I'm trying to find all the points between 0 and 10 where the derivative of y=0 by using (x1(H)+x2(H))/2) as a staring point but I keep on getting this error... Please help :(
Error using fzero (line 169) If FUN is a MATLAB object, it must have an feval method.
h=10;
ti=(1/200);
n=[1.2 2.9];
ni=[1 3];
na=[2 1];
x=[];
for k=1:numel(n)
x=sort([x (((0:1:1+(ceil(h/(n(k)))))-0.5)*(n(k)))]);
end
x1=(x(1:numel(x)-1));
x2=(x(2:numel(x)));
for T=[1:10];
nwa=[na.*sin((T*ti)*(pi./(ni)))];
for H=(1:numel(x1))
syms d;
y=sum(nwa.*cos(d.*(pi./n)));
X(H)=fzero(diff(y),(x1(H)+x2(H))/2);
end
P=(X(X>x1<x2))';
end

 Accepted Answer

dpb
dpb on 20 Oct 2013
From the doc's for fzero
X = fzero(FUN,X0) ... FUN is a function handle.
Your code is
X(H)=fzero(diff(y),(x1(H)+x2(H))/2);
where you've entered an actual function.

4 Comments

I can't figure out how to apply that though...
Well, it's not clear to me what your function is intended to be solving but the basic idea is to write a function either as an anonymous function or give the handle to an existing function and an initial guess and the fzero will find the value that provides a zero to that function on the assumption a zero-crossing can be found near (for some definition of near) x0.
Clarify what it is you're trying to get fzero to do for you here...
I'm trying to find a way to calculate the position of the antinodes (the maximas and minimas) of a waveworm=sum(nwa.*cos(d.*(pi./n))) by using its derivative, and then using fzero starting at all the midpoints between consecutive multiples of n between 0 and 10.
There's an example in the doc on fzero using trig functions for formulation that should be illustrative.

Sign in to comment.

More Answers (0)

Categories

Find more on Optimization in Help Center and File Exchange

Tags

Asked:

on 20 Oct 2013

Commented:

dpb
on 21 Oct 2013

Community Treasure Hunt

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

Start Hunting!