sensitivity analysis of transient simulation?

Guille1811 · November 3, 2017, 14:54

Hi, i have to do a sensitivity analysis of a transient simulation of a system that is transient by nature, ie it does not have a steady state (yes, i run a steady state simulation to verify this).

My question is, since the simulation takes like 4 days to complete, what is a clever way to do sensitivity analysis in cases like this?

ghorrocks · November 3, 2017, 23:02

Some ideas:

* You could run just the the starting little bit of the simulation. Enough to get the important physics happening, but not the whole simulation.
* You could run a starting simulation until the flow gets going, then stop it. Then do the sensitivity analysis from there using the starting simulation as an initial condition
* You could develop a simple analog of the flow. For instance, for automobile aerodynamics don't run the actual car body you are looking at, instead do the sensitivity analysis on the simpler Ahmed body instead.

Guille1811 · November 5, 2017, 00:36

Thanks, i think ill do the first one.

November 3, 2017, 14:54	sensitivity analysis of transient simulation?	#1
Guille1811 Member Join Date: Oct 2017 Posts: 89 Rep Power: 8	Hi, i have to do a sensitivity analysis of a transient simulation of a system that is transient by nature, ie it does not have a steady state (yes, i run a steady state simulation to verify this). My question is, since the simulation takes like 4 days to complete, what is a clever way to do sensitivity analysis in cases like this?

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Problem in initializing transient simulation with a finer mesh	sidd	CFX	8	April 29, 2016 02:25
the problem of my transient simulation "Floating point exception: Overflow "	alloveyou	CFX	15	November 22, 2012 11:14
Beginning a Transient Simulation	Norflow	CFX	2	October 7, 2011 00:38
Resume Transient simulation	HMR	CFX	1	June 28, 2011 21:13
transient simulation of a rotating rectangle	icesniffer	CFX	1	August 8, 2009 07:25

November 3, 2017, 23:02		#2
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,705 Rep Power: 143	Some ideas: * You could run just the the starting little bit of the simulation. Enough to get the important physics happening, but not the whole simulation. * You could run a starting simulation until the flow gets going, then stop it. Then do the sensitivity analysis from there using the starting simulation as an initial condition * You could develop a simple analog of the flow. For instance, for automobile aerodynamics don't run the actual car body you are looking at, instead do the sensitivity analysis on the simpler Ahmed body instead.

November 5, 2017, 00:36		#3
Guille1811 Member Join Date: Oct 2017 Posts: 89 Rep Power: 8	Thanks, i think ill do the first one.