FEM Modelling

Dear Cristina,

It depends a bit on the type of analysis  you're doing, particularly seismic analysis needs far larger models than static analyis.
But as a rule of thumb if you look at the construction zone D x W (depth, width of an excavation, embankment, foundation etc) and you take the largest dimension of that area, then use a model with in all directions at least 3x times that largest dimension. So for an embankment that largest dimension is usually the width W, so use 3*W on either side of the embankment ánd also below the embankment.

Under specific conditions a smaller model may be good enough, but there is not really such a thing as a too large model, but there can be a too small model.
If the model is too small it may underestimate deformations and overestimate factor of safety and/or failure load.

The idea is that a boundary is so far away from the construction zone that stresses close to the boundary remain (almost) unchanged from the initial stresses during the construction. The lower boundary may experience a stress increase that is max 10% of the applied load. 

With kind regards,

Dennis Waterman

Dear Cristina,

It depends a bit on the type of analysis  you're doing, particularly seismic analysis needs far larger models than static analyis.
But as a rule of thumb if you look at the construction zone D x W (depth, width of an excavation, embankment, foundation etc) and you take the largest dimension of that area, then use a model with in all directions at least 3x times that largest dimension. So for an embankment that largest dimension is usually the width W, so use 3*W on either side of the embankment ánd also below the embankment.

Under specific conditions a smaller model may be good enough, but there is not really such a thing as a too large model, but there can be a too small model.
If the model is too small it may underestimate deformations and overestimate factor of safety and/or failure load.

The idea is that a boundary is so far away from the construction zone that stresses close to the boundary remain (almost) unchanged from the initial stresses during the construction. The lower boundary may experience a stress increase that is max 10% of the applied load. 

With kind regards,

Dennis Waterman

Dear  Dennis, thank you for your quick response. It will be helpful. So, you do you have any reference that I can use to cite in my text?

Regards

Hi Cristina,

We have gathered a lot of this information as rules of thumb in a presentation. In this presentation we refer to the following literature, so maybe you want to use them too:

Potts D.M., Zdravkovic L. (2001). Finite element analysis in geotechnical engineering – Application. Thomas Telford, London.

Meiβner H. (2002). Baugruben – Empfehlungen des Arbeitskreises 1.6 “Numerik in der Geotechnik”, Abschnitt 3, Geotechnik 25, 44-46.

Schweiger H.F. (2002). Musterlösung und Parameterstudie für dreifach verankerte Baugrube,
Geotechnik 25, 101-109.

Ruse N.M. (2003). Räumliche Betrachtung der Standsicherheit der Ortsbrust beim Tunnelvortrieb.
PhD thesis. Institut für Geotechnik. Universität Stuttgart.

Vermeer P.A., Wehnert M. (2005). Beispiele von FE-Anwendungen – Man lernt nie aus. In: FEM in der Geotechnik (ed. Grabe et.al.). Technische Universität Hamburg-Harburg.

Brinkgreve R.B.J, Bakker K.J., Bonnier P.G. (2006). The relevance of small-strain stiffness in numerical simulation of excavation and tunnelling projects. In: NUMGE 2006 (ed. Schweiger). Taylor & Francis, London. 133-139.

With kind regards,

Dennis Waterman

Excellent Dennis.

Thank you a lot.

Regards