Jump to content
TUFLOW Forum

DanielCopelin

Members
  • Content Count

    14
  • Joined

  • Last visited

Community Reputation

0 Neutral

About DanielCopelin

  • Rank
    Member

Recent Profile Visitors

The recent visitors block is disabled and is not being shown to other users.

  1. As per the title. Do the more expensive specialised Tesla and Quadro cards offer any advantages compared to GeForce cards for TUFLOW GPU?
  2. Thanks, Kate! This seems to have worked. I have upstream levels within 0.1m of HECRAS now.
  3. Hi, I have a weir (W) channel with a crest level of 19m as part of a 1D network. The weir geometry is being read from a mid cross-section file, and the check files seem to indicate that the geometry is being read correctly and the hydraulic properties seem correct. However, I am seeing very large afflux (approx 1.5m) across the structure even under very high flows where the weir is drowned by ~20m+ of water over the crest. By comparison, a steady state HECRAS model only predicts about 0.15m difference in upstream and downstream water level, which is more reasonable. If I model the W channels as normal S channels, I get results that are similar to HECRAS. I have a suspicion that the W channels are not using the submerged flow regime, but I don't know how to go about diagnosing this. I can't seem to find anything of relevance in the check files and .eof file. I am using the latest version of TUFLOW. Does anyone have any experience with this?
  4. Richie, Which version are you using? We were experiencing this issue on build 2011 AF DP 64, but is seems to have been fixed in 2012 AD DP 64.
  5. Phillip, Again, thanks for your assistance. Simply extending the SX line to select one additional cell upstream and downstream of the culverts (which is plausible and realistic in my case) seemed to fix the issue. I note the point amount smooth flow paths approaching the culvert in low flows, and I will ensure that I pay attention to this in the future.
  6. Thanks for the quick reply! Yes, I am using the Z flag. I will follow your suggestion and modify all of the Z points using a Z shape.
  7. Hi, I am modelling two culverts within a 2d domain. The culverts are both 5 x 5m dia. circular barrels, and the 2d grid size is 10m. I am using the usual SX / CN line method to select a number of cells upstream and downstream of the culverts. The Q results for the culverts show wildly oscillating flow on the rising and falling limbs of the hydrograph (although the results near the peak seem steady). I am seeking any tips on how to resolve this issue. So far I have tried: - Checking which 2d cells have been selected by the SX line. - Optimising the orientation of the 2d domain to ensure that a nice line of cells is selected and that approach flow is smooth. Are there any other things I should try? Is there a usual or common cause of this type of issue? Again, any help you can give would be greatly appreciated.
  8. Hi, I'm wondering if it is possible to create a 2D flood model using the normal Mapinfo / TUFLOW approach, run the model, and then later import that model for further development / interrogation in XP-SWMM? I know that this can be done to some extent layer by layer, but does this preserve all of the attributes in the TUFLOW model files? Also, I imagine that any 1D elements can't be transferred between the two programs. Any help would be appreciated.
  9. This is a great idea. Creating breaklines for ridges and gullies is currently one of the most time consuming tasks in model setup.
  10. Nadia, Thanks for the reply. I considered this. However, at high flows when those smaller HQ boundaries should be acting as one, there will be different relationships between H and Q for different parts of the floodplain would there not? I thought that would give rise to similar errors? Also, would there be an issue with the cell at the snapping point effectively having two boundaries defined?
  11. Hi, I'm having some issues using HQ boundaries at the downstream end of my model. I am modelling a river with a very broad and flat yet uneven floodplain. The river breaks out and flows away in 3 or 4 different directions. I used the results of a preliminary 1000-year ARI model to define a few automatic HQ boundaries normal to the flow direction. The results of the resulting 1000-year ARI event are fine. However, I have issues with the lesser ARI events when trying to use the same boundaries. In these events, there are multiple smaller "perched" flow paths on the section of the floodplain that is defined by a single HQ boundary. The HQ boundary is therefore inappropriately forcing a constant water level. My model doesn't go unstable, and it has a small mass error throughout. However, the Qo bounces between a high value and low value throughout the run. When I check hydrographs near the outlets using WaterRide (although the computational timesteps are obviously not visible), they appear smooth and all of my other results seem OK as well. The messages mif file doesn't show any negative depths associated with the downstream boundaries. I understand why this error is occurring, however my question is whether or not I should be concerned about the bouncing outflow. Could this be giving rise to any other problems that I need to be concerned about and just haven't noticed yet? If it is a significant issue, I cannot think how else to simply model the downstream boundary given the irregular terrain and varying flow conditions in the different events (short of having different boundary conditions for different events). A sample of my tlf showing the issue is included below: 39576 54:58:00 -d 0 0 Wet 203653 CS 28066 4890 CE 0.0% 0.0% 0.0% Qi 838 Qo 624 dV 50 39577 54:58:05 -d 0 0 Wet 203642 CS 28082 4917 CE 0.0% 0.0% 0.0% Qi 793 Qo 1423 dV -2130 39578 54:58:10 -d 0 0 Wet 203652 CS 28081 4913 CE 0.0% 0.0% 0.0% Qi 838 Qo 624 dV 49 39579 54:58:15 -d 0 0 Wet 203664 CS 28073 4890 CE 0.0% 0.0% 0.0% Qi 793 Qo 1423 dV -2132 39580 54:58:20 -d 0 0 Wet 203627 CS 28122 4931 CE 0.0% 0.0% 0.0% Qi 838 Qo 623 dV 49 39581 54:58:25 -d 0 0 Wet 203622 CS 28152 4916 CE 0.0% 0.0% 0.0% Qi 793 Qo 1422 dV -2116 39582 54:58:30 -d 0 0 Wet 203618 CS 28135 4928 CE 0.0% 0.0% 0.0% Qi 838 Qo 623 dV 43 39583 54:58:35 -d 0 0 Wet 203615 CS 28121 4905 CE 0.0% 0.0% 0.0% Qi 793 Qo 1422 dV -2127 39584 54:58:40 -d 0 0 Wet 203601 CS 28100 4920 CE 0.0% 0.0% 0.0% Qi 838 Qo 623 dV 59 39585 54:58:45 -d 0 0 Wet 203604 CS 28116 4948 CE 0.0% 0.0% 0.0% Qi 793 Qo 1422 dV -2133 39586 54:58:50 -d 0 0 Wet 203604 CS 28100 4940 CE 0.0% 0.0% 0.0% Qi 837 Qo 623 dV 42 39587 54:58:55 -d 0 0 Wet 203587 CS 28060 4921 CE 0.0% 0.0% 0.0% Qi 792 Qo 1422 dV -2125 39588 54:59:00 -d 0 0 Wet 203573 CS 28053 4907 CE 0.0% 0.0% 0.0% Qi 837 Qo 623 dV 50 39589 54:59:05 -d 0 0 Wet 203548 CS 28059 4933 CE 0.0% 0.0% 0.0% Qi 792 Qo 1422 dV -2129 39590 54:59:10 -d 0 0 Wet 203550 CS 28065 4922 CE 0.0% 0.0% 0.0% Qi 837 Qo 623 dV 46 39591 54:59:15 -d 0 0 Wet 203527 CS 28049 4921 CE 0.0% 0.0% 0.0% Qi 792 Qo 1422 dV -2133 39592 54:59:20 -d 0 0 Wet 203517 CS 28095 4897 CE 0.0% 0.0% 0.0% Qi 837 Qo 623 dV 43 39593 54:59:25 -d 0 0 Wet 203512 CS 28103 4937 CE 0.0% 0.0% 0.0% Qi 792 Qo 1421 dV -2132 39594 54:59:30 -d 0 0 Wet 203506 CS 28116 4942 CE 0.0% 0.0% 0.0% Qi 837 Qo 623 dV 44 39595 54:59:35 -d 0 0 Wet 203499 CS 28056 4914 CE 0.0% 0.0% 0.0% Qi 792 Qo 1421 dV -2125 39596 54:59:40 -d 0 0 Wet 203481 CS 28078 4914 CE 0.0% 0.0% 0.0% Qi 837 Qo 623 dV 45 39597 54:59:45 -d 0 0 Wet 203469 CS 28057 4950 CE 0.0% 0.0% 0.0% Qi 792 Qo 1421 dV -2129 39598 54:59:50 -d 0 0 Wet 203468 CS 28073 4929 CE 0.0% 0.0% 0.0% Qi 837 Qo 623 dV 45 39599 54:59:55 -d 0 0 Wet 203480 CS 28065 4949 CE 0.0% 0.0% 0.0% Qi 792 Qo 1421 dV -2132 39600 55:00:00 -d 0 0 Wet 203471 CS 28065 4936 CE 0.0% 0.0% 0.0% Qi 837 Qo 622 dV 44 Any help would be greatly appreciated. Regards, Dan Copelin
  12. Hi, I believe that I have found an error with the way TUFLOW handles centre (rather than end-point) cross-section lines when using .shp GIS files. My initial model setup had 1d_xs lines representing weirs that were mid-way along my 1d_nwk lines. I made sure that a vertex from my 1d_xs line was snapped to the 1d_nwk line. However, TUFLOW wouldn't recognise these cross sections and I would receive an error saying that my W channels had no CS data. When I redrew my 1d_xs weir lines as simple lines with only two vertices and made sure they intersected the 1d_nwk lines, the model would run without issue. Has anyone else encountered something similar?
  13. Thanks Phillip, I have just emailed support. The increased number of points will be useful. I used an interim work-around whereby I created a HECRAS project from 12d, used the HECRAS cross-section points filter tool, and then imported the filtered cross-sections back into 12d. Using that data, I exported a .12da file and then used the relevant TUFLOW utilities.
  14. Hi, I am using xsgenerator.exe on points and lines .mif files created by 12da_to_from_mif.exe (from draped cross-section lines in .12da format). However, I get an error saying that the maximum number of points per section has been exceeded. See output below. Nearly all of my cross-sections will have more than 500 points. Is there a way to get around this? Read C point 492 for XS 10 Read C point 493 for XS 10 Read C point 494 for XS 10 Read C point 495 for XS 10 Read C point 496 for XS 10 Read C point 497 for XS 10 Read C point 498 for XS 10 Read C point 499 for XS 10 Read C point 500 for XS 10 ERROR - Exceeded number of allowable points per cross-section section. Cross-section "10" in "trial_xs.12da.Points.mif"
×
×
  • Create New...