How to implement constraints (conditions) to the estimation of parameters by fitting?

4 次查看（过去 30 天）

Luis B. Walter 2024-6-7

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/2126211-how-to-implement-constraints-conditions-to-the-estimation-of-parameters-by-fitting

评论： Luis B. Walter 2024-6-11

Hi to everyone!

How to implement in the Simbiology interface so that the estimated parameters in a fit satisfy a given condition? For example, estimate by a fit the values of the kinetic constants k1 and k-1 for a serie of experimental data, which satisfy the value of the equilibrium constant Keq (Keq= k1/k-1) within a given interval, such as K= 10±2.

Thank you in advance!

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

采纳的回答

Jeremy Huard 2024-6-7

1
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/2126211-how-to-implement-constraints-conditions-to-the-estimation-of-parameters-by-fitting#answer_1468726

Hi @Luis B. Walter,

one idea would be to actually estimate

and

, with

and have

be calculated in the model with an initial assignment

3 个评论
显示 1更早的评论隐藏 1更早的评论

Jeremy Huard 2024-6-10

Hi @Luis B. Walter,

Thanks for this input!

As of R2024a, SimBiology only supports bound constraints, both in the Model analyzer App and programmatically with sbiofit / fitProblem.

That said, if you want more general constraints, you could set up an optimization problem by writing the objective function and constraints using a SimFunction to simulate your SimBiology model and optimproblem to set up the optimization problem.

Best regards,

Jérémy

Luis B. Walter 2024-6-11