CFD Online Logo CFD Online URL
Home > Forums > Software User Forums > SU2 > SU2 Shape Design

SLSQP linesearches

Register Blogs Members List Search Today's Posts Mark Forums Read

LinkBack Thread Tools Search this Thread Display Modes
Old   April 26, 2019, 08:41
Question SLSQP linesearches
tom pratt
Join Date: Jan 2016
Posts: 55
Rep Power: 9
tomp1993 is on a distinguished road
Hi All,
i am using python SLSQP as an optimiser and asking it to call SU2 V6.2.0 for direct and discrete adjoint solutions, and it is doing a pretty good job of reducing drag but when it gets near an optima the process starts doing endless linesearches with only minute changes in design between each major iteration - yet i know that there is still a better solution than the one found and i am confident that it has not found a local minima.
This happens in both fixed CL mode (not using lift adjoint) and also not fixed CL but constrained lift through lift adjoint and addition of FFD_AOA design variable.
i have tried scaling the function and gradient values (as a means of bypassing the armijo condition) and have tried using ever-smaller values for ftol (an argument given to slsqp) but still get the linesearches.
is there a reliable means of 'switching off' this line searching feature or does anyone have any advice as to what exact criteria the optimiser uses to decide wether or not to do one of these searches?
the lift and drag sensitivities are not zero and not necessarily conflicting when the linesearch starts.
thankyou for any suggestions!
tomp1993 is offline   Reply With Quote


Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are On
Refbacks are On

Similar Threads
Thread Thread Starter Forum Replies Last Post
inequality constraints incompatible in (SLSQP exit mode 4) tomp1993 SU2 Shape Design 2 November 25, 2022 10:27
Any description about SLSQP dasssj SU2 4 March 4, 2016 22:17

All times are GMT -4. The time now is 05:04.