Control surface deflection results not matching NeuralFoil whitepaper #170
Description
Bug Description / Observed Behavior
I'm using neuralfoil/aerosandbox to generate lift/drag/moment curves for an airfoil with a deflected flap. I'm getting very different results between neuralfoil and XFOIL and after some digging I realized that I also wasn't matching the NeuralFoil white paper.
Perhaps I am missing something (like rotating or normalizing the airfoil) but based on the examples and the paper it seems like that should be handled internally. Any help is appreciated.
Steps to Reproduce
The code below reproduces the problematic aero results I observed (note that include_360_deg_effects is set to False to better match XFOIL when we have no control surface deflections).
import aerosandbox as asb
af = asb.Airfoil("naca0012")
controlSurfaces_l = [asb.ControlSurface(name = 'aileron', deflection = 30, hinge_point = 0.7)]
aero = af.get_aero_from_neuralfoil(alpha = 0, Re = 1e6, control_surfaces = controlSurfaces_l, include_360_deg_effects = False)
print(aero['CL'], aero['CD'], aero['CM'])
this prints: [1.23500284] [0.62995164] [7.1466822e-05] for CL, CD, and CM
which is off for CL and vastly different for CD and CM.
Expected Behavior
Running xfoil for this setup yields:
alpha CL CD CDp CM
0.000 1.3245 0.05461 0.04791 -0.1789
Looking at the NeuralFoil white paper, I would expect approx. the following:
CL = 1.42, CD = 0.05, CM = -0.21
System Information
- Operating System: Windows 11
- Python Version: Python 3.11.4
- AeroSandbox Version: 4.2.8