Welcome, Guest
Username: Password: Remember me
Forgot your password? Forgot your username? Create an account
Forum
System Modules
GAIA / SISYPHE
Modeling flushing out of a reservoir using BREACH
Flat Threaded
  • Page:
  • 1
  • 2

TOPIC: Modeling flushing out of a reservoir using BREACH

Modeling flushing out of a reservoir using BREACH 11 years 10 months ago #6915

  • pilou1253
  • pilou1253's Avatar
  • OFFLINE
  • openTELEMAC Guru
  • Posts: 584
  • Thank you received: 106
Hi all!

I will model flushing of materials out of a reservoir and I thouht of using the new BREACH function to do this.
Before starting, I would like to know if BREACH can be used coupled with Sisyphe. And especially in the configuration where the dam defined in BREACH will have a non erodible depth equal to 0 (the only erodible area being the reservoir itself).
I don't know if the fact of having ZR variable in time can be solved without problem by Sisyphe.

I have a last question regarding noerod.f. I have been reading some posts here and found that the subroutine can be adapted in order to read the non erodible information directly in the geometry file. Here is an example I found from Dougal's modified file:
! Read rigid bed level from geometry file (variable 'NOER')

      CALL BIEF_ALLVEC(1,NOER,'NOER  ',IELMT,1,2,MESH)
      ALLOCATE(W(NPOIN),STAT=ERR)
      NAME='NOER            '
      CALL FIND_IN_SEL(NOER,NAME,T2D_FILES(T2DGEO)%LU,
     & W,OK)
      CALL OV( 'X=Y+C     ',ZR,NOER%R,ZF,0.D0,NPOIN)

I tried to figure out what are exactely all the parameters of the OV subroutine (I checked in the guide for programming), but it still remains a bit unclear why there are 4 parameters (ZR,NOER%R,ZF,0.D0) with the X=Y+C operand.
Any help on this would be much appreciated, especially on what should be defined in 'NOER'? The depth or the level (altitude) of the rigid bed ?

Thank you very much in advance for your help!
PL
The administrator has disabled public write access.

Modeling flushing out of a reservoir using BREACH 11 years 10 months ago #6916

  • jmhervouet
  • jmhervouet's Avatar
Hello,

On variable ZR: I would say that, at least in the finite element option, the real variable is the sediment height that is above ZR, so if you change ZR it should work, and the new bottom will be the new ZR + the sediment height.

On OV : some arguments in OV are not optional (in OS they are) so they must be present, even if not useful. In the line :

CALL OV( 'X=Y+C ',ZR,NOER%R,ZF,0.D0,NPOIN)

the second ZF is here only as a dummy argument, any double precision array would work. Actually here C in the formula X=Y+C is 0.D0, so the call is equivalent to simply :

CALL OV( 'X=Y ',ZR,NOER%R,ZF,0.D0,NPOIN)

Maybe Christophe Coulet will answer on the BREACH feature, to me it is only a way of moving the bottom, so it should work in a coupling with Sisyphe.

With best regards,

Jean-Michel Hervouet
The administrator has disabled public write access.

Modeling flushing out of a reservoir using BREACH 11 years 10 months ago #6917

  • pilou1253
  • pilou1253's Avatar
  • OFFLINE
  • openTELEMAC Guru
  • Posts: 584
  • Thank you received: 106
Thank you for your reply!

So, if I want to define the erodible bed thickness in NOER, I should have the following OV formulation:

CALL OV( 'X=Y-Z ',ZR,ZF,NOER%R,0.D0,NPOIN)

Right?

It would then be great if C. Coulet could confirm that Sisyphe works with BREACH.

Regards,
PL
The administrator has disabled public write access.

Modeling flushing out of a reservoir using BREACH 11 years 10 months ago #6918

  • jmhervouet
  • jmhervouet's Avatar
Yes, that's it, except that the name NOER for the erodible part is somewhat misleading...

JMH
The administrator has disabled public write access.

Modeling flushing out of a reservoir using BREACH 11 years 10 months ago #6942

  • c.coulet
  • c.coulet's Avatar
  • OFFLINE
  • Moderator
  • Posts: 3722
  • Thank you received: 1031
Hi

The BREACH module hasn't been tested with the Telemac-Sisyphe coupling.
Nevertheless, the aim of this module is to modify the bottom level in a specific location in order to represent a breach.
If this area is non-erodible in sisyphe, I thought this should work.
To check, it could be good to mede a test with ZF and ZR in output. The only risk i could imagine is if ZR=ZF at the beginning and ZF is modified by the breach algorithm. I'm not sure there is a correction of ZR in sisyphe and maybe this could create a problem ...

Regards
Christophe
The administrator has disabled public write access.

Modeling flushing out of a reservoir using BREACH 11 years 10 months ago #6950

  • pilou1253
  • pilou1253's Avatar
  • OFFLINE
  • openTELEMAC Guru
  • Posts: 584
  • Thank you received: 106
Hi

I have tested it. I used the geometry file provided with the bridge validation case and modified the geometry.

The computation works fine (Breach + Sisyphe) ONLY if the bed is set as erodible at the breach location, with a sufficient depth so that instabilities can be avoided (it might be function of time step and breach development rate I guess). I can send the files if you want to have a closer look.

Regards,
PL
The administrator has disabled public write access.

Modeling flushing out of a reservoir using BREACH 11 years 10 months ago #6983

  • pilou1253
  • pilou1253's Avatar
  • OFFLINE
  • openTELEMAC Guru
  • Posts: 584
  • Thank you received: 106
Hi all!

I am now making some tests on a flushing simulation.
I have ran a case with bed-load only and I now try to combine it with suspension. I would like to have a few advices on the following.

The documentation mentions that the use of total transport formulae (EH...) should only be used with steady conditions. What about coupling bed-load (with non total formulae) and suspended load? Are the available equilibrium formulae compatible with unsteady flow? What are the 2 new formulae, nr 3 and 4, that are not described in the manual (6.0) ?

Does someone have some experience to share regarding flushing simulation, with fine sand (d50 = 0,1 mm) ?

In the listing, ERROR ON VOLUME (dimensionless) is in the same magnitude as VOLUME OF DEPOSIT (m3). Is that result normal? Which indication does it give on the accuracy?

Thank you in advance for your help!
PL
The administrator has disabled public write access.

Modeling flushing out of a reservoir using BREACH 11 years 10 months ago #7027

  • pilou1253
  • pilou1253's Avatar
  • OFFLINE
  • openTELEMAC Guru
  • Posts: 584
  • Thank you received: 106
Hi again,

I read some documentation on Reference concentration formulae. I also read the article "Modelling sediment transport with hysteresis effect" in the 2011 user club proceedings by two HR Wallingford engineers, that is very interesting, especially when dealing with unsteady flows.

I know that a new version of the Sisyphe manual is to be released soon, but I would like to have some more information on the following (in addition to my previous post).

Will the method described in that article be included in the code in a next release (modified suspended concentration with lag effect) ?

Should all the reference concentration formulae be used with steady conditions only? If yes, do some people have any recommendation for unsteady flows (only bed-load?, ...)?

Finally, I have a few more "theoretical" questions:

The Meyer Peter formula coefficient used by default is 8. I found several versions of this formula, and especially one in my course material (Transports Solides, P. Lefort) where the coefficient seems to be 25. I also read in one article of the user club proceedings that the coefficient has once been set to 3. Is there any documentation on how can vary this coefficient, under which assumptions ?

According to the manual, the Shields parameter computed with the Van Rijn formula is 0,045 for D* >= 150. I checked in the original article (ref 31 of the manual), and it says 0,055. Is there any reason for that/error in the original article?

Thank you very much in advance for your feedback!
Cheers
PL
The administrator has disabled public write access.

Modeling flushing out of a reservoir using BREACH 11 years 10 months ago #7030

  • mafknaapen
  • mafknaapen's Avatar
  • OFFLINE
  • Expert Boarder
  • Posts: 157
  • Thank you received: 62
Hi PL,

I can confirm that the sediment trasnport option proposed in the paper "Modelling sediment transport with hysteresis effect" will be included in the next release.

I answer to you question
Should all the reference concentration formulae be used with steady conditions only? If yes, do some people have any recommendation for unsteady flows (only bed-load?, ...)?
:

All reference concentration formulae in common literature assume steady conditions. It is very diffficult to analyse non-steady condition sediment concentrations, as they are inherently transient.

However, The reference concentration formulae still provide decent results in a lot of situations and are used widely. So I would suggest use them but with care. (Bed load only models should not be effected to the same effect as suspende load can be).


Regards,

Michiel Knaapen
Dr Michiel Knaapen
Senior Scientist
E This email address is being protected from spambots. You need JavaScript enabled to view it.
T +44 (0)1491 822399

HR Wallingford, Howbery Park, Wallingford, Oxfordshire OX10 8BA, United Kingdom
T +44 (0)1491 835381, F +44 (0)1491 832233
www.hrwallingford.com
The administrator has disabled public write access.
The following user(s) said Thank You: pilou1253

Modeling flushing out of a reservoir using BREACH 11 years 10 months ago #7031

  • Pablo
  • Pablo's Avatar
  • OFFLINE
  • Moderator
  • Posts: 316
  • Thank you received: 121
Dear PL,
I'd like to complete Michiel's remarks.

> The Meyer Peter formula coefficient used by default is 8. I found several >versions of this formula, and especially one in my course material (Transports
...

In the end, some authors use the MPM coefficient as a calibration parameter. Do not forget you can use the keyword "MPM COEFFICIENT" to modify this coefficient via the steering file (=8 by default). Under this aspect, please take also a look to the article:
Wong, M. and Parker, G. (2006). ”Reanalysis and Correction of Bed-Load Relation of Meyer-Peter and Müller Using Their Own Database.” J. Hydraul. Eng., 132(11), 1159–1168.
doi: 10.1061/(ASCE)0733-9429(2006)132:11(1159)

>According to the manual, the Shields parameter computed with the Van Rijn >formula is 0,045 for D* >= 150. I checked in the original article (ref 31 >of the manual), and it says 0,055

Some authors recommend to take this value =0,045 (see also subroutine init_sediment.f)

All the best,

Pablo
The administrator has disabled public write access.
The following user(s) said Thank You: pilou1253
Flat Threaded
  • Page:
  • 1
  • 2
Moderators: Pablo, pavans
Forum
System Modules
GAIA / SISYPHE
Modeling flushing out of a reservoir using BREACH

The open TELEMAC-MASCARET template for Joomla!2.5, the HTML 4 version.