Hai...need help for this Matlab code..why the output can't come out..already try in many way...anyone??...plez help me

%function u=pdehypwave(c,L,Tmax,h,k,c1,c2,f,g)
%Filename:pdehypwave.m
%Usage: u=pdehypwave(c,L,Tmax,h,k,c1,c2,f,g)
%To solve wave equation c^2u_xx=u_tt, 0<x<L, t>0
%with the boundary conditions
% u(0,t=c1, u(L,t)=c2, t>0
%and the initial condtions
%   u(x,0)=f(x), 0<=x<=L
%   u_t(u(x,0))=g(x), 0<=x<=L
%(c)C.R.C.Teh

%INPUT
%Filename: pdehypwave_in.m
c=1;                        %coefficient in PDE; c^2u_xx=u_tt
L=pi;                        %length of string in x direction
Tmax=1;                      %maxmimum time t
h=0.3142;                      %step size for space x
k=0.5;                      %step size for space t
c1=0;                       %BC1: w(0,t)=c1 at x=0
c2=0;                       %BC2: w(pi,t)=c2 at x=pi
f=inline('0','x');            %IC1: w(x,0)=f(x) at t=0
g=inline('0','x');             %IC2: w_t(w(x,0))=g(x) at t=0

%OUTPUT
%u matrix of solution value

n=L/h;           %number of subinterval of [0,a]
m=Tmax/k;        %number of subinterval of [0,b]
N=n+1;           %number of grids [0=x1,x2,....,xN=a]
M=m+1;           %number of grids [0=y1,y2,....,yN=a]
x=0:h:L;         %x grids
t=0:k:Tmax;      %t grids
w=zeros(1.4,1);    %initialization matrix solution u
for i=2:N
    w(i,1)=f(x(i));     %IC1: initial values
end
for j=1:M
    w(i,j)=c1;          %BC1: boundary values
    w(N,j)=c2;          %BC2: boundary values
end
r=c*k/h;

%EXPLICIT METHOD

for j=3:M       %time loop
end
    for i=2:N-1 %space loop
    end
        if j==3
  w(i,j+1)=(2-2*c^2*r^2)*w(i,j)+(c^2*r^2)*w(i-1,j)+(c^2*r^2)*w(i+1,j)-w(i,j-1)+k^2*sin(u(i,j)-w(i,j));
end
disp('--------------------------------------')
fprintf(' u=x\\t')
fprintf(' %4,2f ',t)
fprintf('\n')
disp('--------------------------------------')
for i=1:N
fprintf(' %4.4f',x(i))
for j=1:M
    fprintf('%10.4f',u(i,j))
end
fprintf('\n')
end

*The question need me to come out the value from this code by using Explicit scheme.

*here the value given
k=0.05
r=0.1592
c=1
r=0.15925
a=1
*boundary condition 
w(0,t)=0, w(pi,t)=0    0 less and equal x less and equal pi,  0 less and equal t less and equal a

*Initial condition
w(x,0)=0 wt(x,0)=0

*the formula is
 w(i,j+1)=(2-2*c^2*r^2)*w(i,j)+(c^2*r^2)*w(i-1,j)+(c^2*r^2)*w(i+1,j)-w(i,j-1)+k^2*sin(u(i,j)-w(i,j));

*for i=2 to n-1,evaluate the initial condition.

*for j=3:M,for i=2:n-1 evaluate the formula

[Walter Roberson]

What happens when you try? Is there an error message? If so what is it and on which line?

[ZARITH SOFIA]

yes i already try it...but when i'm trying to run this that is an error Warning: Size vector should be a row vector with integer elements. > In EXPLICITAKU at 34 ??? Attempted to access w(4.18269,1); index must be a positive integer or logical.

Error in ==> EXPLICITAKU at 40 w(N,j)=c2; %BC2: boundary values

[dpb]

Indeed...looking at your initialization we see...

L=pi;                        %length of string in x direction
h=0.3142;                      %step size for space x
...

n=L/h;           %number of subinterval of [0,a]
m=Tmax/k;        %number of subinterval of [0,b]
N=n+1;           %number of grids [0=x1,x2,....,xN=a]
...
for j=1:M
  w(i,j)=c1;          %BC1: boundary values
  w(N,j)=c2;          %BC2: boundary values
  ...

Your value for N won't be integer -- above it will be 10-something, not 4-something at the time of the error. I didn't see in a quick reading where the particular value actually came from but it's the similarly-recognizable problem. The value is ~(pi+1) from some other like calculation no doubt.

The error is quite explicit and exact -- you've got to make the grid points integer to be used as indices into arrays. The previous could work as

h=pi/10; N=round(pi/10); % will return 10, exactly.

You need something similar at the point where the result turned out to be 4.18269... to get, one would presume, 4.

Better, of course, would be to write it all in terms of simply the number of grid points desired and keep the actual grid values stored along with the grid instead of trying to make the grid match exactly which greatly restricts the flexibility of any solution/model.

[ZARITH SOFIA]

sorry not clearly understand with this explanation...can you give me your email address so it much easy to me to contact you..

[dpb]

What, specifically, don't you understand?

Array indices must be integers; your manner of computing them returns other than that as demonstrated.

In general, one would define the dx for a grid and then use x=x0:dx:xEnd; as the coordinate vector and then reference it (and the solution vector associated with same) with the indices. Trying to use the actual x values as indices won't work as, in general, they're not integer-valued.

[ZARITH SOFIA]

ok i get that..but still have problem when all the output come out is zero..

[Image Analyst]

You're not making it easy for us to help you. I copied and pasted your code and it didn't even run. Please attach your m-file, and any data file that we might need to read in so we can continue trying to help you. Use the paper clip icon and don't forget to click the "Attach file" button after you browse to the file(s). ANd is this any different than this question: http://www.mathworks.com/matlabcentral/answers/107349-why-output-can-t-come-out You don't want to have to look in two places for answers.

[ZARITH SOFIA]

actually i'm not familiar with this web...this is my first time that's why it easy for me to communicate via email.

[ZARITH SOFIA]

yes the code is different