An application of quasi-Newton methods for the numerical solution of interface problems

Quasi-Newton methods are applied to solve interface problems which arise from domain decomposition methods. These interface problems are usually sparse systems of linear or nonlinear equations. We are interested in applying these methods to systems of linear equations where we are not able or willing to calculate the Jacobian matrices as well as to systems of nonlinear equations resulting from nonlinear elliptic problems in the context of domain decomposition. Suitability for parallel implementation of these algorithms on coarse-grained parallel computers is discussed.

Image denoise by fourth-order PDE based on the changes of laplacian

Fourth-order partial differential equation (PDE) proposed by You and Kaveh (You-Kaveh fourth-order PDE), which replaces the gradient operator in classical second-order nonlinear diffusion methods with a Laplacian operator, is able to avoid blocky effects often caused by second-order nonlinear PDEs. However, the equation brought forward by You and Kaveh tends to leave the processed images with isolated black and white speckles. Although You and Kaveh use median filters to filter these speckles, median filters can blur the processed images to some extent, which weakens the result of You-Kaveh fourth-order PDE. In this paper, the reason why You-Kaveh fourth-order PDE can leave the processed images with isolated black and white speckles is analyzed, and a new fourth-order PDE based on the changes of Laplacian (LC fourth-order PDE) is proposed and tested. The new fourth-order PDE preserves the advantage of You-Kaveh fourth-order PDE and avoids leaving isolated black and white speckles. Moreover, the new fourth-order PDE keeps the boundary from being blurred and preserves the nuance in the processed images, so, the processed images look very natural.

Coupled 3-D finite difference time domain and finite volume methods for solving microwave heating in porous media

Computational results for the microwave heating of a porous material are presented in this paper. Combined finite difference time domain and finite volume methods were used to solve equations that describe the electromagnetic field and heat and mass transfer in porous media. The coupling between the two schemes is through a change in dielectric properties which were assumed to be dependent both on temperature and moisture content. The model was able to reflect the evolution of temperature and moisture fields as the moisture in the porous medium evaporates. Moisture movement results from internal pressure gradients produced by the internal heating and phase change.

A cell-centred finite volume method for modelling viscoelastic flow

An unstructured cell-centred finite volume method for modelling viscoelastic flow is presented. The method is applied to the flow through a planar channel and the 4:1 planar contraction for creeping flow of an Oldroyd-B fluid. The results are presented for a range of Weissenberg numbers. In the case of the planar channel results are compared with analytical solutions. For the 4:1 planar contraction benchmark problem the convection terms in the constitutive equations are approximated using both first and second order differencing schemes to compare the techniques and the effect of mesh refinement on the solution is investigated. This is the first time that a fully unstructured, cell-centredfinitevolume technique has been used to model the Oldroyd-B fluid for the test cases presented in this paper.

Experience of using a CFD code for estimating the noise generated by gusts along the sun-roof of a car

The problem to be examined here is the fluctuating pressure distribution along the open cavity of the sun-roof at the top of a car compartment due to gusts passing over the sun-roof. The aim of this test is to investigate the capability of a typical commercial CFD package, PHOENICS, in recognising pressure fluctuations occurring in an important automotive industrial problem. In particular to examine the accuracy of transporting pulsatory gusts traveling along the main flow through the use of finite volume methods with higher order schemes in the numercial solutins of the unsteady compressible Navier-Stokes equations. The Helmholtz equation is used to solve the sound distribution inside the car compartment, resulting from the externally induced fluctuations.

Kinematics and EMG analysis of expert pole vaulter's lower limb during take off phase (P218)

Introduction: The critical phase, in jumping events in track and field, appears to be between touchdown and take-off. Since obvious similarities exist between the take off phase in both long jump and pole vault, numerous 3D kinematics and electromyographic studies have only looked at long jump. Currently there are few detailed kinematics electromyographic data on the pole vault take-off phase. The aim of this study was therefore to characterise kinematics and electromyographic variables during the take-off phase to provide a better understanding of this phase in pole vaulting and its role in performance outcome. Material and methods: Six pole-vaulters took part in the study. Kinematics data were captured with retro reflective markers fixed on the body. Hip, knee and ankle angle were calculated. Differential bipolar surface electrodes were placed on the following muscles of the take-off leg: tibialis anterior, lateral gastrocnemius, vastus lateralis, rectus femoris, bicep femoris and gluteus maximus. EMG activity was synchronously acquired with the kinematic data. EMG data were rectified and smoothed using a second order low pass Butterworth Bidirectional filter (resulting in a 4th order filter) with a cut-off frequency of 14 Hz. Results: Evolution of hip, knee and ankle angle show no significant differences during the last step before touchdown, the take-off phase and the beginning of fly phase. Meanwhile, strong differences in EMG signal are noted inter and intra pole vaulter. However for a same subject the EMG activities seem to converge to some phase locked point. Discussion: All pole vaulters have approximately the same visible coordination This coordination reflects a different muscular control among pole vaulters but also for a considered pole vaulter. These phase locked point could be considered as invariant of motor control i.e. a prerequisite for a normal sequence of the movement and performance realization.