open TELEMAC-MASCARET The mathematically superior suite of solvers

Hi,
I want to do the sediment transportation simulation with multi-grain class. In the steering file, I set the sediment properties as following:

/ SEDIMENT PROPERTIES
NUMBER OF SIZE-CLASSES OF BED MATERIAL = 2
INITIAL FRACTION FOR PARTICULAR SIZE CLASS =0.7;0.3
MEAN DIAMETER OF THE SEDIMENT =0.002;0.0039
/SEDIMENT DIAMETERS = 0.0037
SEDIMENT DENSITY = 2650

When I run the telemac, there appear an error message as following:

"error in layer case 5"

The whole information is in attachment file.

What is the reason of the error?

Thanks a lot.
Jinyong ZHAO

Hello,

This would be rather a question for the Sisyphe module, but the error 5 occurs when, due to truncation errors, the sum of all fraction of sediment goes over 1. If it is really a truncation error and not another error, you can try to take subroutine layer.f in your FORTRAN FILE and change the parameter ZERO given in DATA.

With best regards,

Jean-Michel Hervouet

Hi,Jean Michel Hervouet,

Thank you for your reply.
In your message, you say"you can try to take subroutine layer.f in your FORTRAN FILE and change the parameter ZERO given in DATA"
How to do that?

Thanks.

Jinyong Zhao

Hello,

FORTRAN FILE is a keyword, if you put in your STEERING FILE the keyword :

FORTRAN FILE : princi.f

you can then put in the file princi.f subroutines taken from the telemac libraries and change them for your needs, e.g. change the parameter ZERO in layer.f. The compiler will take your modified subroutine and will discard the one in the library.

Subroutine layer.f is in library .../V6P2/sisyphe/sisyphe_v6p2/sources/layer.f

With best regards,

Jean-Michel Hervouet

hi, Jean-Michel Hervouet,
I try what you suggest. But still doesnot work, and compilation error occurs that donot occur before.
I summarize the whole process:
When I do the sediment transport simulation with one single grain size with TELEMAC2D version6.1, everything is well. But when I modify the part of "sediment property" of the steering files and try to do the sediment transport simulation with multi grain size, "error in layer case 5" occurs.
The following are the contents of the steering files that produce the "error in layer case 5":
/

---

/ TELEMAC-2D - cnd
/

---

/
TITLE = 'cnd'
/
/

---

/ PROJECT FILES
/

---

/
GEOMETRY FILE = geo_cnd.slf
BOUNDARY CONDITIONS FILE = geo_cnd.cli
PARALLEL PROCESSORS = 0
/FORTRAN FILE = princi.f
/
SISYPHE STEERING FILE = multi_s.cas
COUPLING WITH = INTER-SISYPHE
/
/ Run#1
RESULTS FILE = result_multi_f.slf
COMPUTATION CONTINUED = NO
/
/ Run#2
/ RESULTS FILE = r2d_cnd_2.slf
/ COMPUTATION CONTINUED = YES
/ PREVIOUS COMPUTATION FILE = r2d_cnd_1.slf
/
/

---

/ GENERAL INFORMATIONS - OUTPUTS
/

---

/
VARIABLES FOR GRAPHIC PRINTOUTS = 'U,V,H,S,B'
GRAPHIC PRINTOUT PERIOD = 5000
LISTING PRINTOUT PERIOD = 1000
VALIDATION = NO
MASS-BALANCE = YES
TIME STEP = 0.1
INFORMATION ABOUT SOLVER = YES
/
/ Run#1
NUMBER OF TIME STEPS = 2500000
/
/

---

/ INITIAL CONDITIONS
/

---

/
INITIAL CONDITIONS = 'CONSTANT DEPTH'
INITIAL DEPTH = 0.1822
/
/

---

/ BOUNDARY CONDITIONS
/

---

/
PRESCRIBED FLOWRATES = 0.105;0.035;0.070
PRESCRIBED ELEVATIONS = 0.2852;0.;0.
VELOCITY PROFILES = 2;2;2
/
/

---

/ PHYSICAL PARAMETERS
/

---

/ LAW OF BOTTOM FRICTION : 3 = Strickler formula
/ FRICTION COEFFICIENT : Value depends on LAW OF BOTTOM FRICTION
/ TURBULENCE MODEL : 1 = Constant viscosity
/

---

/
LAW OF BOTTOM FRICTION = 3
FRICTION COEFFICIENT = 62.
TURBULENCE MODEL = 1
VELOCITY DIFFUSIVITY = 1.E-3
ROUGHNESS COEFFICIENT OF BOUNDARIES = 100.
/
/

---

/ NUMERICAL PARAMETERS
/

---

/
TYPE OF ADVECTION = 1;5
SUPG OPTION = 2;2
TREATMENT OF THE LINEAR SYSTEM = 2
SOLVER = 1
SOLVER OPTION = 3
IMPLICITATION FOR DEPTH = 0.6
IMPLICITATION FOR VELOCITY = 0.6
MASS-LUMPING ON H = 1.
H CLIPPING = NO
/
&FIN


/

---

/ SISYPHE - cnd
/

---

/
TITLE = cnd
/
/

---

/ PROJECT FILES
/

---

/
GEOMETRY FILE = geo_cnd.slf
BOUNDARY CONDITIONS FILE = geo_cnd.cli
/FORTRAN FILE = layer.f
/
/ Run#1
RESULTS FILE = result_multi_s.slf
COMPUTATION CONTINUED = NO
/
/ Run#2
/ RESULTS FILE = ris_cnd_2.slf
/ COMPUTATION CONTINUED = YES
/ PREVIOUS SEDIMENTOLOGICAL COMPUTATION FILE = ris_cnd_1.slf
/
/

---

/ GENERAL INFORMATIONS - OUTPUTS
/

---

/
VARIABLES FOR GRAPHIC PRINTOUTS = U,V,H,S,B,E,CS1,TOB,M,QSBL
VALIDATION = NO
MASS-BALANCE = YES
/
/

---

/ NUMERICAL
/

---

/
FINITE VOLUMES = YES
/
/

---

/ SEDIMENT PROPERTIES
/

---

/
NUMBER OF SIZE-CLASSES OF BED MATERIAL = 2
INITIAL FRACTION FOR PARTICULAR SIZE CLASS =.7;.3
MEAN DIAMETER OF THE SEDIMENT =.002;.0039
/SEDIMENT DIAMETERS =0.002;0.0039
SEDIMENT DENSITY = 2650
/
/

---

/ SEDIMENT TRANSPORT
/ BED-LOAD TRANSPORT FORMULA: 1 = MPM; 7 = van Rijn
/

---

/
BED LOAD = YES
BED-LOAD TRANSPORT FORMULA = 1
SUSPENSION = NO
NON COHESIVE BED POROSITY = 0.4
/
/

---

/ SLOPE EFFECT
/

---

/
SLOPE EFFECT = YES
FORMULA FOR SLOPE EFFECT = 1
FORMULA FOR DEVIATION = 1
FRICTION ANGLE OF THE SEDIMENT = 40
BETA = 1.3
PARAMETER FOR DEVIATION = 0.85
/
/

---

/ SEDIMENT SLIDE
/

---

/
SEDIMENT SLIDE = NO
/
/

---

/ SECONDARY CURRENTS
/

---

/
SECONDARY CURRENTS = YES
/
/

---

/ TREATMENT OF RIGID BEDS
/

---

/
/ 0 = ERODABLE BOTTOMS EVERYWHERE
/ 1 = MINIMISATION OF THE SOLID DISCHARGE
/ 2 = NUL SOLID DISCHARGE
/ 3 = MINIMISATION OF THE SOLID DISCHARGE IN FE / MASS-LUMPING
/ 4 = MINIMISATION OF THE SOLID DISCHARGE IN FINITE VOLUMES
OPTION FOR THE TREATMENT OF NON ERODABLE BEDS = 0
MASS-LUMPING = YES
/
&FIN

Thank you very much!

Jinyong Zhao

Hello,

You can try in Telemac-2D steering file :

TYPE OF ADVECTION = 1;5
/ next line modified compared to what you sent
SUPG OPTION = 2;0
TREATMENT OF THE LINEAR SYSTEM = 2
SOLVER = 1
SOLVER OPTION = 3
IMPLICITATION FOR DEPTH = 0.6
IMPLICITATION FOR VELOCITY = 0.6
MASS-LUMPING ON H = 1.
H CLIPPING = NO
/ this will replace SUPG OPTION for supressing wiggles
COMPATIBILITY OF FREE SURFACE GRADIENT : 0.9
/ even if you have no tidal flats, this has the secondary effect of giving
/ a continuity equation at machine accuracy
TREATMENT OF NEGATIVE DEPTHS : 2
/ this will correct the problem of continuity at exits where depth is imposed
CONTINUITY CORRECTION : YES


then if the problem remains it will be for my colleagues working on Sisyphe, though testing version 6.2 would be also something to try.

With best regards,

Jean-Michel Hervouet

hi, Jean-Michel Hervouet,
I try all the keywords you provide.
Firstly, one error occur, the information as following:
Le mot <COMPATIBILITY OF FREE SURFACE GRADIENT> n'existe pas dans le dictionnaire(The word <COMPATIBILITY OF FREE GRADIENTFREE SURFACE > does not exist in the dictionary)
So, I change the keyword <COMPATIBILITY OF FREE SURFACE GRADIENT> into comments.
After that, another new error occur. I save the information in the attachment file.
Best regards.
Jinyong Zhao

Hello,

Sorry, I did it without checking, it's in fact :

FREE SURFACE GRADIENT COMPATIBILITY : 0.9 / to even 0. to stabilize

the other message is certainly a boundary with prescribed discharge that becomes totally dry, but you did not have this before, so it is certainly crashing, due to the absence of any filtering of wiggles.

Regards,

JMH

Hi，JMH
I modify the steering file based on the information you provide.
The temelec2d runs fluently. And the final screen information is in the attachment file. Whether this means that the Telemac2d finish the simulation very well?
Best regards.
Jinyong Zhao

Hello,

Yes, this is quite OK.

JMH