can anyone help me ?

here is a code matlab of one degree differential equation, i want to plot the error method for different values of N, can anyone help me

[Geoff Hayes]

diadalina - please clarify what you mean by i want to plot the error method for different values of N. Does your code do some or all of this already? If not, where should this happen in your attached code?

[diadalina]

the error wich is equal to norm(the approchsolution-the exactsoltuion), i want to show the accuracy of the numerical method used here , my code doesn't work correctly !!

Answer 1

[Stephan]

clear all;
clc;
close all;
N=[100,200,300,400];
hold on
for k=1:numel(N)
%%les données mathématiques
t      = 0:1/(N(k)-1):1;
%les conditions initiales
c0=0;
%%résolution de notre système 
[~,Y]=ode45(@systest,t,c0 );
sol_exacte=t.*exp(t);
subplot(2,1,1)
plot(t,Y(:,1),t,sol_exacte,'+')
legend('sol-app','sol-exacte')
err(k)=norm(Y(:,1)'-sol_exacte)
e=err(k);
end
hold off
subplot(2,1,2)
semilogy(N,err,'ob--')
legend('error')

function F=systest(t,Y)
F(1,1)=Y(1)+exp(t);
end