How can i use different solvers like ga, patternsearch, particle swarm etc in problem based Mixed integer Linear Programming

7 次查看(过去 30 天)
How can i use different solvers like ga, patternsearch, particle swarm etc in problem based Mixed integer Linear Programming.
I am using MILP and if i try different solvers for example, ga in optimoptions, it shows that the optimization problem automatically uses linprog instead of the solver i specified.Kindly advise .
When i used patternsearch, i got a warning like below:
You have passed PATTERNSEARCH options to LINPROG. LINPROG will use the common options and ignore the PATTERNSEARCH options that do not apply.
To avoid this warning, convert the PATTERNSEARCH options using OPTIMOPTIONS.
showproblem(prob)
options = optimoptions('ga',prob.optimoptions,'Display','iter');
[values,~,exitflag] = solve(prob,'Options',options);
fval = evaluate(prob.Objective,values);

回答(1 个)

Torsten
Torsten 2022-10-27
编辑:Torsten 2022-10-27
MATLAB chooses the solver that is most appropriate (usually means: fastest) for your problem.
Why solving a linear optimization problem with the slowest nonlinear solver ?
But you can choose a different solver if you like:
options = optimoptions("ga",prob.optimoptions,"Display","iter");
[values,fval,exitflag] = solve(prob,"Solver","ga","Options",options)
  9 个评论
Torsten
Torsten 2022-10-31
Seems the Global Optimization Toolbox is installed, but not licenced.
Although I don't understand that the ga testcase seems to work.
Contact Mathworks support for help.

请先登录,再进行评论。

类别

Help CenterFile Exchange 中查找有关 Simulink Supported Hardware 的更多信息

标签

产品

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by