Geometry

A Geometry represents a 3D CAD model. You can import a geometry by uploading a CAD file, and you can modify existing geometries using boolean operations and transformations. This is useful when you want to run a large number of simulations with different geometric variants.

The sections below describe some of the most common operations you can perform on a Geometry. The full reference for the class can be found here.

Creating a Geometry

To create a geometry, you must first create a project to contain it. See Creating a Project for more info.

Once you’ve created a project, you can create a geometry by uploading a CAD file.

my_geometry = project.upload_geometry(
  "path/to/cad/file.stp",
  scaling=1.0,
  wait=True,
)

The only required parameter is the path to the CAD file.

The scaling parameter is optional and defaults to 1.0. If your geometry is in a different unit system, you can use this parameter to scale it to the desired unit system.

The wait parameter is also optional and defaults to False. If wait is True, the method will block until the geometry is fully uploaded and ready to use. It’s recommended to set this to True when you’re first getting started with the SDK, but you can set it to False to achieve higher throughput when setting up a large number of simulations in parallel.

See upload_geometry() for the full syntax and parameters.

Modifying a Geometry

Once you’ve created a geometry, you can modify it.

The recommended way to do this is by using select_volumes(). Once you’ve selected the volumes you want, you can apply the desired set of boolean operations and transformations on them. If an operation creates or destroys a volume, the selection is updated to reflect that.

For example, suppose you have a geometry that contains a model of a car, and you want to convert this into a wind tunnel volume that surrounds the car.

To start, we can select all the volumes in the car, then shrink wrap them into a single volume.

_, all_volumes = my_geometry.list_entities()

car = my_geometry.select_volumes(all_volumes)
car.shrinkwrap(min_resolution=0.01, max_resolution=0.04)

The shrinkwrap() operation replaces the car volumes with a single volume that is the shrink wrapped version of the car, which is guaranteed to be watertight. The volume selection is updated so that instead of containing the car volumes, it now contains the shrink wrapped volume.

Next, we can add a wind tunnel volume around the car.

windtunnel = geometry.select_volumes([])
windtunnel.import_cad(windtunnel_filepath)

Here, we’ve started with an empty selection, then imported a second CAD file to create a wind tunnel volume. The selection is updated to include the newly created wind tunnel volume.

Finally, we can subtract the car volume from the wind tunnel volume, and tag the resulting volume as “windtunnel”. Notice that we can use car.volumes() to get the shrink wrapped car volume from before.

windtunnel.subtract(car.volumes())
windtunnel.tag("windtunnel")

You can find the full reference of supported operations here.