How to use function like `min` in symfun?(I got an error)
6 views (last 30 days)
Show older comments
Liming Fang
on 26 Apr 2020
Commented: Walter Roberson
on 26 Apr 2020
For example, i want to create a symbolic function: f(x,y) = min(x-1,y+1);
when f(x,y) = x+y, everything is ok, but when it comes to min, error occured.
syms x y
f = symfun(min([x-1 y+1])),[x y])
I got error like this:
"Unable to convert expression into double array."
Help...
0 Comments
Accepted Answer
Thiago Henrique Gomes Lobato
on 26 Apr 2020
You can't use min or max in symbolic variables in matlab. A work around can be this one:
syms x1 x2 real;
xlt = symfun(1/2*(x1+x2-abs(x1-x2)),[x1,x2]); %min
xgt = symfun(1/2*(x1+x2+abs(x1-x2)),[x1,x2]); %max
I took this answer for a comment here https://de.mathworks.com/matlabcentral/answers/102485-why-are-symbolic-variables-not-allowed-when-using-the-functions-max-min-in-the-symbolic-math-toolbox , which is also from a question very similar to yours
4 Comments
Walter Roberson
on 26 Apr 2020
It depends upon how the condition is constructed.
Functions you should avoid for symbolic expressions that include unresolved symbolic variables include min(), max(), and mod() . Most other functions are willing to postpone evaluation when they are passed symbolic variables, but those three do odd or undesirable things.
And the answer below menthons "piecewise", i wonder whether it is used to solve functions involved condions.
Yes, it is. You can even differentiate piecewise -- but the boundary conditions might not get the right value, as piecewise() expressions are often discontinuous on the boundaries and differentiating piecewise() does not return a special "undefined" result for the boundary.
However, solve() sometimes has difficulty with piecewise, especially something of the form piecewise(condition, expression, condition, expression...) == value . Sometimes you need to solve() against each of the expressions and then test to see whether the result meets the conditions
More Answers (1)
See Also
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!