SISO model - how to "auto tune"?

2 Ansichten (letzte 30 Tage)
Michael Cody
Michael Cody am 17 Mär. 2018
Kommentiert: Michael Cody am 2 Sep. 2018
Hi,
Have two versions of a "black box" function/model. No visibility on internals of function.
Single Input-Single Output with a set of tuning parameters.
Say for Version A I know tuning parameters. So can obtain step response.
Then for Version B would like some way to "auto tune" a set of parameters that will give same (or very similar) step response.
Can anyone give me an outline of a method how to achieve this?
I am not afraid of bit of scripting, if required.
Any information appreciated.
  1 Kommentar
Birdman
Birdman am 19 Mär. 2018
Firstly, can you share your black box model and what you have done so far?

Melden Sie sich an, um zu kommentieren.

Antworten (1)

Arkadiy Turevskiy
Arkadiy Turevskiy am 16 Aug. 2018
Hi,
One option is to use Simulink Design Optimization .
Essentially, you would set up an optimization problem to tune parameters to minimize mismatch between simulated and desired step response. Please take a look at the doc and examples of this product and respond with any questions.
  1 Kommentar
Michael Cody
Michael Cody am 2 Sep. 2018
Hi, Thanks for suggestion. DOn't have this toolkit. I will try find colleague who does and see if can progress my project.

Melden Sie sich an, um zu kommentieren.

Community Treasure Hunt

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

Start Hunting!

Translated by