Optimization terminated: average change in the penalty fitness value less than options.FunctionTolerance but constraints are not satisfied.

Question

AMAL ALGEDIR 2017-5-24

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/341754-optimization-terminated-average-change-in-the-penalty-fitness-value-less-than-options-functiontoler

编辑： Abderrahmane DADA 2019-6-1

I am trying to solve MINLP problem by using GA algorithm. I have 1250 binary unknown variables {0,1}. and the number of constraints are 186 constraints where 75 is binary { 0,1} and the remaining nonlinear constraints. I attached a copy from the problem and Matlab files. I need an advice or help with this problem. I tried Pattern search as you recommend in one of the previous discussion also but it gives no feasible point. Thank you

1 个评论
显示 -1更早的评论隐藏 -1更早的评论

John D'Errico 2017-5-24

编辑：John D'Errico 2017-5-24

What do you expect someone to do? No feasible point was found. There is no magical way to find a feasible point that we can give you. The optimizer already tried to do that for you, and failed. And since you are the one most likely to be able to know what the parameters mean in your problem, you are the unique person in the universe that has some chance of choosing a feasible start point.

That no feasible point was found may mean that no such solution exists, or just that you supplied a poor choice to start looking. But in a problem of this size, there is no simple way to locate a feasible point, especially since we are given no clue as to what the parameters mean. Did you write one of the equations incorrectly? Again, you are the one best able to determine that.

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Abderrahmane DADA 2019-5-27

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/341754-optimization-terminated-average-change-in-the-penalty-fitness-value-less-than-options-functiontoler#answer_376773

编辑：Abderrahmane DADA 2019-6-1

Amal.

Have tried to increase "MaxGenerations" and "MaxStallGenerations"?

I have increased those two parameters in my own problem and it worked perfectly and all the constraints were satisfied. I have more than 6 constraints, with a number variables that increase 9 times from one iteration to another.

Try my tip and keep me updated.