Tutorial Module04 Archive: Difference between revisions

Content deleted Content added
No edit summary
 
(10 intermediate revisions by 4 users not shown)
Line 1:
<ol>
=Introduction=
As discussed in Module 1 (<u>[[Tutorial_Module01_Archive#Reviewing_Model_Performance | (discussion here)]]</u>), the main creek channel is not very well represented using the 5m 2D cell size. In parts, the creek is only 5-10m wide and the 5m cell size could be considered too coarse to accurately represent the creek topography. Refer to the below figure.<br>
[[file:Poor_2d_rep.png|400px]]<br>
Using a cell size that is coarse relative to the width of the creek channel may reduce the accuracy of the computed conveyance in the channel and introduce a source of mass error to the model. There are two options for improving the representation of the creek channel:<br>
* decrease the width of the 2D cells; and/or
* model the channel as a 1D network, dynamically linked to the 2D domain (the floodplain).
In the optional section of module 1 (<u>[[Tutorial_Module01_Archive#Advanced_-_Model_Resolution_.28Optional.29 | (located here)]]</u>), we looked at reducing the cell size to get a better representation of the channel. In this module we will adopt the second approach of modelling the creek as 1D elements.<br>
Setting up a 1D/2D model where the 1D channel cuts through the 2D domain is probably the most time-consuming type of a TUFLOW model to setup. However, the reduction in simulation time can be beneficial and make this a good approach. For this module, most of the time consuming data entry tasks have been undertaken, so you can progress through the module in a relatively short period of time. However, throughout the tutorial we will provide guidance on making this process as efficient as possible should you have to start from scratch.<br>
 
Line 29:
As with Modules 2 and 3 the GIS inputs are demonstrated in ArcGIS, MapInfo, SMS and QGIS. At each stage please select your GIS package for instructions.
==1D Cross-Sections==
A 1D model requires cross-sections for processing of the channel geometry. There are a number of different formats for cross-sections to be input to TUFLOW. In this module we will be using an offset-elevation approach (XZ). This is a typical format for inputting cross-sections however, other options include height-width (HW) and MIKE11 format inputs. These are not covered in this tutorial, however the general process is very similar. To save time, the locations and XZ profiles for the 1D cross sections are provided. The XZ profiles were automatically extracted from the DEM triangulation using the 12D software (www.12d.com) with the aid of the TUFLOW utility 12da_to_from_mif12da_to_from_GIS.exe using the -xs option (see [[12da_to_from_mif12da_to_from_GIS | 12da_to_from_mif12da_to_from_GIS wiki page]]). The SMS TUFLOW interface has a similar and easier process, noting that a licence for the SMS TUFLOW interface is required to save the cross-sections. It is more common for this information to come from surveyed cross-sections. If the survey has been provided as x-y-z points, or a a GIS layer of points, the xs_Generator utility can be used to create the cross-sections in a TUFLOW format. The xs_Generator utility is described in the [[XsGenerator |XsGenerator wiki page]].<br>
<br>
For each cross-section there is an individual comma separated variable (.csv) file. The 1d_xs layer contains a link to the source .csv file at every cross-section location. <br>
Line 50:
 
==1D Channels==
Now that we have looked at the cross section data, we must add 1D channels for the 1D network. In TUFLOW, cross sections can either be located at the ends of each channel reach, or at some point along the reach. The way that TUFLOW uses the cross-section depends upon the location, for example a "centre" cross-section has a higher priority than "end" sections if both "End" and "Centre" sections are defined. For more information on the 1D topography, youplease are encouragedrefer to readthe Section 4<u>[https://docs.6 of the 2010tuflow.com/classic-hpc/manual/latest/ TUFLOW manualManual]</u>.<br>
<br>
For this tutorial, we will locate most of the cross-sections at the ends of channels. There are a number of locations where a "centre" section will be used. The final channel will be a weir type channel; this represents a weir with a known stage-discharge curve. In later modules we will also use "centre" cross sections for the modelling of structures.<br>
Line 63:
 
==1D/2D Links==
In this section we define lines where there is flow exchange between the 1D and 2D components of the model. At first it may seem slightly complicated, but it is actually very logical and the method allows the user to change the 2D cell size and orientation at a later stage without having to change the 1D/2D linking. Before we begin creating the link we will spend a bit of time describing how the link works. More details on the 1D/2D link can be found in Sectionthe 4<u>[https://docs.10tuflow.5 of the 2010com/classic-hpc/manual/latest/ TUFLOW manualManual]</u>.<br>
<br>
'''Description of linking concept'''<br>
Line 90:
 
==2D Breaklines==
As mentioned above, the water level from the 1D is transferred out to the 2D HX cells. Therefore, it is important that the 2D cell elevation reflects the level when water can spill out into the 2D. The water level computation point for the 2D cells is the cell centre, therefore it is important to use a "thick" breakline. The types of breaklines was discussed further when reviewing the check files for the previous module <u>[[Tutorial_Module03_Archive_ArchiveTutorial_Module03_Archive#Review_Check_Files | here]]</u>.<br>
In the image below, the lightly shaded cells are the 1D/2D boundary cells. The black labels are the elevations with no breakline (elevation as read from the DEM). At the circled cell, the cell centre elevation is 44.35, this controls when water can spill from the 1D to the 2D. The red line indicates the true top of bank; the labels are elevations along this line. If this is included as a breakline, the elevations are as labelled in yellow. In this case, the level at the circled cell is actually 44.57. Including breaklines for the top of bank is important, particularly if there is an embankment or levee.
<br>
Line 149:
<li>After this command in the .tcf, add the following command to link the new 1D control file:</li>
<font color="blue"><tt>ESTRY Control File </tt></font> <font color="red"><tt>== </tt></font> <tt>..\model\M04_1d_001.ecf</tt><br>
<li>Depending on the speed of your computer, you may have noticed in the previous three modules that when TUFLOW is running, it is hard to read the TUFLOW console output because it updates too quickly. We can reduce the frequency of output to the DOSconsole screenwindow and also the TUFLOW Log File (.tlf). Include the following command at the end of the .tcf with the other output commands:<br>
<font color="blue"><tt>Screen/Log Display Interval </tt></font> <font color="red"><tt>== </tt></font> <tt>40</tt><br>
TUFLOW will write the output every 40 timesteps. For the 1.5 second timestep this means we will get output every 60 seconds of simulation time.<br>
Line 235:
 
=Run the Simulation=
Using your preferred method for starting TUFLOW, run the recently created '''M04_5m_001.tcf'''. Please refer to <u>[[Tutorial_Module01_Archive#Running_TUFLOWRun_Simulation | moduleModule 1]]</u> for a detailed description of the various methods for running a TUFLOW simulation.<br>
If the model fails to start correctly please refer to the <u>[[Tutorial_Module04_Archive#Troubleshooting |troubleshooting section]]</u> at the end of this page.<br>
 
Line 258:
[[File:Tute M04 results no 1d 02.jpg|500px]]
<br><br>
If you look at the 1D results (by importing the time series (_TS.mif or _TS_P.shp from the 1D results folder, or opening the M04_5m_001_1d_Q.csv) the 1D channels are conveying flow. In order to get map output for the 1D we need to use Water Level Line (WLL) inputs. The underlying water level from the 1D model is displayed across this line and as such it should be digitised perpendicular to flow. The theory behind water level lines and their use is detailed in the Water Level Lines section of the<u>[https://docs.tuflow.com/classic-hpc/manual/latest/ TUFLOW Manual]</u>. <br>
 
<li> For the purpose of this tutorial the 1D_WLL GIS inputs have been provided for you.<br>
Line 281:
If the TUFLOW simulation fails to start TUFLOW will output the error in a number of locations. Firstly, check the Console Window or the TUFLOW Log File (.tlf) (located in the '''TUFLOW\runs\log\''' folder with the name '''M04_5m_001.tlf'''). This file can be opened in a text editor and the error is generally located at the end of the file. You can however search for "Error" if you can not see the error. In most cases there is also a spatial location for the error message (if the error reported in the log file is prefixed by XY:). To check the location of the geographic errors, open the '''M04_5m_001_messages.mif''' or '''M04_5m_001_messages_P.shp''' file in your GIS package.<br>
<br>
This section contains links to some possible problems that may occur when progressing through the fourth tutorial module. If you experience an issue that is not detailed, please email [mailto:support@tuflow.com or add and describe the problem on the <u>[[Talk%3ATutorial_Module04_Archive | discussion pagesupport@tuflow.com]]</u>.<br>
 
*[[Save_Date_Error_Archive|Save date of .tab file is later than .mif or .mid]]
*[[Model_Changes_Not_Seen_Archive | When reviewing check files, I can't see the changes]]
*[[Tute_Error_2051_Archive | Error 2051 - Connection object unused]]
*[[TUFLOW_Message_2060_ArchiveTUFLOW_Message_2060 | Error 2060 - Could not find a CN connection snapped to the end of HX line]]
*[[TUFLOW_Message_2024_ArchiveTUFLOW_Message_2024 | Error 2024 - Could not find a 1D node snapped to CN line.]]
*[[Only_TUFLOW_DOS_Window_ArchiveOnly_TUFLOW_Console_Window_Archive | Why does the TUFLOW DOSconsole Windowwindow open, though the TCF isn’t read and the simulation doesn’t execute?]]
 
= Advanced - HPC Solver (Optional) =
This section will introduce how to run the model TUFLOW’s HPC (Heavily Parallelised Compute) solver, and how to fix some common issues that may occur when trying to run a simulation using Graphics Processing Unit (GPU) hardware. Please see [[HPC_Features_Archive | HPC Features Archive]] for more information on TUFLOW HPC features supported in the 2017 release.
 
TUFLOW HPC can run between 10 and 100 times faster than TUFLOW Classic using NVidia Graphics Processing Units (GPU)(depending on the model configuration and hardware performance).
Line 309:
==Results==
Using the methods described above in the <u>[[Tutorial_Module01_Archive#Viewing_Results | Viewing Results]]</u> section of Module 1.
* Check the simulation logs in the DOSconsole window, .tlf and .hpc.tlf log files.
* View the results in your preferred package.
Do the logs and results appear different to the TUFLOW Classic simulation?