Jump to content

Wave simulation for CumulusX!


Peter Lürkens

Recommended Posts

Hi folks,

not being to enthusiastic at this time, I'd like to keep you informed about a first milestone of simulating waves over mountains. Based mainly on this paper of Rottman and Broutman, I succeded in realising a simulation of a wave over a single mountain:

wave15k.png

The green-brown surface depicts the ground, while the blue plane indicates the lift in 15.000 m. The intensity of the lift is expressed in 1000m/(m/s) displacement. The mountain height is 1500m, the wind 10m/s. The area shown covers 120x160 km^2.

This a very first step and only shows, that one can calculate something reasonable. It's still quite a way until it will become possible in real-time (without drop of FSP, of course).

Cheers,

Peter

  • Upvote 3
Link to comment
Share on other sites

Eh Peter

The graphics link to your picture is broken at this point. Wonderful paper! Getting the efficiency of the algorithm up is a big challenge.

Here are other resources for interested enthusiasts of wave lift:

B21's post on wave lift June last year:

http://www.forum.aer...showtopic=35874

Great real world practical information on wave lift flying:

http://www.canberrag...waveFlying.html

Cheers

Link to comment
Share on other sites

Awesome stuff. So I can maybe hope that 2011 will be the year of "The WAVe" in FSX.

Great work, Peter, and keep it as real as it gets inside FSX if the wave can be achieved. Thanks for your continued efforts to keep the latest in the Flight Simulator series going strong.

Scott

Link to comment
Share on other sites

Yeah beautiful shot awe inspiring paper as well. What to do in terms of pre-compute/real time calculation with two mountains next to each other in FSX unsure.gif

Cheers

Link to comment
Share on other sites

Hi folks,

another update, see this animation of a vertical profile of gravitational waves behind an obstacle. Data as above.

Cheers,

Peter

Yeah thanks. The bit I'm amazed at is that the wave height oscillates with altitude. blink.gif I thought that it would set up wave's but that each wave would be x-shape, y distance apart, and taper off in height gradually by z-amount. This is a difficult problem for FSX when you see what is actually going on.

Cheers

Link to comment
Share on other sites

Hi folks,

another update, see this animation of a vertical profile of gravitational waves behind an obstacle. Data as above.

Cheers,

Peter

The wave seems to move forward with altitude. blink.gif

Cheers

Link to comment
Share on other sites

I thought that it would set up wave's but that each wave would be x-shape, y distance apart, and taper off in height gradually by z-amount.

That's a mistake I made initially, too. But literature is quite clear here. I must state, however, that the above situation is heavily simplified, as it assumes a homogeneous wind situation ("background"), which is typically not given in reality.

The question now is, if that method generates wave conditions which are sufficiently in coincidence with what is known from various regions, that it can be accepted. Otherwise, Ian's proposal of a "wave map" is possibly more attractive, however, would be limited to known regions and a one or two wind situations only. It would be a lot of work to assemble these information into "wave maps".

I truly hope that the analytical approach will eventually lead to acceptable results.

Cheers,

Peter

(Edit) To avoidmisunderstanding: The animation shows the quasi-static situation in each layer, not a wave travelling from bottom to the top. I.e., the wave situation in each layer is constant over time. However, it is not the same pattern which is merely shifted downstream.

Link to comment
Share on other sites

I think the landmark feature of the ridgelift calculation in FSX is that a '3D' problem was turned into a '2D' problem yet still gave realistic results, and this is the fundamental breakthrough required in wave calculation to allow per-frame effect calculation. My suggestion of the use of geographic 'wave map' textures in part suggests the use of GPU operators might be possible, i.e. literally use image textures.

The proper compute step could be as sophisticated as we like, and crank through wave textures at different heights and for different wind directions taking as long as necessary. But the real-time calculation would use these pre-compiled textures.

B21

Link to comment
Share on other sites

Guest
This topic is now closed to further replies.
×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue. Privacy Policy & Terms of Use