can anyone tell me why this code doesn't stop running

w=1;
d=200;
B=530;
Ca=300;
Cb=300;
Fh=70;
L=600;
da=100;
db=100;
iteri=0;
iterj=0;
iterk=0;
x=0;
Tk=Ca-Cb;
df=abs(Tk-Fh);
while df>0.8 
  iterk=iterk+1;
  x=x+0.1;
  while abs(da)>0.0001
    iteri=iteri+1;
    Ba=B+x;
    Dca=L-Ba-sqrt(d*(d+2*Ca))+Ca*acosh(1+(d/Ca))
    Ja=-d/(sqrt(d*(d+2*Ca)))+acosh(1+(d/Ca))-d/(Ca*sqrt(((1+(d/Ca))^2)-1))
    da=Dca/Ja;
    Ca=Ca-da;
  end
  while abs(db)>0.0001
    iterj=iterj+1;
    Bb=B-x;
    Dcb=L-Bb-sqrt(d*(d+2*Cb))+Cb*acosh(1+(d/Cb))
    Jb=-d/(sqrt(d*(d+2*Cb)))+acosh(1+(d/Cb))-d/(Cb*sqrt(((1+(d/Cb))^2)-1))
    db=Dcb/Jb;
    Cb=Cb-db;
  end
  Tk=Ca-Cb;
  df=abs(Tk-Fh);
end

Answer 1

[Andreas Goser]

Well, the answer is that df is never <= 0.8. Actually it becomes stable at 67.9.

But I understand that does not help you. Maybe you can share what type of algorithm this is / should be and scientists and that area could comment?

[dpb]

Looks like doesn't actually compute a convergence value but the end result maybe???

Answer 2

[Image Analyst]

I'd guess that one of the 3 "while" conditions is not ever false so you're getting an infinite loop. You can find out which one after you look at this link: http://blogs.mathworks.com/videos/2012/07/03/debugging-in-matlab/ This is why I say time and time again that while loops must always use a failsafe, like a loop counter where you break out if it ever gets bigger than you might possibly expect, like a million or something

loopCounter = 1;
maxCount = 1000000;
while someCondition && loopCounter < maxCount% While loop with failsafe
    % Some code you want to run goes here.
    % Now at the end of the loop, increment the loop counter.
    loopCounter = loopCounter + 1;
end
% Alert user if while loop finished only because of the failsafe.
if loopCounter >= maxCount
  message = sprintf('Warning: loop did not finish after % iterations', loopCounter-1);
  uiwait(warndlg(message));
end