Drag and lift understimation on a rigid body

flinde · January 21, 2015, 08:27

Hi all,
I am currently using CFD to predict hydrodynamic forces on a ship in restricted waters. For that I am using ansys fluent software. So far I obtained a good correspondance between my numerical results and experimental data. However, for higher fluid velocity, I am underestimating the drag and lift coefficient (ie horizontal and vertical forces).
I studied the influence of several parameters such as mesh density, turbulence model (kE, kW, kW-SST), boundary conditions at the inlet (mass flow inlet, pressure inlet) and outlet (pressure outlet, outflow), domain size, change of the spatial discretizaion scheme,... But none did have an influence on the results. I am inclined to think that the pressure around the hull is underestimated but i do not know where this could come from...
Any help would be appreciated.
Thanks

t.teschner · January 21, 2015, 15:15

have you checked the pressure distribution with the laminar model (i.e. no turbulence model) and checked if the lift is predicted correctly?
i would argue that at least the pressure field should be captured somewhat realistically (of course you don't get drag). if the pressure is off then there might be some issue with the set up / model.

flinde · January 23, 2015, 08:59

I ran a simulation with the laminar model and just like with the turbulence model, the lift is underestimated. According to you, there might be a problem with the setup / model. But given that i already checked parameters such as mesh sensitivity and numerical parameters i don't know where the problem could come from, especially given that it works well with lower speeds... Maybe I didin't check thoroughly enough those parameters... But then, what should I check first, from what could this problem arise from?
Thanks in advance for your help.

jpando · January 23, 2015, 09:54

Flinde,
The issue to me looks like your y+ value may be wrong. As the velocity of the fluid increases, this also increases your y+. To maintain the same value of y+ as the lower velocity cases, you must decrease the first cell height on the wall. You also want to make sure your inflation layers traverse the entire height of boundary layer. This will make sure you have good quality cells in the most imporant areas of the mesh (where your lift and drag forces are generated).

If this doesn't work, then you will want to check your compressibility model. Velocities around the body can reach speeds where comperessibility can affect the solution. If you are reaching speeds in excess of Mach 0.5, consider a density based solver. Pressure based solvers are more efficient at lower Mach numbers but the D.B.S. is more accurate when there is density changes are present.

Alex C. · January 23, 2015, 11:54

Quote:

Originally Posted by jpando

If you are reaching speeds in excess of Mach 0.5, consider a density based solver..

I strongly prefer to use density solver from Mach 0.3. The importance of compressibility in the equation is following Mach^2. So Mach=0.3 means the compressibility effect will at maximum have an order of magnitude less than you pressure gradient in the momentum equation.

Although, it is unlikely that a ship is going at compressible velocities. Maximum speed reached by boat is around 140 m/s. And speed of sound in water is around 1,490 m/s. Therefore, Mach is around 0.1. In that case, the compressiblity effects will be 1% of the pressure gradient.

flinde · February 1, 2015, 08:05

I tried to decrease the first cell on the wall as well as increasing the number of layers in my boundary layer cell mesh, but it did not change the results. I also tried the compressibility, but unsurprisingly it did not change anything...
Any other idea where this might come from?
Thx

Nasper95 · July 16, 2021, 01:40

I know it's a very old thread but did you find the answer? I'm in a somewhat similar situation. Thanks in advance!

January 21, 2015, 08:27	Drag and lift understimation on a rigid body	#1
flinde New Member Florian Join Date: Jun 2013 Posts: 8 Rep Power: 12	Hi all, I am currently using CFD to predict hydrodynamic forces on a ship in restricted waters. For that I am using ansys fluent software. So far I obtained a good correspondance between my numerical results and experimental data. However, for higher fluid velocity, I am underestimating the drag and lift coefficient (ie horizontal and vertical forces). I studied the influence of several parameters such as mesh density, turbulence model (kE, kW, kW-SST), boundary conditions at the inlet (mass flow inlet, pressure inlet) and outlet (pressure outlet, outflow), domain size, change of the spatial discretizaion scheme,... But none did have an influence on the results. I am inclined to think that the pressure around the hull is underestimated but i do not know where this could come from... Any help would be appreciated. Thanks

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
wrong SU2 calculation for lift and drag coefficient for NAC4421	mechy	SU2	7	January 9, 2017 05:18
Beginner questions - modelling lift and drag on a towed body	ABF	Main CFD Forum	15	October 19, 2014 23:27
[OpenFOAM] Display lift and Drag in paraview	SamerAli	ParaView	1	May 16, 2013 12:51
Measuring Lift and Drag on Moving Body	sheth	FLUENT	0	March 20, 2012 10:32
CFX - Flow Over a Moving Body (Lift & Drag)	rossbardon	CFX	2	January 6, 2010 19:14

January 21, 2015, 15:15		#2
t.teschner Senior Member Tom-Robin Teschner Join Date: Dec 2011 Location: Cranfield, UK Posts: 204 Rep Power: 16	have you checked the pressure distribution with the laminar model (i.e. no turbulence model) and checked if the lift is predicted correctly? i would argue that at least the pressure field should be captured somewhat realistically (of course you don't get drag). if the pressure is off then there might be some issue with the set up / model.

January 23, 2015, 08:59		#3
flinde New Member Florian Join Date: Jun 2013 Posts: 8 Rep Power: 12	I ran a simulation with the laminar model and just like with the turbulence model, the lift is underestimated. According to you, there might be a problem with the setup / model. But given that i already checked parameters such as mesh sensitivity and numerical parameters i don't know where the problem could come from, especially given that it works well with lower speeds... Maybe I didin't check thoroughly enough those parameters... But then, what should I check first, from what could this problem arise from? Thanks in advance for your help.

January 23, 2015, 09:54		#4
jpando New Member Join Date: May 2012 Posts: 26 Rep Power: 13	Flinde, The issue to me looks like your y+ value may be wrong. As the velocity of the fluid increases, this also increases your y+. To maintain the same value of y+ as the lower velocity cases, you must decrease the first cell height on the wall. You also want to make sure your inflation layers traverse the entire height of boundary layer. This will make sure you have good quality cells in the most imporant areas of the mesh (where your lift and drag forces are generated). If this doesn't work, then you will want to check your compressibility model. Velocities around the body can reach speeds where comperessibility can affect the solution. If you are reaching speeds in excess of Mach 0.5, consider a density based solver. Pressure based solvers are more efficient at lower Mach numbers but the D.B.S. is more accurate when there is density changes are present.

February 1, 2015, 08:05		#6
flinde New Member Florian Join Date: Jun 2013 Posts: 8 Rep Power: 12	I tried to decrease the first cell on the wall as well as increasing the number of layers in my boundary layer cell mesh, but it did not change the results. I also tried the compressibility, but unsurprisingly it did not change anything... Any other idea where this might come from? Thx

July 16, 2021, 01:40		#7
Nasper95 New Member Join Date: May 2021 Posts: 1 Rep Power: 0	I know it's a very old thread but did you find the answer? I'm in a somewhat similar situation. Thanks in advance!