Version 3.0.4 (Released at 06/09/2020)
- Minor bug in command widget is fixed.
- Minor bug in scripting language is fixed.

solve relax problem(Read 88 times)
solve relax problem on: June 28, 2020, 02:44:23 am

I encountered the following problem (for me) when playing around with the new solve-relax functionality:

modelling a tunnel (basically like tutorial 1) and using the top heading / bench geometry & rockmass properties from tutorial 7 I observe that when I am using solve-relax for the excavated top heading alone I get a max. total displacement of 0.0452m and if I use it for an excavated full face (i.e. excavate top heading & bench) I get a max. total displacement of 0.0439m.
If I do the same with the "normal" solve function and I solve to equilibrium, the top heading alone shows 0.276m - and the full face excavation 0.544m of max. tot displacement.

Where the "normal" solve function produces results as I would expect them (the bigger the hole, the bigger the deformations...), I don't understand what the reason is that the solve-relax function produces equally big displacements for both excavation geometries (the top heading even slightly bigger than the full face !?).

A sample script is attached to this post and I'd highly appreciate any help!

Re: solve relax problem Reply #1 on: June 29, 2020, 01:23:20 am
Unfortunately, I'm not sure if I understand the reasoning behind your comparison!

if you use "solve relax" command with 50% relaxation this means that only 50% of initial load applied on the boundary of the tunnel will be relaxed, as a result you should expect less displacement. it's like you applied some sort of support that resists against 50% of rock load.

but if you use "solve" command since there is no relaxation, 100% of the load initially applied on the boundary of the tunnel will be relaxed which should create maximum tunnel displacement (assuming tunnel is still stable).

did I answer your question correctly or you are looking for something else?

« Last Edit: June 29, 2020, 08:54:31 am by Roozbeh »

Re: solve relax problem Reply #2 on: July 01, 2020, 01:13:17 am
Hi Roozbeh,

thanks for the reply. My point / observation is that 50% "solve relax" generates roughly the same amount of displacement irrespective of the two tunnel geometries in the code attached to my first post. I find this strange, as I'd expect that 50% "solve relax" on the top heading excavation alone would lead to less displacement than 50% "solve relax" on the full face excavation without a preceding top heading excavation.

Re: solve relax problem Reply #3 on: July 01, 2020, 09:25:59 am
I see the reduction in the numbers. I measured 43 mm crown displacement for full face and 39 mm for top heading case.

there are couple of points:

1- please note that the example I provided in the tutorial is based on imaginary soil/rock properties and if you plot the element state you will realize that there is hardly any plastic zone around tunnel. I would recommend you to change the parameters and give it try.

2- Personally I have never checked it myself. But intuitively for this particular case, I would not expect to get huge changes in crown displacement since curvature of the tunnel has not been changed and also width of the tunnel is pretty much the same. Again it's my gut feeling that need to be tested.