New features of Telemac-2D version 7.2
New features and differences with version 7.1
Estimated release date: end 2016
Latest article update: 14/03/2016
Main new features of this version 7.2
* New distributive scheme, option 15, called ERIA (Element by element Residual distributive Iterative Advection scheme). It is like scheme NERD (13 or 14) but second order in space and with upwinding of the derivative in time. Comparable to LIPS scheme for the results, but without linear system to solve. For more details on all keywords related to advection schemes see post #19861 in the forum (Generalities/Announcements...).
* New option for the treatment of negative depths.
* New turbulence model (mixing-length).
* All advection schemes now enabled for the advection of k and epsilon.
* New formulation added for culverts in TELEMAC-2D and 3D, proposed and tested at Flanders Hydraulics.
* Algorithmic (or automatic) differentiation using the Nag Compiler.
* For Finite volumes, the management of the graphical output is common with finite elements i.e. with a TIME STEP = 1s and GRAPHIC PRINTOUT PERIOD = 100, Telemac2d will generate a graphic output in the result file every 100s (instead of 100 iterations (with variable time step) before).
* For Finite volumes, new hydrostatic reconstruction for the well-balanceness of source terms.
* New rainfall-runoff model based on the Curve Number (CN) method implemented by P.L. Ligier from Sweco (Sweden). See validation case Pluie (examples/telemac2d) and user manual for more détails.
* New algorithm for the computation of cross-section.
New or modified key-words
"TREATMENT OF NEGATIVE DEPTHS" (TRAITEMENT DES HAUTEURS NEGATIVES)
Option 3 is new, it is like option 2 but the fluxes are limited element by element. The result is similar but this option is necessary if the new advection scheme 15 (ERIA) is used for the tracers.
"TURBULENCE MODEL" (MODELE DE TURBULENCE)
Option 5 is a new mixing-length model, programmed and provided by Clemens Dorfmann. It combines the classical mixing-length in 2D and a parabolic model for the vertical. Some constants of the model may be tuned, see subroutine mixlength.f.
"TURBULENCE REGIME FOR SOLID BOUNDARIES"
It replaces the misnamed "TURBULENCE MODEL FOR SOLID BOUNDARIES".
"OPTION FOR CULVERTS" ("OPTION POUR LES BUSES")
Option for the treatment of culverts. There are two options in TELEMAC.
"NUMBER OF SIPHONS"
It replaces the misnamed "NUMBER OF CULVERTS"
"NUMBER OF CULVERTS"
It replaces the misnamed 'NUMBER OF TUBES'
"SIPHONS DATA FILE"
It replaces the misnamed "CULVERT DATA FILE"
"CULVERTS DATA FILE"
It replaces the misnamed "TUBES DATA FILE"
"NUMBER OF DIFFERENTIATORS"
Defines the number of user differentiators
"NAMES OF DIFFERENTIATORS"
Name of user differentiators in 32 characters, 16 for the name, 16 for the unit
"OPTION OF THE HYDROSTATIC RECONSTRUCTION"
Gives the option for hydrostatic reconstruction (used only for finite volumes):
1: option of Audusse
2: option of Noelle
Option for the rainfall-runoff model. Available options are:
0 : No infiltration
1 : CN runoff model (Curve Number method of the SCS)
"ANTECEDENT MOISTURE CONDITIONS"
Gives the antecedent moisture conditions before a rainfall
event for the SCS CN runoff model. Available options are:
1 : dry antecedent conditions
2 : normal antecedent conditions
3 : wet antecedent conditions
This keyword is only usefull for runoff model 1 (SCS CN model)
"OPTION FOR INITIAL ABSTRACTION RATIO"
Gives the ratio for Initial Abstraction to Maximal Potential
Retention S for the SCS CN runoff model. Available options are:
1 : IA/S = 0.2 (standard method)
2 : IA/S = 0.05 (revised method, see Woodward, Hawkins et al. 2003. With
this option the CN values given in input are automatically conversed,
see user manual).
This keyword is only useful for runoff model 1 (SCS CN model)
"DURATION OF RAIN OR EVAPORATION IN HOURS"
Gives the duration of the rain in hour, default value is infinite
Use Fluxline to compute flux over lines
"FLUXLINE INPUT FILE"
Name of the Fluxline file, with data on cross-sections
Modifications in user subroutines or new subroutines
Calls to cvtrvf.f and cvtrvf_pos have been added to enable all advection schemes for the advection of k and epsilon.