A node-numbering-invariant directional length scale for simplex elements

Variational multiscale methods, and their precursors, stabilized methods, have been very popular in flow computations in the past several decades. Stabilization parameters embedded in most of these methods play a significant role. The parameters almost always involve element length scales, most of the time in specific directions, such as the direction of the flow or solution gradient. We require the length scales, including the directional length scales, to have node-numbering invariance for all element types, including simplex elements. We propose a length scale expression meeting that requirement. We analytically evaluate the expression in the context of simplex elements and compared to one of the most widely used length scale expressions and show the levels of noninvariance avoided.

A Method to Solve the Reaction-Diffusion-Chemotaxis System

The objective of this article is to use a recent developed scheme to simulate reaction-diffusion-chemotaxis equations. The solution gradient required for an accurate discretization is computed directly as an additional variable rather than interpolated from solution values around neighboring computational nodes. To achieve this goal, a supplementary equation and its associated control volume are introduced to retain a compact and accurate discretization. Scheme essentials are exposed by the numerical analysis on two-dimensional chemotaxis problems to reveal its formal accuracy. Due to its highly comprehensible and practical features, this formulation can be easily extended to solve problems for other two-dimensional rectangular grid systems. One- and two-dimensional problems are solved to verify its simulation accuracy and to study the possible formation of bacteria bands. We further perform the linearization technique to the reaction term to increase the stability of the current scheme. From the numerical analysis and computational results, it is found that the present formulation is a useful tool to solve reaction-diffusion-chemotaxis equations.

Fluid Flow Simulations Based on an Equation-Solving Solution Gradient Strategy

A compact and accurate discretization for fluid flow simulations is introduced in this paper. Contrary to the common wisdom in a convectional scheme, the solution gradient required for a high-resolution scheme is provided by solving its corresponding difference equation rather than by interpolation from solution values at neighboring computational nodes. To achieve this goal, a supplementary equation and its associated control volume are proposed to retain a compact and accurate discretization. Scheme essentials are exposed by numerical analyses on simple one-dimensional modeled problems to reveal its formal accuracy. Several test problems are solved to illustrate the feasibility of present formulation. From the obtained numerical results, it is evident that the proposed scheme will be a useful tool to simulate fluid flow problems in arbitrary domains.