International Journal of Computational Engineering Science

The elasto-viscoplastic constitutive equation considering damage is formulated, based on the concept of continuum damage mechanics. It employs the viscoplastic strain, which was originally given by Perzyna and extended by Murakami to consider the effect of damage. The unified form of damage evolution equation given by Lemaitre is extended to consider the effect of damage type and strain rate. The constitutive modeling is identified, based on static/dynamic tensile tests and fatigue tests for steel (SM490A) and aluminum (2219-T87). The identified models are used to predict the static/dynamic, tensile behaviors of pre-strained steel and pre-fatigued aluminum. The predicted results have agreed well with the corresponding experimental results.

Parallel fast multipole method is developed for radar cross section (RCS) computation of large targets on multiple processor systems. The sparse matrices in fast multipole method (FMM) are calculated and stored in distributed processors and message passing interface (MPI) is used to perform the matrix-vector product in parallel iterative solvor. The accuracy of the code is verified by comparison with the results of benchmark targets. Either the computational complexity or speed-up ratio of parallel FMM and MoM are compared.

Models of acoustic scattering that predict the echo time history of targets are essential tools for the development of signal processing algorithms in underwater active detection systems. Realistic simulation of echoes requires a multi-frequency evaluation of acoustic scattering. This can be achieved by numerically modeling the surface of the object of interest as a collection of facets and calculating the scattered field using the Helmholtz-Kirchhoff approximation. Time and frequency domain analysis of a simple object (e.g. a sphere) and a complex structure using this technique in monostatic and bistatic configurations are given as examples.

A lattice Boltzmann method on nonuniform mesh is developed for the simulation of incompressible flows. The computation mesh is uncoupled from the lattice. The particle distribution function on the mesh is obtained using the Lagrange interpolation technique at each time step. Comparison with the finite difference solutions for the two-dimensional driven cavity flow and the experimental results for flow around a circular cylinder shows that the present scheme not only can predict the flow separation characteristic satisfactorily with much less mesh but also increase the computational stability. The influence of the mesh structure and interpolation on the results will also be discussed.

The Multi Priority Dual Queue (MPDQ) is a new scheduling regime shown to reduce congestion for multi-class traffic over conventional scheduling disciplines in isolation. As the MPDQ improves traffic congestion, most notably in communication systems, it is not necessary to include it at every node, or service centre, within a network. To gain insight into the advantages of a mixed MPDQ / First In First Out network with prioritised traffic, simulations are performed with and without a dual queue and various Quality of Service (QoS) criteria are obtained. Queueing networks containing differentiated traffic, also known as multi-class networks, are complicated to solve analytically using existing queueing theory techniques. We discuss the complications of exact solutions and describe how we used simulation to obtain performance statistics. We also look at high-class traffic under different offered loads, and provide a comparison of the delay characteristics. These findings provide communication service providers valuable information in determining the improvement in QoS to differentiated networks with the MPDQ. They also highlight the importance of simulation as a tool of evaluation of networks, and the best MPDQ network scenario.

This paper presents the dynamic response of structures under high velocity impact loading using Smooth Particle Hydrodynamics (SPH) approach. The SPH equations governing the elastic and elasto-plastic large deformation dynamic response of solid structure are derived. The proposed additional stress points are introduced in the formulation in order to mitigate the tensile instability inherent in the SPH approach. The incremental rate approach is combined with the leap-frog scheme of time integration forming solution algorithm adopted in present study. Examples on high velocity impact of the solids are presented and results from the proposed SPH approach compared with available finite element solution illustrating that the transient dynamic response under high velocity impact can be effectively solved by the proposed SPH approach.

This paper presents an algorithm for solving a class of large scale nonlinear programming which is originally derived from the multi-stage stochastic convex nonlinear programming. With the Lagrangian-dual method and the Moreau-Yosida regularization, the primal problem is transformed into a smooth convex problem. By introducing a self-concordant barrier function, an approximate generalized Newton method is designed in this paper. The algorithm is shown to be of superlinear convergence. At last, some preliminary numerical results are provided.

Numerical methods such as the finite element analysis (FEA) are revolutionizing the conventional trial-and error methods for laser beam scanning industry today. These methods have proven to be very useful in product development and process design. With simulation procedure being proposed, 3D numerical simulations on a micro-scale suspension flexure due to laser beam scanning have been carried out, which includes a non-linear transient indirect coupled thermal-mechanical analysis accounting for the temperature dependency of the thermal and mechanical properties of materials. From the simulation results, relationships between the permanent bending angle of the flexure and the scanning parameters are established and this will help to predict the bending angle of the micro-scale sheet metal under any other scanning condition.

The present work considered fluid circulation in an annular region between two cylinders with the inner cylinder heated and rotating. The Prandtl number (Pr) considered here varies from 0.01 to 1.0 and Rayleigh number (Ra) is of the order of 10⁵. Reynolds number Re in the range of 0 to 1120 was considered. Mono-thermal plume above the stationary inner was observed for higher Prandtl number fluids (Pr≫0.1) while bi-thermal plume above the stationary inner cylinder was observed for lower Prandtl number fluids (Pr≪0.1). However, when the inner cylinder is made to rotate, the thermal plume for higher and lower Prandtl number fluids were observed to move in different directions. The mechanism of the mono- and bi-thermal plumes movements were investigated through numerical flow simulations.

This paper presents a methodology/model for elastro-plastic analysis of the wire bonding packages. The methodology is based on finite element method for rate sensitive materials and applicable to very large deformation processes. Ultrasonic power is simulated as displacement boundary condition. Stress induced by the mismatch of the coefficient of thermal expansion (CTE) in the system because of temperature increase due to application of the ultrasonic or the reflow process is considered. The effect of temperature increase on material property is also accounted for. Nonlinear spring elements are adopted to simulate the gradual formation of intermetallic bonding layer between the ball and the electric pad. With these spring elements, the effect of stress and strain relaxation of the ball in the subsequent procedures such as reflow process is taken into account. The mechanical behavior of the silicon chip during the wire bonding with the ultrasonic irradiation and reflow process is thus revealed. Failure probability with reference to the stress distribution is discussed. The developed FE methodology is simple but effective to evaluate the failure probability of wire-bonding packages.

In this paper we consider Prandtl's boundary layer problem for incompressible laminar flow past a plate with transfer of fluid through the surface of the plate. When the Reynolds number is large the solution of this problem has a parabolic boundary layer. In a neighbourhood of the plate the solution of the problem has an additional singularity which is caused by the absence of the compartability conditions. To solve this problem outside nearest neighbourhood of the leading edge, we construct a direct numerical method for computing approximations to the solution of the problem using a piecewise uniform mesh appropriately fitted to the parabolic boundary layer. To validate this numerical method, the model Prandtl problem with self-similar solution was examined, for which a reference solution can be computed using the Blasius problem for a nonlinear ordinary differential equation. For the model problem, suction/blowing of the flow rate density is v₀(x)=-v_i2^-1/2Re^1/2x^-1/2, where the Reynolds number Re can be arbitrarily large and v_i is the intensity of the mass transfer with arbitrary values in the segment [-.3,.3]. We considered the Prandtl problem in a finite rectangle excluding the leading edge of the plate for various values of Re which can be arbitrary large and for some values of v_i, when meshes with different number of mesh points were used. To find reference solutions for the the velocity components and their derivatives with required accuracy, we solved the Blasius problem using a semi–analytical numerical method. By extensive numerical experiments we showed that the direct numerical method constructed in this paper allows us to approximate both the solution and its derivatives Re–uniformly for different values of v_i.

A multiresolution curve representation, based on wavelets, provides more flexibility for curve smoothing, data compressing, and editing at different resolution levels. It requires no extra storage apart from that of the original control points. This paper presents the technique for wavelet-based multiresolution representations of C² continuous closed cubic B-spline curves. Due to the requirement on continuity at the start/end point, closed B-spline curves need special processing when wavelets are applied to decompose or reconstruct them. The method for multiresolution editing of closed B-spline curves is also introduced. Users can edit the overall shape of a closed curve while preserving its details, or change its details without affecting its overall shape. The corresponding algorithms have been implemented and some examples are given to illustrate the editing of C² continuous closed cubic B-spline curves at multiresolution levels.

Equal channel angular extrusion (ECAE) is a promising technique for producing ultra-fine grained (UFG) or nanostructured materials based on the principle of simple shearing. Through analysis, it is shown that only the geometrical factor Φ, namely, the half-angle of the two intersecting channels, and the number of ECAE passes, N, affect the effective strain. The equivalent linear reduction ratio, r₀/r₁, is derived to describe the size reduction effect of an object such as a grain. The most effective intersecting angle (2Φ) is 90°. Compared to traditional area reduction extrusion, the deformation effect is equivalent to an area reduction ratio of 1 million or a linear reduction ratio of 1022 after 12 passes of ECAE. Magnesium AZ31B, aluminium 6061 and commercially-pure Titanium were used for the study. Three types of die designs for ECAE of each alloy were proposed and numerically analysed. The effective strain, von Mises stress, equivalent area reduction ratio and equivalent linear reduction ratio were compared for the three types of die designs based on the simulation results using ANSYS/LS-DYNA. The parameter N_μ→nm, namely, the number of passes of ECAE required to reduce 100μ structures into 100nm structures, was calculated for each design. A grain size of 100μ can be deformed into a nanostructure through as few as 12-17 passes of ECAE.

Fluid structure Interaction (FSI) is important in analysis of: (1) Cardio-vascular dynamics; (2) Underwater/ Offshore structures; (3) Aircraft wings and turbine blade designs; (4) Design of tall structures/buildings; (5) high speed hard disk drives etc. At present, such capabilities are being incorporated, if any, only to a limited extent in commercial CFD codes leaving the engineers to develop their own codes. In the current study, a hydro-elastic problem of an underwater structure has been considered. Traditionally, the hydrodynamics and the structural dynamics are solved using finite difference/boundary element methods and finite element method (FEM) respectively. Moreover, the governing equation for fluid and structure are usually written in Eulerian and Lagrangian reference frames - posing further difficulties for coupling the two systems. In this research, both the fluid and structural systems are solved by the finite element method using a mixed Eulerian-Lagrangian scheme, where, fluid mesh moves and adapts to new free surface and structural positions. A full non-linear free surface implementation is considered. A mesh adaptation using Laplacian smoothing is performed to reduce the need for re-meshing the domain frequently. The scheme is validated with solutions available in the literature and extended to the present FSI problem with non-linear free surface boundary condition.

In this paper, the Wilson-θ integration method is thoroughly studied. By using Wilson-θ integration method and single particle shear model, the dynamic response of brick-concrete structure is given. The simulation results of dynamic response of brick-concrete structure under sine wave load are compared with analytical results. From this study, it demonstrated that the present approach of combining Wilson-θ integration method and single particle shear model method is a useful technique in the dynamic and vibration analysis of brick-concrete structure. Furthermore, by optimizing and selecting suitable parameters of Wilson-θ method, this approach also can be used to analyze the seismic dynamic response analysis of the brick-concrete structure.