Mix-integer optimization problem using GA

5 次查看（过去 30 天）
Hoang Trinh
Hoang Trinh 2019-7-15
编辑： Bruno Luong 2019-7-17
Is there anyway we can assign some specific values to a variable in the optimization problem? For example, x1 can only take values of 32 40 50 and 65.
When I read through this material: https://www.mathworks.com/help/gads/mixed-integer-optimization.html
I know that GA can find integer solutions. However, in my problem, some of my variables can only be assigned with specific values. For example, variable X1 only takes the value of 32 40 50 and 65. Is there anyway we can define it ?
Thank you so much,
Regards,
Kim,
  1 个评论
Bruno Luong
Bruno Luong 2019-7-17
编辑：Bruno Luong 2019-7-17
For example, variable X1 only takes the value of 32 40 50 and 65.
Reformulate your score/variable, instead of minimizing (by GA)
f(x1,x2, ...,xn) with x1 in [32,40,50,65]
minimize
g(y1,x2,...xn) := f(yourset(y1),x2,...,xn) with y1 (integer) in 1:4
and with
yourset := [32,40,50,65]

请先登录，再进行评论。

请先登录，再回答此问题。

回答（2 个）

KALYAN ACHARJYA
KALYAN ACHARJYA 2019-7-15
编辑：KALYAN ACHARJYA 2019-7-15
Define x1 as array with allowable values,
x1 =[32 40 50 65];
if sum(x1==input_value)>=1
% Allowed % Do something
else
%Not allowed with revert message
end
Is this you are looking for?
Bruno Luong
Bruno Luong 2019-7-15
编辑：Bruno Luong 2019-7-17
Take x1 fix allowed value, e.g. 32, optimize wrt (x2,...), repeat 4 times pick the solutions (among 4) with smallest score value.
If your allowed array is large you might use integer programming GA method.
  1 个评论
Walter Roberson
Walter Roberson 2019-7-15
intlinprog permits discrete values but not general objective functions.
ga permits discrete values and general objective functions but uses a randomized approach in its choices.

请先登录，再进行评论。

类别

在 Help Center 和 File Exchange 中查找有关 Surrogate Optimization 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