Pile group using Strain Wedge Model

[BernardoLima]

Hi all,

I have been studying the SWM for pile groups. Some of the documents I’ve read is “Pile Group Program for Full Material Modeling and Progressive Failure” and “Lateral Behavior of Pile Groups in Layered Soils” both from M. Ashour and Gary Norris. I programed the method based on the algorithm described in Fig.8 of “Lateral Behavior of Pile Groups in Layered Soils” and it is at an advanced stage already. However, I still have some doubts about which I would be very grateful if you could clarify me.
Let's consider this example:
A pile cap with two piles is loaded along the strong axis.

For all loading increments (values of ε) converged wedge geometries are assessed for piles 1 and 2.
Converged wedge geometries at the top soil layer, corresponding do Pile 1 head displacement Δ, are geometry 1A and geometry 2A for pile 1 and 2 respectively.
However, after the reorganization described in step 11 of Fig.8, when Pile 2 has the same pile head displacement Δ (Pile 2 behaves like an isolated pile), the wedge geometry in pile 2 is different.
This way, when the pile cap has a displacement Δ, we have two different possible geometries for Pile 2:
- Wedge geometry 2A – when Pile 1 has displacement Δ;
- Wedge geometry 2B – when Pile 2 has displacement Δ.
Shouldn’t the geometry for Pile 2 be unique for a given pile cap displacement? Or am I missing something here?

If anyone can help me i would be very grateful!

 

[EireChch]

Post a sketch. Too much visualisation for me.

 

[BernardoLima]

Hi EireChch. Thank you for your reply. Attached the figueres of the described example.

 

[LRJ]

I'm a bit dubious about the strain wedge method, personally. It's okay for some applications, but you have to be careful that the underlying assumptions are still applicable.

 

[oldestguy]

When it comes to what really happens out there, a comportment contractor will drive his piling with minimum displacements happening by the sequence that he uses, such as starting at the center and progressing out.

 

[BernardoLima]

Hi LRJ,
In my opinion, SWM is a great method for isolated piles. However for pile groups, it seems to me that further investigation on a rational mathematical model, comproved by experimental "in situ" data, still have to come. But that's my personal opinion, of course. What do you think about it?

 

[LRJ]

My major gripe with SWM is that it is only really applicable for shallower piles where there is a wedge failure; for long slender piles, it is not appropriate. However, given that the most important influences are nearly always shallow, I appreciate it has some uses.

I've not looked at SWM extensively for pile group calculations - I vaguely recall looking at it (last year?) and thinking that it does have some promise since the wedge failure mechanism is fairly reasonable for pile group analyses (most interaction occurs at the top of the pile where wedge failure occurs) and offers advantages over some of the elasticity-based solutions. However, as pile group response is normally related to single pile response in some way, the single pile response needs to be appropriately calculated (i.e. check the validity of SWM assumptions).