— Inspired by Hebb’s cell assembly theory about how the brain worked, we have developed a function localization neural network (FLNN). The main part of a FLNN is structurally t...
Finite difference methods continue to provide an important and parallelisable approach to many numerical simulations problems. Iterative multigrid and multilevel algorithms can co...
We consider a control volume(covolume) method for second order elliptic PDEs with the rotated-Q1 nonconforming finite element on rectangular grids. The coefficient may a variable,...
ct 9 A new discrete-velocity method for the numerical solution of large eddy simulation (LES) is presented based on relax10 ation type formulation. The discrete-velocity equations ...
In this paper, we propose a stochastic mixed multiscale finite element method. The proposed method solves the stochastic porous media flow equation on the coarse grid using a set ...