Fix Secant method bounds between two positive values

Hi,

I am using Secant Mehtod to find roots of a system of nonlinear algebraic equations. all the x[i] varibales should be at specific range, for instance, [0 50]. My question is how to set the lower and upper bounds when we know x[i] is loacted between lower and upper values?

In Secant Method, after 4 iteration, the new x becomes negative. Is there any way to avoid the bounds?

It would be appreciated if you comment on these question.

Thank you.

[Geoff Hayes]

sina - you may need to show some of the code that you have written so that we can get an idea of how you have implemented this algorithm.

[sina]

Hi Geoff,

You're right. But I defined a function which was linked to another software (Aspen) through

actxserver('Apwn.Document.36.0');

so, it cannot be run on your PC.

Answer 1

[Walter Roberson]

lb = 0; ub = 50;
x(1) = lb;
x(2) = ub;
while ...
        newx = ...
        if newx < lb; newx = lb; end
        if newx > ub; newx = ub; end
        x(K) = newx;
end

[sina]

Thank you Walter for the answer.

I modified my code and added the new if statement. But f(x) remained constant for the rest of iteration!

I am thinking about defineing t as a varibale. When t chnages, then newx = f(t). And newx stands between 0 and 50. This idea, I am not sure whether technically is correct.