Year:2003 | Volume:37 | Issue:3 (81)|Papers Count:14

A semi-Lagrangian algorithm is associated with semi-implicit method in the integration of the shallow water equations. The resulting numerical algorithm is unconditionally stable. To obtain accurate results from a semi-Lagrangian integration scheme, it is necessary to choose the order of interpolation carefully. Cubic interpolation is used and it is a good compromise between accuracy and computational cost. The semi-Lagrangian semi-implicit scheme is applied to the shallow water equations with two kinds of boundary conditions, velocity component type or geopotential height type. When the boundary values are imposed on the velocity component some new relations are obtained using governing equations to solve geopotential height equation numerically. The integral invariants of the shallow water equations, i.e., mass, total energy and entropy are well conserved, ensuring that a realistic nonlinear structure is obtained.

The objective of the present work is to develop a highly accurate code for numerical solution of the conservation-law form of the shallow water equations in the beta plane. The second-order delta formulation of the trapezoidal time differencing scheme is used. The Hirsch fourth-order compact finite-difference method is also applied to discretize the spatial factored form of the equations obtained using the ADI method. Because of the large aliasing error introduced by the fourth-order scheme a very selective low-pass filter is used to overcome the error generated by the interaction of the nonlinear terms of the equations. It was found that the best results can be obtained by a periodical application of the filter. The integral invariants of the shallow water equations, i.e. the total energy and entropy are well conserved during the numerical integration. This fact shows that the nonlinear structure of the equations is correctly modeled. The validation of the fourth-order compact results is investigated by comparing them with an accurate nonlinear ADI method performed by Gustafsson. In addition, a comparison is made between the results of the fourth-order compact scheme and a second-order finite-difference method (developed by authors) for different grid resolutions to confirm the high accuracy of the compact scheme.

An analytical solution for forging processes using an elemental upper bound method was given in this paper. The analysis was carried out by dividing the cross section into some standard pre-defined elements. For each of these elements the velocity field was defined. When the cross section was divided into the elements, using the cinematically admissible velocity field of each element the boundary conditions the general cinematically admissible velocity field for the deforming region was completely defined. Using the general cinematically admissible velocity field upper bound solutions were arrived at for both plane strain and ax symmetric forging processes. For the case of plane strain forging processes rectangular and trapezoidal elements were employed. Ring type elements of triangular and square cross sections were used for the analysis of ax symmetric forging problems. In this work the velocity fields were improved by modifying the previous work equations. Based on the above theory a computer program was written with the use of which simulation of both plane strain and ax symmetric forgings were presented. The influence of different process parameters such as friction factor, flash and geometrical dimensions were considered. The results obtained from this work were compared with previous work and improvements were observed.

The handling quality of a passenger car is noticeably influenced by the structural and functional characteristics of the individual components of the vehicles various mechanisms, including the steering, suspension and braking systems. This implies that in designing each of these mechanisms, its contribution to the dynamic behavior of the vehicle should be taken into account. In this paper the sensitivity of the vehicles handling quality to the geometry of a rack-and-pinion steering mechanism, specially to the coordinates of its joints, is analytically studied and a new Genetic-based approach the determination of the joint coordinates is proposed. In the proposed approach, the dynamic equations and kinematics constraints of the steering mechanism are derived and incorporated in a dynamic model of the vehicle. The model is then solved for a step-input to the steering wheel, using MATLAB, and the values of the handling parameters are calculated. An Elitist Genetic Algorithm is used to search the hyper-space of the joint coordinates and to find the ones resulting in the best handling quality. The advantages of the proposed method over the classical solution methods are also discussed.

In this paper pin-fin cooling by mixed convection has been considered. The system consisting of a pin-fin heat sink and a chimney is presented for increasing heat transfer and draft force. The flow regime is laminar. The results is applied to problems in which the size of the overall system is constrained. For a given heat dissipation and total system size optimal values of the pin-fin diameter and heat sink porosity are obtained. It is shown hat the optimum occur for systems with and without chimney. The pin-fin array is modeled as a Forchheimer-flow porous medium (Non Darcy-flow). The results of optimization are compared with existing data in literature. Also it is shown that the results obtained grom Forchheimer-flow porous model agree much better with the experimental results compared with the natural convection dary model.

In the present paper a new scheme is presented which combines the Sequential Function Specification and the Dual Reciprocity Boundary Element Methods. In this scheme the unknown boundary condition is estimated sequentially by using two transformation matrices. The matrices defined in the direct heat conduction calculations by the Dual Reciprocity Boundary Element Method are used as a basis for the definition of the transformation matrices, and the mathematical derivations for the inverse estimation are in accordance with the Sequential Function Specification of Beck. In order to compare the speed and the accuracy of the method with the existing Sequential Function Specification Method, the exact analytical solution of a simulated test case in utilized. Results indicate an impressive improvement in the efficiency as well as the accuracy of this scheme as compared with the conventional ones. The results are accurate and stable foe error amplitudes up to one percent of measured temperatures.

Using underwater shock waves produced by underwater explosion of high explosives (HE) has been studied, as a manufacturing process of metals, for many years. Although early studies on this subject had mainly military motivation, many investigations have recently been conducted in other industrial fields. The main aim of these research are to understand the details of the motion of spherical bubble produced by HE explosion as well as the subsequent flow field generated by the motion of a strong shock wave in the water. Due to the limitation of computing facilities, the early numerical simulations of underwater explosion phenomenon were performed by crude numerical methods (e.g., Stenberg et al. 1971). In those works the linear artificial viscosity of von Neumann was used to prevent the numerical instability close to the shock (the so-called q-method). Flores and Holt (1981) applied Glimms method to underwater explosion of an uniform spherical bubble. Their study showed the capability of new methods to study the problem. Molyneaux et al. (1996) used the finite element program DYNA3D and observed that numerical tools are able to predict well both the magnitude and form of the pressure transient field. Their simulation was performed for a cylindrical charge. In the present study, a third order Godunov method has been utilized in the framework of a Lagrangian approach to investigate the entire flow field generated by underwater explosion of a spherical bubble. The developed code is based on the Piecewise Parabolic Method (PPM) of Collela (Colella et. al.1984). The original PPM which was developed for ideal gases, here is extended to treat the real gas effect. The water and the explosion products were treated using Mie-Gruneisen and JWL equation of states, respectively. The processes of shock and expansion waves reflection from the origin and interaction with the gas/water interface are fully resolved with the present method.

Nonlinear analysis of a structure that includes two connected circular and annular thin plates under uniform loads has been performed by shooting method. Plates are connected by linear tensional spring. The von Karman equations with corresponding boundary conditions (clamped-clamped) have been solved by employing parallel shooting method. The variation of displacements, bending and membrane stresses of plates under variety of loading intensity are obtained. Finally the variation of maximum stresses due to different parameters such as spring tensional rigidity, plates thickness ratio and ratio of annular inner radius to circular outer radius are considered

Material selection of tool steels for satisfying the customers needs is not limited to one type of steel. In fact, each customer expresses his or her needs in a specific range. On the other hand, technical specifications of tool steels are not absolute and it consistently deviates from the mean. Therefore in this paper, fuzzy logic theory has been used for selecting appropriate tool steels. The main steps follow as:1)Four technical features have been used for selecting tool steels: untreated hardness, hardness after heat treatment, maximum work temperature and toughness, and price variable.2)Technical features and prices of tool steels manufactured in Iran were obtained from manufacturers.3)The mean and standard deviation of some technical features were calculated.4)Membership functions for the a.m. technical features and prices have been defined and established.5)The tool steels were rated according to the amount of their favourability to customers.6)Customers desirable tool steels were determined by using fuzzy logic theory. Finally, the appropriate tool steels were selected for customers.In this paper, in addition to formulate the tool steel selection, the possibility of reducing a variety of tool steels for manufacturers was also provided which the associated result leads to the decrease in the total cost of tool steels for the manufacturer.

The formation of columnar structure in Alnico alloys enhances markedly its magnetic properties. The addition of Ti to these alloys also improves the magnetic properties but inhabitants formation of columnar structure. By addition of some elements such as C, S, Se, Te or P the columnar structure could be obtained again. In this work, the effects of C and S on the formation of columnar structure have been studied. Experimental results and thermo dynamical analysis show that carbon by deoxidation of the melt results in not formation of titanium oxide inclusions and sulfur precipitates as titanium sulphide on the surface of titanium nitride inclusions. Hence nucleation effects of titanium inclusions which makes the structure non columnar is controlled.

As is well accepted, any heavy duty parts of a vehicle, such as suspension and transmission systems, should be made of special grades of micro alloyed steels. In addition the experience is shown that the temperature of these parts may locally rise up to about 200 °C. In the present work, relying on the in-house steel making potential, a series of vanadium microalloyed steels were prepared and successively heat treated to obtain different desired multi-phase microstructures. The related fatigue properties were then examined to investigate the effects of material and process variables. The results indicate that the increase of temperature would result in strain aging phenomena and this in turn may lead to higher fatigue life in ferrite-marten site dual phase microstructure.

The objective of the present work was to study the preparation a fully dense MgO+C dopped alumina. The sintered materials was characterized by their microstructure and the microstructure variation during the different thermal treatment. The atmosphere is suggested to play an important role during the annealing of MgO+C dopped Al2O3. From the results it can be seen that MgO+C increased the densification rate of Al2O3 and the density is arrived very near the theoritical density. The Vaccum secondary and the absence eventually gas in the pores have been played in the densification rate. The other role of MgO+C in powder is to lower the boundary mobility which helps to keep the pores attached to the moving boundary even though the pores are controlling grain growth and not to change the grain size/density significantly. That is to say, the pesence of carbon or MgO in powder is to inhibit grain growth and to promote the development of more uniform grain structure by formation the precipitate and segregation. Because of the above roles of MgO+C in Al2O3, it is possible to arrive the theoritical density without the abnormal grain.

One of the most problem during solidification of ductile iron is segregation of alloying element. Segregation of alloying element decreases mechanical properties of ductile iron and austempered ductile iron. In this research, alloying element profile as a function intergraphite distance was studied. Homogenization of ductile iron using partial melting is carried out, for first time, to homogenize the distribution of alloying element in the matrix of ductile iron. Results of image analyzer for homogenize and austember specimen showed that homogenization using partial melting against conventional homogenization, is very effective to uniform microstructure of ductile iron. The results show that morphology of graphite before and after homogenization using partial melting has minor effect on the morphology of graphite.