1 ACCEPTED SOLUTION
Accepted Solutions
The less equations (or the more variables) the more solutions, resp. the more degrees of freedom.
So its easier for the solver to find a solution but this solution will heavily depend an the choice of the guess values.
As you can see below you can check the quality of the solution by evaluating the built-in variable ERR (its exactly this variable which Mathcad tries to minimize) and you may consider adding some additional constraints like aL>0 or the like, to avoid unwanted solutions.
The less equations (or the more variables) the more solutions, resp. the more degrees of freedom.
So its easier for the solver to find a solution but this solution will heavily depend an the choice of the guess values.
As you can see below you can check the quality of the solution by evaluating the built-in variable ERR (its exactly this variable which Mathcad tries to minimize) and you may consider adding some additional constraints like aL>0 or the like, to avoid unwanted solutions.
Thanks, Werner!
I think (guess) it (degrees of freedom) is more for the Find not for Minimize/Maximize Function.
See please the problem with 9 variables and 5 equations
@ValeryOchkov wrote:
Thanks, Werner!
I think (guess) it (degrees of freedom) is more for the Find not for Minimize/Maximize Function.
It applies in a similar way to a solve block with "minimize", too.
The less constraints and the more variables to change, the more possibilities to find different but equally good solutions. Ever so often in such cases depending on the equations given Mathcad possibly won't even change the guess value of the one or other variable given.
In your new sheet its quite interesting that we get the very same results using "find" and without referencing PE in any way!?
@ValeryOchkov wrote:
Yes!
Now I see - we need n variables and >=n equations!
We have different guess values nd get different answers
