Getting same final value for various time-steps

1 次查看（过去 30 天）
Turgut Ataseven
Turgut Ataseven 2022-5-31
评论： Turgut Ataseven 2022-5-31
Hi.
I am trying to modify my Crank Nicholson method code so that it breaks when 14.9<T(11,11,k+1)<15.1 condition is satisfied.
This is the equation
Since time interval is unknown,time variable t, time step variable p and variables including p and t are commented out. I arbitrarily picked p values to obtain the time when 14.9<T(11,11,k+1)<15.1 but whatever the p interval is, I always face with the same result ("final" matrix). What should be done to make the code work properly?
%t=4; % Total time (s)
delta_t=0.05; % Time step
L=2; % Length of each edge
delta_x=0.1; % Spacing
% Boundary temperatures
T1=10;
T2=10;
T3=10;
T4=10;
% Initial temperature
T5=400;
n=((L/delta_x)+1)^2; % Total no of nodes
m=sqrt(n); % Number of nodes in each row and column
%r1=(t/delta_t)+1; % Number of time steps
%p=round(r1); % Number of time steps (rounded)
r=0.033;
% Placing initial and boundary conditions
%T=zeros(m,m,p); % Pre-allocating
for k=1:8000 % Time-step loop
for i=1:m % x coord loop
for j=1:m % y coord loop
if(i==1)&&(j==1)
T(i,j,k)=(T4+T1)/2;
elseif(i==1)&&(j==m)
T(i,j,k)=(T2+T1)/2;
elseif(i==m)&&(j==m)
T(i,j,k)=(T2+T3)/2;
elseif(i==m)&&(j==1)
T(i,j,k)=(T4+T3)/2;
elseif(i==1&&(j>1&&j<m))
T(i,j,k)=T1;
elseif(j==m&&(i>1&&i<m))
T(i,j,k)=T2;
elseif(i==m&&(j>1&&j<m))
T(i,j,k)=T3;
elseif(j==1&&(i>1&&i<m))
T(i,j,k)=T4;
else
T(i,j,k)=T5;
end
end
end
end
% Solution
keepgoing = true;
for k=1:7999
for i=2:m-1
for j=2:m-1
T(i,j,k+1)=1/(1+2*r)*(r/2*(T(i-1,j,k+1)+ T(i+1,j,k+1)+ T(i,j+1,k+1)+ T(i,j-1,k+1))+(1-2*r)*T(i,j,k)+r/2*(T(i-1,j,k)+ T(i+1,j,k)+ T(i,j+1,k)+ T(i,j-1,k)));
if (i==11 && j==11 && T(i,j,k+1)>14.9 && T(i,j,k+1)<15.1)
keepgoing =false;
break;
end
end
if ~keepgoing; break; end
end
if ~keepgoing; break; end
end
final=T(:,:,k);
imagesc(T(:,:,k));
colorbar;
title('Temperature Profile')
Thanks.
  3 个评论
显示 1更早的评论
Jan
Jan 2022-5-31
Easier to debug:
T=zeros(m,m, 8000);
% for k=1:8000 The body does not depend on k, so omit this loop!
for i=1:m % x coord loop
for j=1:m % y coord loop
if i==1
if j==1
T(i,j,:) = (T4+T1)/2;
elseif j==m
T(i,j,:) = (T2+T1)/2;
else
T(i,j,:) = T1;
end
elseif i==m
if j==1
T(i,j,:) = (T4+T3)/2;
elseif j==m
T(i,j,:) = (T2+T3)/2;
else
T(i,j,:) = T3;
end
else
if j==1
T(i,j,:) = T4;
elseif j==m
T(i,j,:) = T2;
else
T(i,j,:) = T5;
end
end
end
end
Or what about:
T = zeros(m, m, 8000);
T(1, 1, :) = (T4+T1) / 2;
T(1, m, :) = (T2+T1) / 2;
T(1, 2:m-1, :) = T1;
T(m, 1, :) = (T4+T3) / 2;
T(m, m, :) = (T2+T3) / 2;
T(m, 2:m-1, :) = T3;
T(2:m-1, 1, :) = T4;
T(2:m-1, m, :) = T2;
T(2:m-1, 2:m-1, :) = T5;
Now, what is your problem? You get the same output final if you vary what?
Turgut Ataseven
Turgut Ataseven 2022-5-31
@Jan Thanks for the reply.
I get the same output "final" as I vary ending value of k (number of time steps).
keepgoing = true;
for k=1:7999 % this value %%
for i=2:m-1
for j=2:m-1
T(i,j,k+1)=1/(1+2*r)*(r/2*(T(i-1,j,k+1)+ T(i+1,j,k+1)+ T(i,j+1,k+1)+ T(i,j-1,k+1))+(1-2*r)*T(i,j,k)+r/2*(T(i-1,j,k)+ T(i+1,j,k)+ T(i,j+1,k)+ T(i,j-1,k)));
if (i==11 && j==11 && T(i,j,k+1)>14.9 && T(i,j,k+1)<15.1)
keepgoing =false;
break;
end
end
if ~keepgoing; break; end
end
if ~keepgoing; break; end
end
end
T matrix converges to same values after approximately 260 steps. It is probably because both sides of the equation has "k+1".

请先登录，再进行评论。

请先登录，再回答此问题。

回答（1 个）

Torsten
Torsten 2022-5-31
for k=1:7999
for i=2:m-1
for j=2:m-1
T(i,j,k+1)=1/(1+2*r)*(r/2*(T(i-1,j,k+1)+ T(i+1,j,k+1)+ T(i,j+1,k+1)+ T(i,j-1,k+1))+(1-2*r)*T(i,j,k)+r/2*(T(i-1,j,k)+ T(i+1,j,k)+ T(i,j+1,k)+ T(i,j-1,k)));
end
end
if T(11,11,k+1)>14.9 && T(11,11,k+1)<15.1
break
end
end

类别

在 Help Center 和 File Exchange 中查找有关 Loops and Conditional Statements 的更多信息

标签

产品


版本

R2021a

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by