WSP pile Process in Plaxis 2D

Question

Hi all and Dennis Waterman Rushan Gizzatullin Stefanos Papavasileiou 
 
 Task: I need to model the WSP pile installation in the soil and see the influence zone of it to do the impact assessment. The pressure varies from 2MPa to 20MPa as shown in the table below. 
 
 This huge pressure can't be applied as line load as soil will be collapsed. 
 Solution: So applied as the water pressure using user-defined option. 
 
 Problem encountered: Normally while pipe is moving grouting is ejected circumference as shown below. 
 
 but when we provided that pressure as water pressure (user defined) the pressure distribution is all four sides as shown below. 
 
 Due to this additional pressure up and down applied and when the model reached the top surface (due to the vertical pressure) soil seemed to be collapsed and to encounter this same pressure should be applied from the top to pass (if it is small one). If it is large pressure then line load tips are seemed to be collapsed as shown in the below figures.

Output => 
 
 So please tell how to model this grouting pressure in Plaxis 2D ? 
 
 Thanks.

Stefanos Papavasileiou · Accepted Answer

Dear Nitha, 
 Typically I recommend running a simple, but realistic situation and try to understand the behaviour. 
 The way the volumetric strain works is explained here: https://communities.bentley.com/products/geotech-analysis/w/plaxis-soilvision-wiki/45993/how-does-the-option-volumetric-strain-work 
 In your case, you see no change in stresses as there is no constraint to the cluster you apply the volumetric strain to. As the article explains, a stress equilibrium is found between all the relevant clusters and/or boundary conditions with regards to the cluster that the volumetric strain is applied. 
 Try creating a polygon inside a polygon and then add a volumetric strain in the inside polygon.

Dennis Waterman · Answer

Dear Nitha, Some remarks: 
 1) Note that you're applying a volume strain....hence, a positive volume strain is an increase of volume and a negative volume strain is a decrease of volume. The positive/negative is not linked to the positive/negative axis as you assume. Hence, currently you tell PLAXIS that the left half of the pile should shrink 0.207 and the right half should expand 0.207. This will complete level out, so no force is applied to the soil. You should fill in both volume strains as positive to achieve an expansion of the pile. 2) Be aware that the final diameter of the pile is the area over which you apply the strain + the expansion. So if you need to have a pile of 1.6m diameter, you should model the initial area over which you apply the strain smaller than 1.6m so that the strain can increase it to 1.6m 3) Why would you model the JGP as MC model? It's liquid grout and I presume failure of liquid grout is not of importance here. It's easier to model the JGP as Linear Elastic... 4) In reality the liquid grout can move relative to the soil (it flows), but in the PLAXIS model the liquid grout is a soil and thus connected to the surrounding soil which may give strange strains and local failure in the soil directly next to the volume strain. This can be overcome by assuring the grout material set has sufficient low stiffness, or sometimes it may be necessary to apply an interface around the grout area. The interface should have it's own material set with the stiffness according to the soil and low strength. Or alternatively, use the ks/ kn setting and specify normal stiffness (kn) equal to the soil and shear stiffness (ks) low. 5) Finally, as an alternative, you could decide to not use a volume strain but a user-defined pore pressure. To do so, deactivate the soil that represents the grout, apply the user-defined pore pressure according to the JGP you calculated and after that activate the material set representing the hardened (stiff) grout and deactivate the JGP. With kind regards, Dennis Waterman

Dennis Waterman · Answer

Dear Nitha, 
 I already explained that when using a user-defined pore pressure the pressure should decrease with height as in reality the grout pressure just below ground level is smaller than the grout pressure at depth in order to get a regular column. But you applied a constant pressure over the full height. So basically you have 2 systems that work now, you only have to fine-tune either the volume strain or the user-defined pressure in order to get the correct pile diameter. In order to get reliable displacements/forces at the tunnel you need to use an advanced soil model as already concluded before. 
 With kind regards, 
 Dennis Waterman

Stefanos Papavasileiou · Answer

Dear Nitha, 
 Indeed, your approach is correct. Alternatively, you can also simulate it by applying a volumetric strain to the cluster. 
 Now, the problem arises at the top material as there is no top pressure to handle the water pressure you apply. That is to be expected as if you apply a pressure to the model boundary the soil has an easy way out which is towards the free sky. That is a problem as we try to simulate what happens in reality but in a finite element world. 
 You show a small vertical load applied which seems to solve the problem. That can be acceptable if the vertical load is rather small. Another option is to define a separate material only for the top cluster and add some extra strength to prevent local failure. In all cases, make sure to refine that area to get more accurate results.

Stefanos Papavasileiou · Answer

Dear Nitha, 
 We have information about this feature in our Reference manual. You can apply a volumetric strain per direction or a total value. I would recommend you try a few models and if you have questions, you can reach out via a service request. 
 Applying a high load can cause local failure. That happens in reality which sometimes requires some practical solutions on the field but do not affect the overall project. In principle whether you apply a load or a soil volume with equivalent weight should not have an impact on the results. The only difference is that a load is concentrated on a single line while a volume is physically represented in the model. 
 You can consider applying smaller loading steps, e.g. max load fraction per step, to make it easier for the calculation and solve smaller loading steps.

Dennis Waterman · Answer

Dear Abdirahman, 
 1) From a quick check it seems ok, but I didn't check all the details as this falls outside the scope of this forum 
 2) A 10 MPa pressure, hence a 100 bar pressure, is not a small pressure. If in reality one would maintain a 100 bar pressure in the last meter to ground level a massive blowout would occur. So yes, the high displacements are realistic, the 100 bar is not. 
 3) Why would you assign a material at all when the grout pressure is still applied? The grout is at this stage still liquid and doesn't have a mechanical stiffness. By applying the material with a stiffness you limit the displacement due to the pressure. The procedure is to a) Excavate soil + apply pressure and then b) Activate + assign pile material + remove pressure 
 4) I'm really not sure what you're asking here. You apply a user-defined pore pressure and pore pressures are the same in any direction. But I don't see what that has to do with rotating piles or copy them to get 4 piles? 
 5) I could think of a couple of reasons for this but based on the provided information I wouldn't know. With kind regards, 
 Dennis Waterman

Dennis Waterman · Answer

Dear Abdirahman, Happy New Year to you too I'm afraid you didn't understand it correctly .... but I now understand that you're doing it a bit different as I originally intended. In it's simplest form it's enough to only define the nozzle area and then use 2 phases per section that you grout: 
 
 Phase A
 
 Excavate nozzle area 
 Apply water pressure

Phase B
 
 Activate soil in the nozzle area again 
 Change the nozzle area to the grout material 
 Remove water pressure

Since PLAXIS for weights will always consider the original volume and not the volume after deformation, the weight of your grout column will be based on the nozzle area. You can compensate this by making the grout heavier with a factor equal to the ratio of volume of the jet grout area and nozzle area (hence sqr(D_grout / D_nozzle)) What you do with a separate nozzle &aacute;nd grout area probably works too, I just never tried it before. In this case in your Phase B you also have to set the grout area to the grout material and there is no need for weight compensation because the grout material is now assigned to an area that is initially already the entire grout area. So what you do has the advantage that you have better control over the weight of the grout column. As I mentioned, a water pressure works in all directions because that is how fluids are. Kind regards, Dennis Waterman

WSP pile Process in Plaxis 2D

Top Replies