0 votes
in AEM3D by [email protected] (470 points)
edited by [email protected]

Hi Chris. I am simulating a reservoir that has very large changes in the surface level, so the inflow changes its bc location. So far I have been using flow_multi_sides to define the thalweg and, also, INFLOW_MAXDEPTH with the aim to set the inflow bc  which actually would be contribuing to the mass and momentum. However, the slope of the river in the entrance is very small, so I am not sure if the entire mass (and the momentum) is being distribuided in several cells. Should I use Inflow_mindepth as well?. Is my interpretation correct?

(1 day later) I want to complement my question by using one figure [see the link :  https://drive.google.com/file/d/1Dqaw7PeExcYgvnAgQqgEQ8n5fC69ZAd5/view?usp=sharing). I defined the flow_multi_sides from cells 1 to 6. At time T0 the surface level was such as high that the water comes from cell # 6 to cell 5 and, then, to the remaing cells (H in the cell 5 was higher than INFLOW_MAXDEPTH = Hmax). It means green arrows indicate the flow from bc and blue arrows are the flow computed by the NS equations. At Ti the bc moves from cell 6 to cell 5 as the water level dropped, it means the water mass comes from cell 5 to cell 4 and, then, to the remaing ones. I want to know if I understand this process correctly. My concern is about the mass and momentum distribution in different cells from bc. If, for example, the H (estimated by the model) is higher than the H-max in two cells defined as multi-sides (let's say cells 4 and 5 at time T0 in my figure) I suppose the algorithm must define the entrance only in cell 5, it means the water from outside just comes from cell 6 to cell 5. If the above is correct, the model would not distribuite water mass (from bc) into cell 5 and 4 at the same time (see green arrow + yellow arrow), decreasing the momentum at the entrance, which is my concern. Does it make sense?

Thank you in advance

Ricardo RB

1 Answer

0 votes
by [email protected] (19.4k points)
HI Ricardo,

The key component of the flow boundaries that I think your missing (it's not really clear in any of the documentation sorry) is that flow can only enter/leave the domain from "ghost cells",  that is, cells that can't be wet.

So in your 2nd image (T = Ti) the green arrows are incorrect.

What happens in the pre-processor is that all the valid bc cells are calculated from the bc definition in bc.dat.  So it looks at all cells/faces specified and see if the face has an interior cell on one face and a ghost cell on the other.  If so then this cell is included in the valid bc cells set.

If no inflow mindepth or maxdepth is specified then the when the flow is calculated in the model it comes through all these faces (the flow is area weighted so the velocity at each valid face is the same).

If mindepth or max depth is given then the code checks to see if any part of the face is within the valid range.  If there is no valid area for that face then the cell is removed from the valid set for that time step.

Hopefully that clears it up a little.  

I am planning on adding some additional output to the pre code to make it easier to see which cells are valid for each bc ref but I'm not sure when I'll have chance to do this.

Chris
by [email protected] (470 points)
edited by [email protected]
Hi Chris, Thank you for your answer. I was really confused as you said. Well, just let me know if in this case I understood your point. In this link (https://drive.google.com/file/d/1O9yJAxnOMERHsO6ph6XeOVsgbrRmdd3l/view?usp=sharing) I show a simulation of your old example InflowDepthExample where the level is rising as there is just one inflow, no outflow. Bathymetry is each 1m and the maxdepth is 1 m as well. In the figures I labeled "ghost cells" and the black arrows indicate the flow from outside and the blue ones are those (of course i am showing some of them just as example) computed by NS equations. In this case the bc is all i (:), j=2, k=all (:) it means the cells/faces which I marked should be the "ghost cells".

is it correct?
by [email protected] (19.4k points)
Hi, your image is correct as labelled, though if INFLOW_MAXDEPTH = 1m then in the 2nd image the 2nd layer will also have a valid inflow bc cell as part of the cell face will be above 1m depth.
Welcome to Hydronumerics Q&A, where you can ask questions and receive answers about the software we develop and distribute

Categories

Hydronumerics

  • Suite 103 757 Bourke St Docklands Victoria 3008
  • Suite 103 757 Bourke St Docklands Victoria 3008
  • +61 3 9037 0892
  • [email protected]
  • Twitter
...