Regolith (or soil) load

[Ale22pic]

Hi everybody, I'm working on my thesis and the project is about a martian habitat. So for the moment, I'm trying to analyze a hemispherical structure that is represented in the figure. The problem is that I'm searching for a rapid way to describe the shielding regolith (or soil it depends on how do you want to call it) around the structure. Even the shielding regolith has a hemispherical shape. I already studied a single arch and now I'm studying a complete structure. I thought to use a contact analysis, for example, to create a deformable structure, with regolith mass properties, on the habitat but I'm not totally sure is possible.

 

[STrctPono]

Would this regolith be unconsolidated? Because that would make a huge difference in how the analysis was run. You describe an analysis where you would model the hemispherical structure with the non-linear properties of the regolith mass above. I actually have some great technical articles on how to do this and the key is in the mohr-coulomb properties you assign to the soil coupled with the way in which you model your contact boundary conditions with the structure. How thick is the soil mass? Because if it is not substantially thick and is not consolidated, I'm not sure you would see any soil arching benefits that you could gain from running a non-linear soil structure interaction analysis. In essence, with the way you describe it, I would think the regolith would be nothing more than a dead load mass atop your structure.

 

[Ale22pic]

Hi STrctPono, thanks for your answer. Actually my structure is covered with a thick soil mass, circa 1.650 meters and it is not solid but deformable. Image to have a sphere and cover it with sand to achieve a kind of dome.

 

[STrctPono]

This is a pretty significant analysis that is going to take a ton of reading on your part. Do you have access to a FEA program with an explicit solver that can model soil continuums? Typically this is done using plane-strain elements to represent the backfill material. You will also need to model the soil foundation elements. Current practice is to use Drucker-Prager or Mohr-Coulomb constitutive models. You will need to know the soils elastic properties such as the elastic modulus, poisson's ratio, and weight density. You will additionally need the material non-linear properties for the constitutive model that can handle cohesion, friction angle, and either isotropic or kinematic hardening. Modeling the material boundary contact connections to the structure can be handled in a multitude of ways. I have modeled a thin layer of material between the soil continuum and structure using the same properties as the soil but with 1/2 cohesion and friction angle.

You should start small and run a 2D analysis to make sure you are getting reasonably expected results before you dive into a full 3D model.

Good luck. I attached a few links to articles that helped me when I ran my first model.

Link

Link

Link