Ground effect models

[seanmcleod70]

I've written up a brief description and comparison of 3 different ground effect models, one being a fairly common one found in a number of JSBSim aircraft like the 737 model, and another one being a Boeing supplied full simulator model for the 747-100 years ago for NASA Ames, and lastly an Airbus provided model for use in a challenge for developing a better autoland control system.

https://seanmcleod70.github.io/FlightDynamicsCalcs/GroundEffectModels.html

Hopefully it might be useful for JSBSim modelers looking to add or improve the ground effect model in their particular FDM.

[bcoconni]

Very interesting @seanmcleod70.

This gave me an idea and I have created a notebook on Google Colab where, following your inspiration, a simple aerodynamics model (using the Vortex Lattice Method) has been used to compute the ground effect on a basic 737 unswept wing.

Numerical computation of the ground effect - Google Colab.

Well, believe or not, the computations are in the same ballpark than JSBSim. Of course, my code most likely contain some bugs but I found this interesting nonetheless.

[agodemar]

@bcoconni very good discussion, thank you for sharing.
For the sake of completeness, I would add proper labels to the axes in your final plots: $h/b$, for x-axes; $k_{C_L,\mathrm{ge}}$ and $k_{C_D,\mathrm{ge}}$ for the y-axes.

[seanmcleod70]

@bcoconni nice notebook and implementation.

While looking into ground effect models I came across a number of references to using mirror image modelling to model the ground effect.

One of the references I had in my writeup that I took a look at was this paper - Lifting-Line Predictions for Induced Drag and Lift in Ground Effect

But hadn't come across any code examples like your notebook in terms of actually modelling it.

I also came across this from Flight Dynamics - Robert Stengel, Second Edition pg. 105

While the ground effect on the isolated wing often is estimated using lifting-line theory, Kuchemann warns that this is an oversimplication. It ignores wing thickness effects, and it is inaccurate for swept wings, where the height of the root and tip are quite different at even modest angle of attack. Furthermore, ground effect is affected by the rate of change of altitude. With these caveats in mind, the simplified approach predicts that ground proximity has the same effect on the flow as increasing the wing's aspect ratio. This, in turn, increases the lift slope and decreases induced drag, as generally observed in practice.

I didn't figure out the original source for the JSBSim ground effect data.