1D river simulations - times tab

In this tab, simulation type and timing details are specified.

Simulation type

The simulation type is selected using the radio buttons provided.

Steady runs

A steady run can be chosen for any time for which data is available in the boundary conditions. Steady runs can utilise the direct method or the timestepping method. For 1D simulations, the direct method merely requires a starting time for which the run is required, whilst timestepping additionally requires a time step to be entered (default value of 20s). The choice of time step is discussed further in the Convergence Information topic.

If previous steady runs have been executed on the same file and the steady results file (filename extension .zzs) still exists in the working directory then the user can choose whether to use this file for the run's initial conditions. This option can be toggled on and off by clicking the appropriate option in the Files Tab of the 1D Simulation Interface. Note that if the node labels have been changed between the previous run producing the steady results file and the current datafile then the initial conditions must be taken from the current datafile (i.e. turn off the 'Initial Conditions from Initial Conditions File' option).

The run is started by clicking the 'Run' button. The program now performs a check on the node labels. If no errors are detected then the message 'no errors found in node labels' is displayed in the Message Log, otherwise the program lists the errors detected.

When the direct method solution is found, the results are saved in the steady results file (filename extension '.zzs').
For the pseudo-timestepping method the convergence ratios are displayed after each iteration. Every five time steps the program displays the following prompts:

continue (<cr>), save (s) or quit (q)>

Thus, depending on the Convergence Information, the user can enter a null return to continue iterations, an 's' to save the current results in the steady results file (filename extension '.zzs'), or a 'q' to quit the program. If iterations are continued then the following prompt is displayed:

enter new time step, continue (<cr>)>

When the solution ceases to change significantly at the required time step the steady results may be saved and the program quit. If the user wishes to continue iterations with the same time step a null return can be entered; if a different time step is required this should be entered in units of seconds. The choice of time step is discussed further in the Convergence Information topic.

Following the successful completion of either type of steady run the program may display one or more messages in the Log window warning the user of possible problems encountered during the simulation, as well as a message dialog informing of the successful completion of the run. When this dialog is cleared (by pressing the 'View' button) the Flood Modeller File Inspection utility displays the steady results produced by the run.

Further details of the warnings referred to above, along with other information on the run are written to the end of the diagnostics file (filename extension '.zzd'), which can be viewed using the 1D Model File Inspection utility at this time. This file will also be displayed if the 1D Model run terminates because of an error at any time. This file will contain information that should enable the user to correctly diagnose the problem that caused Flood Modeller to fail.

Unsteady runs

Unsteady runs require a start and finish time to be entered to define the length of the run. The data given in the boundary conditions must encompass the simulation start and finish times.

A computational time step will also need to be entered if you do not wish to use the default value of 20s. For information on the selection of timestep please refer to the topics relating to Non-tidal and Tidal reaches.

Flood Modeller allows the user to save the results of the hydraulic calculations at selective time intervals. This is achieved by setting the value in the 'Save Interval ...' edit box. The value entered represents the time step in seconds at which results will be saved. This value should be a multiple of the time step used, if it is not then a dialog box will appear suggesting a suitable save interval. This can either be selected or an alternative value may be chosen.

It is important to be aware that saving results at all nodes and for every time step when using a many-noded model can result in very large binary results files of many megabytes in size. If, whilst creating these files, all of the space on the hard disk were to be used up, the Flood Modeller run will terminate prematurely.

If previous steady runs have been executed on the same file and the steady results file (filename extension .zzs) still exists in the working directory then the user can choose whether to use this file for the run's initial conditions. This option can be toggled on and off by clicking the appropriate check box in the Files Tab of the 1D Simulation Interface. Note that if the node labels have been changed between the previous run producing the steady results file and the current datafile then the initial conditions must be taken from the current datafile (ie turn off the 'Initial Conditions from Initial Conditions File' option).

The run is started by clicking the 'Run' button. The program now performs a check on the node labels. If no errors are detected then the message 'no errors found in node labels' is displayed in the Message Log, otherwise the program lists the errors detected.
At this point the program starts calculations for the hydraulic simulation. An additional graphical display will now appear showing the number of iterations performed for each time step, convergence information and total model inflows and outflows.

Between 2 and 6 iterations are used per time step by default; these values can be changed from within the 'Advanced Options' dialog if required. If the computations have not converged after the maximum permissible number of iterations the convergence ratios are displayed in the Message Log and computations move on to the next time step.

Following the successful completion of an unsteady run the program may display one or more messages warning the user of possible problems encountered during the simulation, as well as a message dialog informing of the successful completion of the run. When this dialog is cleared (by pressing the 'View' button) the 1D Model File Inspection utility displays the steady results file produced by the run.

Further details of the warnings referred to above, along with other information on the run are written to the end of the diagnostics file (filename extension '.zzd'), which can be viewed using the 1D Model File Inspection utility at this time. This file will also be displayed if the 1D model run terminates because of an error at any time. This file will contain information that should enable the user to correctly diagnose the problem that caused Flood Modeller to fail.

Boundary mode

This mode of operation is to enable users to extract and examine data from boundary type nodes (FEHBDY, FSRBDY, QTBDY, HTBDY etc) at regular time steps over a given time interval without the necessity for performing a full unsteady run. This mode also enables the user to examine rating curves for the CONQH, PUMP and QHBDY units and provide intermediate results for the Bridge units.

Boundary Mode runs require a start time, end time and a time step to be entered in Times Tab of the 1D Simulation Interface. The data given in the boundary conditions must encompass the simulation start and finish times.

On clicking the 'Run' button, Flood Modeller will extract boundary data for every time step within the specified range from all relevant nodes in the model and write them to the .zzn binary results file. Head-time data is extracted for HTBDY and TIDBDY units, while flow-time data is extracted for QTBDY, FSRBDY, FEHBDY, ReFHBDY, USSCSBDY and FRQSIM units. (in the last five cases intermediate results can be examined in the .zzb and .zzh files, which are written in ASCII format).

Boundary data can now be examined in the same way as for unsteady runs (i.e. graphically or by using the Tabular or Tabular CSV Processors).

Rating curve data for the units QHBDY, CONQH, and PUMP are written in ASCII format to the .zzu file and can be examined directly using the 1D Model File Inspection utility.

Timing details

A simulation Start Time and Finish Time should be provided. Absolute or relative times can be used. The Use Time Zero option also allows for a time zero to be provided.

If a 'Time Zero' is set, then this will equate to t=0 for the simulation; otherwise a real Start Time, if set, will be assumed to be at t=0. A real Finish Time may only be set if a Time Zero or real Start Time has been set.

The results postprocessors and time dependent units (e.g. QTBDY , Sluices, etc.) have not yet been enabled to accept real times, and for these purposes, the times will be stored internally as absolute times.

Time intervals

A Timestep and Save Interval should be provided. A minimum and maximum timestep should be provided if the unsteady (adaptive timestep ) option is selected.

If a warm-up start is selected from the Options tab, then the warm-up time must be specified (in hours). This equates to the number of warm-up timesteps to be performed multiplied by the specified timestep (in hours).

.mmm file details

A check-box is provided to adjust the mean/max/min calculation times from the default values of the start and finish of the simulation. If this option is chosen, the calculation start and end times should be provided in the fields below, entered in hours from time zero.