How do I optimize the inputs to a function?
显示 更早的评论
I have a measured pressure-signal (much like the one in the image below) and a function, which when solved by an ode-solver, roughly generates the same dataset.
How do I fine-tune the parameters for the match to improve even further?
My idea of a "plan-of-attack" is as follows:
% Load in raw data.
% Set up initial attempt at finding parameters manually
%
% Run ode23 to generate a dataset using the assigned parameters.
% Subtract the modelled dataset from the raw one - best-fit being achieved when the value is low.
%
% Set up new parameters (Monte-Carlo or based on findings, I'm not picky),
% and redo the process above for e.g. 50 iterations while storing the parameters yielding a best-fit solution.
I have virtually all toolboxes available to me, so this should not be an obstacle if your solution requires one :)
Best Regards,
采纳的回答
Star Strider
2018-2-3
0 个投票
See Solving Coupled Differential Equations (link), Parameter Estimation for a System of Differential Equations (link), Monod kinetics and curve fitting (link) and others.
更多回答（2 个）
Paul Sutton
2019-3-20
0 个投票
Do you mind sharing how you achieved this?
类别
在 帮助中心 和 File Exchange 中查找有关 Ordinary Differential Equations 的更多信息
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!
Translated by 
