Is there a tigl function to query fuselage points on an arbitrary plane?

[kgntci]

Dear community,

I have fuselage with three dimensional profile points with curved points with coordinate components in the x-direction. Querying the points to generate a mesh using the standard eta-zeta convention builds up a cross section in space rather than on plane. In order overcome this issue, I want query points for a given eta and on a plane. Is there any function on tigl to provide such query for the fuselages? I think a similar function is provided for the wings (e.g. tiglWingGetUpperPointAtDirection).

Thank you.

[joergbrech]

TiGL provides some simple Boolean operations via its C-interface, see https://dlr-sc.github.io/tigl/doc/latest/group__BooleanFunctions.html. For instance, with tiglIntersectWithPlane you can calculate the intersection of a geometric component which is specified by its uid (e.g. a fuselage) with a plane. Then you can use tiglIntersectGetPoint to query points on the intersection of the plane and the geometric component.

Note however, that TiGL only supports planar profiles. If you use nonplanar profiles, there is no guarantee that the geometry generation works. What is your use case? Why do you need non-planar fuselage profiles?

[kgntci]

Hi Jan, thank you for your answer. I've used the functions you've mentioned and it helped a lot.

About my case: I've been reproducing the configurations that I size in openVSP for CPACS using the openVSP API. Currently, I can rebuild most of the VSP Geoms in CPACS as <wing>, as <rotor> or as <fuselage>. Here, I go by the simple section and segment implementations (no positioning), where I place as many <sections> as the tessellations in the VSP Geom and connect these sections sequentially in <segments>. Then I query the surface points in the Geom and write them in <profiles> (not the best way to go but it works!). In certain cases, the VSP Geom may consist of 3D profiles in the tessellation, which leads to the origin of my question above. The resulting surfaces look OK in the TIGLViewer, however querying the surface points using eta / zeta coordinates cause wavey meshes, since the eta/zeta coordinates mainly follow the curved section. In order to achieve 2D profiles, I thought of slicing the fuselage surfaces with planes and read the intersecting points to create the mesh.

tiglIntersectWithPlane and tiglIntersectGetPoint mainly solves the problem in flattening the section curves, however I realize that each intersecting line starts at an arbitrary point on the surface such that the section points not align properly, altgough the dimensions match. The picture below shows a portion (7 sections, 6 segments) of the fuselage, where such misalignment takes place.

Is this a known issue? I'm thinking of some quick solutions in the post processing but I would be very pleased to know, if there's already a solution to that.

Regards
Kagan

[joergbrech]

No, this is not a known issue and seems to be a peculiarity of the Boolean operation performed with OCCT under the hood.

The picture looks like the third intersection curve is oriented in the other direction as all other intersection curves...

From which language are you using TiGL? Are you limited to using the C API?