Hello everyone, I am currently working on my master thesis in structural engineering, focusing on advanced modelling and parametric analysis of lateral load capacity of a concrete pile with varying pile cap rigidity .
The aim is to develop model for a laterally loaded concrete pile, with numerical simulations through a Python script integrated with the RFEM 6 API.
What I am looking for:
- Best practices for modelling nonlinear soil springs together with second order effects
- Tips on automating model creation and parameter variation with Python.
- Any references, workflows, or examples that could help validate or benchmark my approach.
I would greatly appreciate any guidance, shared experiences, or resources from the community.
Thank you in advance for your support!
Hi studentInu11,
This is a very interesting topic!
I would recommend the following workflow:
- Try to model one variant with RFEM 6, using the member type pile from the geotechnical analysis add-on. Define global parameters for easy adaptability. For second-order effects see this knowledge base article.
- Export your model as a Python script.
- Afterwards, you can adjust the model easily and perform a parameter study. Helpful will be our API documentation by changing the code.
You can also compare our implemented member type āpileā by other approaches, p.e., p-y-curves. This method will also be available for RFEM 6 in the near future. But for now, you may model this behavior by yourself; please see these references for more information (1, 2, 3 and 4).
Some related examples from our website are this knowledge base article for parametric pile modeling and this recorded webinar for pile analysis. In the last one, a comparison to literature is also done.
If you wish, you can also submit your thesis to us after completion and receive great benefits:
āŗ https://www.dlubal.com/en/education/students-and-schools/why-submit-your-thesis
Best regrads,
Marc Gebhardt
1 Like
Hello Marc! Iām very glad you found my post.
Iāve recently started working fullātime on this project, and my current approach is to model the pile as a beam of length šæ, discretized into a series of beam elements. At each node, I assign a nonlinear (diagram-based) lateral support that represents the soil stiffness. For this, I use pāy curves based on Matlockās formulation for soft clays. I also draw the pile according to the buckling mode shapes to see the effect from initial imperfections.
I know there are modules for doing this, and after reading your suggestion Iām considering creating a separate analysis variant using the approach you described.
Much appriciated for the links and I will most likely sumbit my work upon completion!
2 Likes
Is there any downside to model the beam in multiple segments. The following page, states that its not needed when acounting for second order effects.
Great, I'm glad my answer was helpful!
In my opinion there are none essential - besides a more or less rise for computational effort if this segmentation leads to a finer mesh. At least this is true in our software, because beams sharing a node are also sharing all DOFās by default. (There are just two subordinate points that are coming to my mind at the moment: The segments are not treated like one for torsional warping (the 7th DOF) - at least in the current state and for some checks in the design add-on. If this is a problem, you can separate the members by nodes on them, without ārealā separation, or by defining member sets.)
Looking forward to reading your thesis upon completion! 
1 Like