Community header

Forum

  1. Home
  2. Elv
  3. ELV downstream normal flow condition

ELV downstream normal flow condition

Hermjan Barneveld, modified 4 Years ago.

ELV downstream normal flow condition

Rookie Crystal gazer Posts: 10 Join Date: 6/24/20 Recent Posts
I try to impose normal flow donwnstream boundary condition in a simulation with variable discharge. I select nput.bch.dotype=3. But I do need to include the 2 lines for etaw0 otherwise get the error message "Reference to non-existent field 'timeetaw0'. in  boundary_conditions_construction_interp, at line 255 "
Including the 2 last lines does work, but I have constant water level at the downstream boundary then. Any suggestion?
input.bch.uptype=12; %type of hydrodynamic boundary condition at the upstream end: 1=water discharge; 2=cyclic hydrograph; 12=input from file [-] [1x1 double]; e.g. [1]
input.bch.path_file_Q=fullfile(source_path,'data','Q_Meuse_50yr.mat'); %path to the mat file containing the time series
input.bch.dotype=3; %type of hydrodynamic boundary condition at the downstream end: 
input.bch.timeetaw0=[0;0]; %time at which the downstream water level is specified ; [nix1 double]; e.g. [1800;3600] or [0,input.mdv.Tstop]
input.bch.etaw0=[0;0]; %downstream water level at the specified times [m]; [nix1 double]; e.g. [1;1.5]
Victor Chavarrias, modified 4 Years ago.

RE: ELV downstream normal flow condition

Famous Fortune teller Posts: 97 Join Date: 4/24/20 Recent Posts
Hi Hermjan, 

I think that you are using the initial condition type 4 (i.e., normal flow for given conditions). In this case, I wanted to automatically compute the water level downstream that would keep the equilibrium situation. Hence, you cannot impose anything at the downstream end. the downstream water level is computed as a constant value that keeps the normal flow. It is what I needed when I implemented it, sorry emoticon.

There are several solutions. The simplest is that you impose the initial condition rather than letting Elv compute it (i.e., input.ini.initype=2). Then, you can impose whatever boundary condition you want. Check the logfile of the simulation you have run and you will see that the normal flow initial condition is printed. Use it as input.
You are modelling a flood wave. The actual correct boundary condition to prevent information to reflect at the downstream end is not normal flow depth but Riemman boundary. This is unfortunately not implemented. Hence, the best you can do independently of the boundary condition you impose is to place the downstream part of the domain far from your region of interest.