Tutorial Module01 Archive: Difference between revisions
Content deleted Content added
No edit summary |
|||
Line 431:
Simulation FINISHED
</pre>
The next part of the ''Simulation Summary'' contains information on the messages issued by TUFLOW. Of note is the 2D Negative Depths (1). This indicates that the numerical solution has "overshot" and calculated a negative depth. Repeated negative depths are an indication that the model is not performing well. We will look at where these occur in the next [[
<pre>
Total 1D Negative Depths: 0
Line 489:
:The '''_ TUFLOW Simulations.log''' also contains a variety of other information, such as date, computer name, TUFLOW build version, the CPU hours and a other useful data.
*Time-series mass balance output. This is written to the same location as the 2D model results (in this case '''TUFLOW\results\M01\2d\'''). This file has the same name as the runfile (TCF) with the suffix, '''_MB.csv'''. As such the mass balance spreadsheet for this simulation is called '''M01_5m_002_MB.csv'''. Open the file in Excel. It contains a summary of the mass entering and leaving the model. Plot the first (time) and last (cumulative mass error %) columns within the spreadsheet to review the time varying mass error values.
*MB1 and MB2 map output. These 2D map outputs can be viewed using the methods described above in <u>[[
The MB1 output tracks the convergence of the solution at each cell since the previous output time. The MB2 is the cumulative sum of the MB1 output. In the image below the MB2 output is shown at the final timestep (3 hours). It highlights areas in which the 2D solution has not performed well during the simulation.<br>
It can be seen that these are the areas along the main channel. The creek is approximately 5-10m wide and the flowpath is being represented by 1-2 grid cells. This is likely to be causing the issues.<br>
Line 504:
See the <u>[[Tutorial_Module01_Archive#Advanced_.28Optional.29 |Optional Section]]</u> below, in which we look at halving the cell size and the effect this has on the model results and simulation time.
Whilst reviewing the results, you may have noticed that the water was being held behind these road embankments (which were essentially acting as dams in the model. We will embed 1D culverts through the road embankments in <u>[[
==Troubleshooting==
Line 533:
<br>
===Results===
Using the methods described above <u>[[
[[File:Tute M01 2p5m results SMS.png|450px]]
| |||