperature dependent effects observed experimentally and the accumulation of the plastic deformation and damage during dynamic cyclic loading process the kinetics of microdamage and the kinematic hardening law have been modified. The relaxation time is used as a regularization parameter. By assuming that the relaxation time tends to zero, the rate independent elastic-plastic response can be obtained. The viscoplastic regularization procedure assures the stable integration algorithm by using the finite difference method. Particular attention is focused on the well-posedness of the evolution problem (the initial-boundary value problem) as well as on its numerical solutions. The Lax-Richtmyer equivalence theorem is formulated and conditions under which this theory is valid are examined. Utilizing the finite difference method for regularized elasto-viscoplastic model, the numerical investigation of the three-dimensional dynamic adiabatic deformation in a particular body under cyclic loading condition is presented. Particular examples have been considered, namely dynamic, adiabatic and isothermal, cyclic loading processes for a thin steel plate with small rectangular hole located in the centre. To the upper edge of the plate the normal and parallel displacements are applied while the lower edge is supported rigidly. Both these displacements change in time cyclically. Small two asymmetric regions which undergo significant deformations and temperature rise have been determined. Their evolution until occurrence of final fracture has been simulated. The accumulation of damage and equivalent plastic deformation on each considered cycle has been obtained. It has been found that this accumulation distinctly depends on the wave shape of the assumed loading cycle.