HI All,
I referred to the tutorial manual to check the suggested procedure for modelling TBM tunnelling in Plaxis 2D. Chapter 6 suggests the following stages:
1- activation of the lining
2- application of contraction representing the volume loss (i.e. the volume loss method is being used)
3- deactivation of the lining and application of the grouting pressure
4- reactivation of the lining
This procedure seems reasonable from a 3D perspective. However from a 2D plain strain perspective stages 3 and 4 are problematic. Stage 3 is assuming a case where the soil across the whole length of the tunnel is supported by grout pressure. This is resulting in stress redistribution in which radial support of the soil is all that is needed to support the ground. Once the lining is reactivated in stage 4, the forces in the lining are mostly normal forces with a very small bending moment which is not what is expected.
Would you please help in clarifying the logic behind those stages? Can someone point to a published paper in which this procedure is suggested?
Thank you!
Dear Omar,
To start with, you are referring to a PLAXIS tutorial example. This is not supposed to be an engineering guide on how to model the TBM but merely a functionality tutorial on how to use PLAXIS 2D. Therefore, if you disagree with or have a better way of modelling this on PLAXIS 2D, please use your engineering judgment and experience.
Back to the tutorial we offer, in phase 3, in addition to the installation of tunnel lining, excavation of soil and dewatering of tunnel is done. However, the volume loss is simulated by activating the Line contraction feature. In phase 4 the grout pressure is applied, and the tunnel lining is deactivated because in PLAXIS 2D it is the only way to apply grout pressure to the surrounding soil/rock.
Reality is different, of course, the tail void grout is applied in between the soil/rock and lining but in the model, it is applied inside the whole tunnel (since the exact behaviour of the TBM during driving remains uncertain, considering the lining in the model is not that feasible). The only way to do this is by temporarily deactivating the lining. So, in the model, the grout is considered to take all the load. Hence the phase 4 calculation is not what is done in reality, but it is the net effect that should be as close to what is possible in 2D plane strain conditions.
Should you have any better suggestions, please feel free to share.
Thank you for the clarification Stefanos! I appreciate taking the time to reply to my question!
Based on your response, I think that I will not model the grouting phase and stop at the stage where the contraction is applied (or use the deconfinement method or just do a quick 3D).
Thanks again!
Both options you mention are good.
Especially if you have PLAXIS 3D, then there is no doubt, as the effect of excavating a tunnel with a TBM is more of a 3D problem.