Optimization of 4 cost functions altogether with specified constraints

1 次查看（过去 30 天）
Ali Rezaei
Ali Rezaei 2013-6-20
How can I optimize 4 simple cost functions altogether with specified constraints on each of them using MATLAB and optimization toolbox?
  6 个评论
显示 4更早的评论
Matt J
Matt J 2013-6-21
编辑：Matt J 2013-6-21
Maybe I wasn't clear. Do you mean you want to calculate
[argmin(f1), argmin(f2), argmin(f3), argmin(f4)]
without using a for-loop? If instead, you're looking for a single x that simultaneously minimizes f1(x), f2(x), f3(x), and f4(x), then see Jan's Answer.
Ali Rezaei
Ali Rezaei 2013-6-23
Yes, that is true. Each f_i (i=1,2,3,4) can be shown as f_i(x1,x2,x3,x4). The x1, x2, x3 and x4 are the same for all of 4 functions.

请先登录，再进行评论。

请先登录，再回答此问题。

回答（1 个）

Jan
Jan 2013-6-21
编辑：Jan 2013-6-21
You cannot optimize for 4 different criteria at the same time. An optimization requires a scalar cost function, and this is related to the nature of the term "optimimum".
If you want to consider 4 parameters, you still need a norm to create a scalar: E.g. a weighted or unweighted sum(), max() or mean(). Using max() is not smooth anymore and the optimization method need to consider this explicitly.
Example:
Find the price of an article, which maximizes the income of the sellers and minimizes the costs for the buyers simultaneously.
This must fail for obvious reasons.

类别

在 Help Center 和 File Exchange 中查找有关 Get Started with Optimization Toolbox 的更多信息

标签

Community Treasure Hunt

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

Start Hunting!

Translated by