>
Would you please provide a simple example on how to do a scale/shift/rotation on a bdf model after it is loaded.
Those are all very different. There isn't a one liner to do scale/shift/rotation. The BDF class is focused mainly on querying data (e.g., mass, load, any attribute on any Nastran card).
If you just use the global cid=0 coordinate system, it's very straightforward to modify the nodes directly (see the links below). If you have nodes referencing coords that reference nodes that reference coords, etc., the `node.get_position()` method is very useful. There is also a `node.get_position_wrt(model, cid)` method. It just depends on what you have and what you want (e.g., complex coordinate references, but you want to scale and rotate and maintain the same Cp coordinate system). If you don't mind switching to Cp=0, it's easier.
Once you apply your transform, just set the `node.xyz` and/or `node.cp` appropriately.
> If you can provide some further examples on how to remove part of the model
I don't follow. Is that not just `del model.elements[10]` to delete element 10 or was there something else you had in mind? Almost everything is a dictionary of objects or a dictionary of lists of objects, so it should be fairly straightforward if you're familiar with python and have the QRG handy.
> merge two models
Have you looked into bdf_merge? It's a blind merge, so it's going to assume that PSHELL 10 and PSHELL 10 from different decks are different properties. That might not be what you want.
> but it seems I have to understand lots of details before I can do the operations by myself.
There is quite the learning curve as you need to know Python, Nastran, and it's a lot of code. Once you understand the layout, it's pretty easy to find things. At some level, you just need to open the code, but I'd start by playing around in a Jupyter notebook. Most of it is fairly well documented. Looking at methods like bdf_interface/write_bdf.py are very useful for understanding how to loop over objects.
I highly recommend going through the tutorial found in the Jupyter notebooks. They're available online (and are in the docs folder).