Optimization objective

shsreekanth · May 9, 2014, 15:40

Hi,

In the configuration file, there is an option to specify objective function with the scale. Objective functions are drag , lift etc. How does the solver know whether its a minimization or maximization problem.

In my case , i am interested in minimizing the force on an object. So how should the objective function be specified?

Thanks,

Sreekanth

copeland · May 15, 2014, 16:13

Hello Sreekanth,

Generally, objective functions in SU2 are always minimized (there are some exceptions like efficiency). For an example on how to specify a design problem in SU2, visit the TestCases/euler/naca0012 and look at the optimal design settings in the NACA 0012 configuration file provided.

Hope that helps,
Sean

shsreekanth · May 16, 2014, 02:11

Hi,
Thanks for the reply.
You say there are some exceptions.
1. By default, is "Lift" being minimized or is it maximized?
2. If its being maximized, can i convert it to minimization problem by multiplying with -1 ?
3. Can i have a linear combination, or a ratio of the two terms, say (LIFT + 10* DRAG )? I tried giving like that, but it doesn't work.

Sreekanth

rktchip · May 16, 2014, 03:10

hey sreekanth,
lift is maximized by default. you can set the scale to be negative to change that --
OPT_OBJECTIVE = LIFT * -1.0
You can't combine objectives here at the moment, tho be sure to explore posing your problem with constraints, like lift constrained drag minimization. You can also take a look inside of shape_optimization.py to see how you can build your own objective functions in python.

May 9, 2014, 15:40	Optimization objective	#1
shsreekanth Member Sreekanth Join Date: Jun 2013 Location: India Posts: 30 Rep Power: 11	Hi, In the configuration file, there is an option to specify objective function with the scale. Objective functions are drag , lift etc. How does the solver know whether its a minimization or maximization problem. In my case , i am interested in minimizing the force on an object. So how should the objective function be specified? Thanks, Sreekanth

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
what are the scaling factors of Optimization objective and constraint functions ?	mechy	SU2 Shape Design	4	September 22, 2017 03:23
Total force optimization or pressure force optimization?	robyTKD	SU2 Shape Design	1	August 19, 2013 15:54
COMSOL: Setting up simple optimization problem	tdh89	COMSOL	1	May 23, 2012 03:55
COMSOL: optimization applied to a laminar flow	tdh89	COMSOL	0	April 30, 2012 08:57
How to do Design Optimization in Fluent 13.0?	p36288	Fluent UDF and Scheme Programming	1	July 20, 2011 00:37

May 15, 2014, 16:13		#2
copeland Member Sean R. Copeland Join Date: Jan 2013 Posts: 40 Rep Power: 12	Hello Sreekanth, Generally, objective functions in SU2 are always minimized (there are some exceptions like efficiency). For an example on how to specify a design problem in SU2, visit the TestCases/euler/naca0012 and look at the optimal design settings in the NACA 0012 configuration file provided. Hope that helps, Sean

May 16, 2014, 02:11		#3
shsreekanth Member Sreekanth Join Date: Jun 2013 Location: India Posts: 30 Rep Power: 11	Hi, Thanks for the reply. You say there are some exceptions. 1. By default, is "Lift" being minimized or is it maximized? 2. If its being maximized, can i convert it to minimization problem by multiplying with -1 ? 3. Can i have a linear combination, or a ratio of the two terms, say (LIFT + 10* DRAG )? I tried giving like that, but it doesn't work. Sreekanth

May 16, 2014, 03:10		#4
rktchip Member Trent Lukaczyk Join Date: Feb 2011 Location: Stanford, CA Posts: 75 Rep Power: 14	hey sreekanth, lift is maximized by default. you can set the scale to be negative to change that -- OPT_OBJECTIVE = LIFT * -1.0 You can't combine objectives here at the moment, tho be sure to explore posing your problem with constraints, like lift constrained drag minimization. You can also take a look inside of shape_optimization.py to see how you can build your own objective functions in python.