Drexel dragonThe Math ForumDonate to the Math Forum



Search All of the Math Forum:

Views expressed in these public forums are not endorsed by Drexel University or The Math Forum.


Math Forum » Discussions » Software » comp.soft-sys.matlab

Topic: Simulannealbnd constraints
Replies: 4   Last Post: Aug 11, 2014 3:45 PM

Advanced Search

Back to Topic List Back to Topic List Jump to Tree View Jump to Tree View   Messages: [ Previous | Next ]
Christoph

Posts: 3
Registered: 8/9/14
Simulannealbnd constraints
Posted: Aug 9, 2014 4:45 AM
  Click to see the message monospaced in plain text Plain Text   Click to reply to this topic Reply

Hi community,

I hope you can help me with this. I want to use the simulated annealing solver (with GUI) for an optimization (minimization) problem that uses a simulink modell. For my problem i have got different variables that can be changed by the solver. I know how the set up a .m file including the cost function and how to define lower and upper boundary conditions for each of my variables, but I wonder if there is a possibility to set additional constraints.
Especially i would need these:

1) The variables that can be change by the solver have to be integers (i.g. I set lb & up [1] & [4] the solver normally can vary the variable like 1.0, 1.1, 1.1545...3.93, 4.0 (double), I'd like the solver to vary those variables only as integers 1,2,3,4)

2) In my simulink modell I calculate different values (with my variables) which should be >, < or == as a defined value. Otherwise the set of variables that the solver has chosen for this iteration step do not meet my needs.

To give you an example:

Variables x(1) and x(2), lb [-3 6], ub [1 9] (in my modell there are 5 variables, chose just one to keep it simple for the example)
The variable have to be an integer x(1) = -3,-2,-1,0,1.
The simulink modell does some calculations like 0.5 * x(1)^2 + x(2) + 5 = val_result;
val_result should be minimized;
additional constraints:
1) ( x(1) + x(2) ) / 2 == integer
2) x(1) * 5 < 4 (I know this is kind of ridiculous in this example because I could also change the ub but I need something like that in my modell)

Is there any possibility to solve such an problem with matlab/simulink optimization?
(I can also change the solver if necessary, just thought of simulanneal because it should not get stuck in a local minima that easy)

Do not hesitate to ask if you need further explanation!

Thanks in advance for your help!
Chris



Point your RSS reader here for a feed of the latest messages in this topic.

[Privacy Policy] [Terms of Use]

© Drexel University 1994-2014. All Rights Reserved.
The Math Forum is a research and educational enterprise of the Drexel University School of Education.