Using simulation to determine thrust vs flapping amplitude relation

quarkz · November 19, 2015, 00:58

Hi,

I have a robotic flapper with a motor powered fin behind for propulsion. In a water expt, the fin flaps at different flapping rate and I got its constant velocity in water. I hope to validate this thru simulation, and also obtain the thrust vs flapping frequency relation.

One way would be to do the simulation directly ie a body with a fin moving in the water domain but that will require a lot of grids since the whole body moves. For every time step, I get the thrust and use it to get the acceleration of the body and use v = u + at and s = ut + 0.5at^2 to get the new velocity and displacement. The body is then moved forward in the domain.

Another way is to keep the body fixed with the fin flapping with incoming flow that was obtained from the expt. If the prediction is correct, I should get zero thrust since it is travelling at this constant velocity. Is this fair? Since I already know the velocity beforehand.

Or is there a better way to do it?

Thanks.

quarkz · November 19, 2015, 08:18

Maybe another mtd is to solve in a non-inertial reference frame for the NS eqn.

So at 1st the body will be accelerating. The incoming flow will start from zero and increases as the body accelerates, until at a certain equilibrium pt where the acc = 0 and it reaches a constant incoming velocity. Is it possible to solve this way?

November 19, 2015, 00:58	Using simulation to determine thrust vs flapping amplitude relation	#1
quarkz Senior Member TWB Join Date: Mar 2009 Posts: 414 Rep Power: 19	Hi, I have a robotic flapper with a motor powered fin behind for propulsion. In a water expt, the fin flaps at different flapping rate and I got its constant velocity in water. I hope to validate this thru simulation, and also obtain the thrust vs flapping frequency relation. One way would be to do the simulation directly ie a body with a fin moving in the water domain but that will require a lot of grids since the whole body moves. For every time step, I get the thrust and use it to get the acceleration of the body and use v = u + at and s = ut + 0.5at^2 to get the new velocity and displacement. The body is then moved forward in the domain. Another way is to keep the body fixed with the fin flapping with incoming flow that was obtained from the expt. If the prediction is correct, I should get zero thrust since it is travelling at this constant velocity. Is this fair? Since I already know the velocity beforehand. Or is there a better way to do it? Thanks.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
combustion flow simulation in liquid rocket thrust chamber	geetha sri	FLUENT	1	February 4, 2016 18:04
Flapping Airfoil Simulation in fluent + Udf	RajeshAero	Main CFD Forum	1	February 8, 2011 05:50
Flapping Airfoil Simulation in Low Reynolds Number Flow	M.Sc_Student	FLUENT	4	April 7, 2010 11:28
Flapping Airfoil Simulation in Low Reynolds Number Flow	M.Sc_Student	Main CFD Forum	0	April 6, 2010 10:30
Flapping simulation	M-A	FLUENT	3	June 2, 2005 05:46

November 19, 2015, 08:18		#2
quarkz Senior Member TWB Join Date: Mar 2009 Posts: 414 Rep Power: 19	Maybe another mtd is to solve in a non-inertial reference frame for the NS eqn. So at 1st the body will be accelerating. The incoming flow will start from zero and increases as the body accelerates, until at a certain equilibrium pt where the acc = 0 and it reaches a constant incoming velocity. Is it possible to solve this way?