|
[Sponsors] |
January 15, 2007, 16:38 |
WENO SCHEME BASICS Help Required!!!
|
#1 |
Guest
Posts: n/a
|
Hi Guys,
I am trying to implement a basic 1-D hyperbolic equation using WENO scheme in Matlab. I am facing lot of trouble due to my boundary and also . Could you guys have a look at my Matlab code and send a quick comment..!! Itwill be of great help fweno.m function [DT]=fweno(u,ng,delx,eps) DTP=zeros(ng,1); DTN=zeros(ng,1); DT=zeros(ng,1); % %extrapolate the beginning and end points of data %% u is the data with 3 extra points at the beginining and %at the end % ng is the number of spatial divisions u(3) = 2*u(4)-u(5); u(2) = 2*u(3)-u(4); u(1) = 2*u(2)-u(3); u(ng-2) = 2*u(ng-3)-u(ng-4); u(ng-1) = 2*u(ng-2)-u(ng-3); u(ng) = 2*u(ng-1)-u(ng-2); D = (u(2:ng)-u(1:ng-1))/delx; % WENO scheme based on left-based stencil ( k=i-3, ........i+2) for i=1ng-6) k = i; f1 = D(i); f2 = D(i+1); f3 = D(i+2); f4 = D(i+3); f5 = D(i+4); SP1=(13/12)*(f1-2*f2+f3).^2+0.25.*(f1-4*f2+3*f3).^2; SP2=(13/12)*(f2-2*f3+f4).^2+0.25.*(f2-f4).^2; SP3=(13/12)*(f3-2*f4+f5).^2+0.25.*(3*f3-4*f4+f5).^2; alphap1= 0.1/((eps+SP1).^2); alphap2=0.6/((eps+SP2).^2); alphap3=0.3/((eps+SP3).^2); WPT=alphap1+alphap2+alphap3; WP1=alphap1./WPT; WP2=alphap2./WPT; WP3=alphap3./WPT; DP1= f1/3- 7*f2/6 + 11*f3/6; DP2= -f2/6 + 5*f3/6 + f4/3; DP3= f3/3 + 5*f4/6 - f5/6; DTP(i+3)=WP1.*DP1+WP2.*DP2+WP3.*DP3; DT(i+3)=DTP(i+3); end |
|
January 18, 2007, 04:16 |
Re: WENO SCHEME BASICS Help Required!!!
|
#2 |
Guest
Posts: n/a
|
I don't know Matlab coding syntax but I will try to help you.
First of all, what is the meaning of left-based or biased stencil? Do these (k=i-3, ........i+2) indicating nodes or cell centers? Cell centers should be from i-2 to i+2 for the reconstruction to i+1/2(-) and i-1/2(+). (-) (+) indicating the left- and right-side of the cell boundary respectively. Can you expalin, what do this expression calculates? D = (u(2:ng)-u(1:ng-1))/delx; i starts to count from 1, so f1=D(1), if it is your 3rd ghost cell it is ok. But the end-side uses ng-2 at most, I think it is not you 3rd ghost cell. For finding i+1/2(-) and i-1/2(+) you should use different linear weights, and the characteristic decomposition should be done at each boundary of the cell wrt the reconstruction point. Saygin |
|
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
ENO / WENO scheme | phy4me | Main CFD Forum | 4 | November 5, 2010 04:58 |
Version 15 on Mac OS X | gschaider | OpenFOAM Installation | 113 | December 2, 2009 11:23 |
Finite Volume WENO Scheme | DSS | Main CFD Forum | 3 | January 15, 2007 04:47 |
Flux splitting for WENO scheme | DSS | Main CFD Forum | 6 | January 10, 2007 11:59 |
Conservative TVD or WENO scheme? | Wen Long | Main CFD Forum | 3 | July 9, 2004 17:40 |