Hi Maria, let’s hope im helpful:
First the mathematical blabla: If I understand correctly, your model is defined by a differential-algebraic equation system. There are several approaches that have been described for the optimization of such models (a dynamic optimization problem), and which one to choose depends on the characteristics of your problem. They basically vary on (if anything does) what gets discretized and what doesn’t and so the size of the nlp and the ability of the solver to follow certain constrains gets affected. If you have access to this paper (http://www.sciencedirect.com/science/article/pii/S0255270107001122) , it is directed mainly towards some of this strategies, but the references and introduction should prove useful anyway.
Computational blabla: If you want help you as to why GAMS is interpreting things as such or such, I would need to see the code (or at least part of it).
I have solved dynamic optimization problems using GAMS.
Let’s see if that helps you !!
Claudio
On Mon, May 6, 2013 at 8:45 PM, Maria Giuliana wrote:
Hello Claudio, thank you for your help!
actually, I’m having a lot of trouble to decide how am I going to write my problem…
I have two ideias:
-
first:
i write my problem as a black box model
i want to minimize the time of a reaction using I0 as a positive variable
so, the black box model uses the first guess that i give, returns the total time and the solver vary my variable I0 and finds the solution
but this ins’t hapening and, as i said, my problem is read as a LP model…
i think that this is the best way to solve the problem, but it’s not working
-
second
-
i write my problem with equations for the model and then it’s read as a NLP
but with equations i con’t see a way to write how the solve statement can be called…
have you alread solved any problem with differential equations using gams?
do you have any idea to explain why my problem is not being read as a NLP?
i don’t know if i’m explaing it very well, but thank you…
Em segunda-feira, 29 de abril de 2013 17h46min16s UTC-3, Maria Giuliana escreveu:
hello people!
I’m trying to solve a model that has a differential equation on GAMS
to do this I’m using a simple method to approximate the result of the differential equation
but I’m finding a lot of problems doing this…
for example, the solver does not find a solution for the problem, it only uses my first guess and says that it is the best answer
other problem that i’m having is that the solver is reading my problem as a LP, but actually, it is a NLP
can somebodyhelp me, please?!!
To unsubscribe from this group and stop receiving emails from it, send an email to gamsworld+unsubscribe@googlegroups.com.
To post to this group, send email to gamsworld@googlegroups.com.
Visit this group at http://groups.google.com/group/gamsworld?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.
\