Density driven flow caused by heat source in porous media

Mar_Bitt · November 11, 2015, 7:56pm

Ladies and gentelemen,

I started working with OGS to compare it to other platforms solving heat transport coupled with richards flow.

In fluid properties I can define a line where the density is dependant of the temperature.

I did a bit research and found this page Inputfile .mfp density as fard as I understood 4. gives a relation between density and temperature. But I need to know how exactly OGS is calculating the density. The formula given at point 4. and the values lower on the page are rather confusing.
Density approximations are dependant of temperature ranges. And I can’t see something that is modelling something like that here.

A co-worker said he is using this equation:

$DENSITY
4 1.000000e+003 0 0.2 ; C0 is 0. beta_C (drho_dC) is 0.2

but this looks for me on the first sight more like a concentration related behaviour, also on the website it is described like that.

In addition viscosity should change too with temperature.

I’m new to OGS and I have to say I didnt find a lot about this stuff in the web.

Best regards,

MB

Christof_Beyer · November 11, 2015, 8:39pm

Hi Marius,

density model 4 is a linear density model, e.g.

$DENSITY
4 999.7 283.15 -2.5e-4

Parameters are rho(T_ref), T_ref, droh/dT. There are many other density models implemented, some of them are also taking into account concentration and/or pressure changes by such linearization.

You can also provide your own non-linear density model by ascii text input ("model 0"):

$DENSITY
0 1

Parameter "1" refers to the first data table in a *.rfd ascii input file, which you will have to provide in this case:

;Curve 1 Temp Density
#CURVES
273.15 999.8675792
274.15 999.9265054
...
#STOP

For T dependent viscosity, you can also provide a data file using the same syntax as for density, or you can use the empirical Yaws (model 3) or deMarsily (model 4) models

  $VISCOSITY
   3 ; Yaws
   ;4 ; DeMarsily

where the parameters are hardcoded in OGS. There are other model, which also take into account pressure and/or concentrations
Btw, you should define temperature in Kelvin.

Best, Christof

···

Am 11.11.2015 um 20:56 schrieb Mar. Bitt.:

Ladies and gentelemen,

I started working with OGS to compare it to other platforms solving heat
transport coupled with richards flow.

In fluid properties I can define a line where the density is dependant
of the temperature.

I did a bit research and found this page Inputfile .mfp density
<https://svn.ufz.de/ogs/wiki/public/doc-auto/by_ext/mfp/S_density> as
fard as I understood 4. gives a relation between density and
temperature. But I need to know how exactly OGS is calculating the
density. The formula given at point 4. and the values lower on the page
are rather confusing.
Density approximations are dependant of temperature ranges. And I can't
see something that is modelling something like that here.

A co-worker said he is using this equation:

$DENSITY
4 1.000000e+00300.2; C0 is 0. beta_C (drho_dC) is 0.2

but this looks for me on the first sight more like a concentration
related behaviour, also on the website it is described like that.

In addition viscosity should change too with temperature.

I'm new to OGS and I have to say I didnt find a lot about this stuff in
the web.

Best regards,

MB

--
You received this message because you are subscribed to the Google
Groups "ogs-users" group.
To unsubscribe from this group and stop receiving emails from it, send
an email to [email protected]
<mailto:[email protected]>.
For more options, visit https://groups.google.com/d/optout\.

--
_______________________________________
Dr. Christof Beyer
Institute of Geosciences
Geohydromodelling
Christian-Albrechts-University Kiel
Ludewig-Meyn-Str. 10
24118 Kiel
Germany

phone: +49(0)431-8803172
fax: +49(0)431-8807606
mobile: +49(0)176-24297908
email: [email protected]
home: http://www.ifg.uni-kiel.de/
_______________________________________

jens-olaf_Delfs · November 12, 2015, 12:27pm

i added density and viscosity models into the cheat sheet https://github.com/drjod/ogs_kb1/blob/master/README.md, since we often need that

···

2015-11-11 21:39 GMT+01:00 Christof Beyer [email protected]:

Hi Marius,

density model 4 is a linear density model, e.g.

$DENSITY

4 999.7 283.15 -2.5e-4

Parameters are rho(T_ref), T_ref, droh/dT. There are many other density models implemented, some of them are also taking into account concentration and/or pressure changes by such linearization.

You can also provide your own non-linear density model by ascii text input (“model 0”):

$DENSITY

0 1

Parameter “1” refers to the first data table in a *.rfd ascii input file, which you will have to provide in this case:

;Curve 1 Temp Density

#CURVES

273.15 999.8675792

274.15 999.9265054

…

#STOP

For T dependent viscosity, you can also provide a data file using the same syntax as for density, or you can use the empirical Yaws (model 3) or deMarsily (model 4) models

$VISCOSITY

3 ; Yaws

;4 ; DeMarsily

where the parameters are hardcoded in OGS. There are other model, which also take into account pressure and/or concentrations

Btw, you should define temperature in Kelvin.

Best, Christof

Am 11.11.2015 um 20:56 schrieb Mar. Bitt.:
Ladies and gentelemen,

I started working with OGS to compare it to other platforms solving heat

transport coupled with richards flow.

In fluid properties I can define a line where the density is dependant

of the temperature.

I did a bit research and found this page Inputfile .mfp density

<https://svn.ufz.de/ogs/wiki/public/doc-auto/by_ext/mfp/S_density> as

fard as I understood 4. gives a relation between density and

temperature. But I need to know how exactly OGS is calculating the

density. The formula given at point 4. and the values lower on the page

are rather confusing.

Density approximations are dependant of temperature ranges. And I can’t

see something that is modelling something like that here.

A co-worker said he is using this equation:
$DENSITY

  4 1.000000e+00300.2; C0 is 0. beta_C (drho_dC) is 0.2
but this looks for me on the first sight more like a concentration

related behaviour, also on the website it is described like that.

In addition viscosity should change too with temperature.

I’m new to OGS and I have to say I didnt find a lot about this stuff in

the web.

Best regards,

MB

–

You received this message because you are subscribed to the Google

Groups “ogs-users” group.

To unsubscribe from this group and stop receiving emails from it, send

an email to [email protected]

mailto:[email protected].

For more options, visit https://groups.google.com/d/optout.
–

Dr. Christof Beyer

Institute of Geosciences

Geohydromodelling

Christian-Albrechts-University Kiel

Ludewig-Meyn-Str. 10

24118 Kiel

Germany

phone: +49(0)431-8803172

fax: +49(0)431-8807606

mobile: +49(0)176-24297908

email: [email protected]

home: http://www.ifg.uni-kiel.de/

–

You received this message because you are subscribed to the Google Groups “ogs-users” group.

To unsubscribe from this group and stop receiving emails from it, send an email to [email protected].

For more options, visit https://groups.google.com/d/optout.

Mar_Bitt · December 4, 2015, 11:08am

Good evening ladies and gentlemen,

thank you for the answer cb!

I took into account what you said and investigated more about this.

I still does not understand how this linear approach is looking. From the input parameters maybe something like this is logical:

(The second formula shows the equation with the parameters that I gave to OGS in my example)

but this really would be only a rough approach. In addition this forumla only works with °C.

When it is really the case that the density is calculated like the formula that is given above I calculated my own values and put them as a #CURVE in a *.rfd file.

I worte a programm to write me a nice approach from 0°C-250°C or when needed 273.15°K-523.15°K. All my files were depending on °C and I dont get any nice results on the contrary at a certain time step the programm stops and windows tells me it stopped running properly also I get really high temperature results and I dont know from what they occur.

My #CURVES file looks like this:

link to .rfd file

I tried both: °K and °C but the same error occurs. I thought that maybe 273,15°K or (0°C) as initial value may be insufficient? Or is the accuracy not high enough?

Thank you for helping. All may have a nice weekend.

Best regards,

MB

(sorry for deleting my post three times, the link expired)

···

Am Mittwoch, 11. November 2015 20:56:15 UTC+1 schrieb Marius:

Ladies and gentelemen,

I started working with OGS to compare it to other platforms solving heat transport coupled with richards flow.

In fluid properties I can define a line where the density is dependant of the temperature.

I did a bit research and found this page Inputfile .mfp density as fard as I understood 4. gives a relation between density and temperature. But I need to know how exactly OGS is calculating the density. The formula given at point 4. and the values lower on the page are rather confusing.
Density approximations are dependant of temperature ranges. And I can’t see something that is modelling something like that here.

A co-worker said he is using this equation:

$DENSITY
4 1.000000e+003 0 0.2 ; C0 is 0. beta_C (drho_dC) is 0.2

but this looks for me on the first sight more like a concentration related behaviour, also on the website it is described like that.

In addition viscosity should change too with temperature.

I’m new to OGS and I have to say I didnt find a lot about this stuff in the web.

Best regards,

MB

Mar_Bitt · December 6, 2015, 7:34pm

I’m sorry somehow the link expires all the time.Hopefully this one works:

LINK TO *.rfd file

In addition I have another question. Is it possible to do exactly this for a temperature dependant viscosity?

···

Am Mittwoch, 11. November 2015 20:56:15 UTC+1 schrieb Marius:

Ladies and gentelemen,

I started working with OGS to compare it to other platforms solving heat transport coupled with richards flow.

In fluid properties I can define a line where the density is dependant of the temperature.

I did a bit research and found this page Inputfile .mfp density as fard as I understood 4. gives a relation between density and temperature. But I need to know how exactly OGS is calculating the density. The formula given at point 4. and the values lower on the page are rather confusing.
Density approximations are dependant of temperature ranges. And I can’t see something that is modelling something like that here.

A co-worker said he is using this equation:

$DENSITY
4 1.000000e+003 0 0.2 ; C0 is 0. beta_C (drho_dC) is 0.2

but this looks for me on the first sight more like a concentration related behaviour, also on the website it is described like that.

In addition viscosity should change too with temperature.

I’m new to OGS and I have to say I didnt find a lot about this stuff in the web.

Best regards,

MB

Wolf_Tilmann_Pfeiff1 · December 6, 2015, 7:51pm

Dear Marius,

I have to admit that I did not follow all replies concerning your question. If I understand it correctly you want to get the correct density for a given temperature using curves in the *.rfd file but your simulation crashes. I had a quick glance at your file and you are using comma as the decimal mark. You should switch that to periods:

#CURVES

273.15 999.8732768

273.151 999.8733405

…

Give it a try and see if that helps.

Concerning your question regarding temperature dependent viscosity: I think yes, but i have not checked it. The integer flag for reading data from a curve is always 0. So you could set your viscosity in addition to your density like this:

$VISCOSITY

0 2

In this case the second curve in the *.rfd file should contain the viscosity data.

All the best, Tilmann

Marius schrieb:

···

I’m sorry somehow the link expires all the time.Hopefully this one works:

LINK TO *.rfd file

https://drive.google.com/file/d/0B_S15Ubp4R77VzNvbTZVY1hsdEU/view?usp=sharing

In addition I have another question. Is it possible to do exactly this

for a temperature dependant viscosity?

Am Mittwoch, 11. November 2015 20:56:15 UTC+1 schrieb Marius:
Ladies and gentelemen,



I started working with OGS to compare it to other platforms

solving heat transport coupled with richards flow.



In fluid properties I can define a line where the density is

dependant of the temperature.



I did a bit research and found this page Inputfile .mfp density

<https://svn.ufz.de/ogs/wiki/public/doc-auto/by_ext/mfp/S_density> as

fard as I understood 4. gives a relation between density and

temperature. But I need to know how exactly OGS is calculating the

density. The formula given at point 4. and the values lower on the

page are rather confusing.

Density approximations are dependant of temperature ranges. And I

can't see something that is modelling something like that here.



A co-worker said he is using this equation:



    $DENSITY

     4 1.000000e+00300.2; C0 is 0. beta_C (drho_dC) is 0.2





but this looks for me on the first sight more like a concentration

related behaviour, also on the  website it is described like that.



In addition viscosity should change too with temperature.



I'm new to OGS and I have to say I didnt find a lot about this

stuff in the web.



Best regards,



MB
–

You received this message because you are subscribed to the Google

Groups “ogs-users” group.

To unsubscribe from this group and stop receiving emails from it, send

an email to [email protected]

mailto:[email protected].

For more options, visit https://groups.google.com/d/optout.

Wolf_Tilmann_Pfeiff1 · December 6, 2015, 8:05pm

Hey,

I just realized that maybe I should show you the syntax you should follow if you have multiple curves in the *.rfd file. Do not put a #STOP between the curves. The #STOP goes wherever you want ogs to stop reading the file. Everything behind it will not be used in any way. If you have multiple curves just separate them by a blank line followed by the second #CURVES keyword:

; Curve 1: kr_w(S) for mat group 1

#CURVES

0.6 0

…

0.9 0.00154

; Curve 2: kr_w(S) for mat group 2

#CURVES

0.3 0

…

#STOP

Wolf Tilmann Pfeiffer schrieb:

···

Dear Marius,

I have to admit that I did not follow all replies concerning your

question. If I understand it correctly you want to get the correct

density for a given temperature using curves in the *.rfd file but

your simulation
crashes. I had a quick glance at your file and you are

using comma as the decimal mark. You should switch that to periods:

#CURVES

273.15 999.8732768

273.151 999.8733405

…

Give it a try and see if that helps.

Concerning your question regarding temperature dependent viscosity: I

think yes, but i have not checked it. The integer flag for reading

data from a curve is always 0. So you could set your viscosity in

addition to your density like this:

$VISCOSITY

0 2

In this case the second curve in the *.rfd file should contain the

viscosity data.

All the best, Tilmann

Marius schrieb:

I’m sorry somehow the link expires all the time.Hopefully this one works:

LINK TO *.rfd file

https://drive.google.com/file/d/0B_S15Ubp4R77VzNvbTZVY1hsdEU/view?usp=sharing

In addition I have another question. Is it possible to do exactly this

for a
temperature dependant viscosity?

Am Mittwoch, 11. November 2015 20:56:15 UTC+1 schrieb Marius:

Ladies and gentelemen,

I started working with OGS to compare it to other platforms

solving heat transport coupled with richards flow.

In fluid properties I can define a line where the density is

dependant of the temperature.

I did a bit research and found this page Inputfile .mfp density

https://svn.ufz.de/ogs/wiki/public/doc-auto/by_ext/mfp/S_density as

fard as I understood 4. gives a relation between density and

temperature. But I need to know how exactly OGS is calculating the

density. The formula given at point 4. and the values lower on the

page are rather confusing.

Density approximations are dependant of temperature ranges. And I

can’t see something that is modelling something like that here.

A co-worker said he is using this equation:

$DENSITY

4 1.000000e+00300.2; C0 is 0. beta_C (drho_dC) is 0.2

but this looks for me on the first sight more like a concentration

related behaviour, also on the website it is described like that.

In addition viscosity should change too with temperature.

I’m new to OGS and I have to say I didnt find a lot about this

stuff in the web.

Best regards,

MB

–

You received this message because you are subscribed to the Google

Groups “ogs-users” group.

To unsubscribe from this group and stop receiving emails from it, send

an email to [email protected]

mailto:[email protected].

For more options, visit https://groups.google.com/d/optout.

–

You received this message because you are subscribed to the Google

Groups “ogs-users” group.

To unsubscribe from this group and stop receiving emails from it,

send an email to [email protected]
om

mailto:[email protected].

For more options, visit https://groups.google.com/d/optout.

Mar_Bitt · January 5, 2016, 8:23pm

Good evening ladies and gentlemen,

Even after changing the sings from comma as decimal sign to a point the results were not satisfactory. Howsoever I decided to take the linear (since the temperature differences are really small) approach that hopefully works as I understood it.

For temperature dependant viscosity I chosed the number 4 it is a approach by Marsily. (1987)

Now I just still have the problem, that the temperatures in my domain remain pretty low.

This case can be seen as closed.

Thanks everyone for the replies.

I wish a good start in the new year whoever might read this.

Greetings,

MB

···

Am Mittwoch, 11. November 2015 20:56:15 UTC+1 schrieb Marius:

Ladies and gentelemen,

I started working with OGS to compare it to other platforms solving heat transport coupled with richards flow.

In fluid properties I can define a line where the density is dependant of the temperature.

I did a bit research and found this page Inputfile .mfp density as fard as I understood 4. gives a relation between density and temperature. But I need to know how exactly OGS is calculating the density. The formula given at point 4. and the values lower on the page are rather confusing.
Density approximations are dependant of temperature ranges. And I can’t see something that is modelling something like that here.

A co-worker said he is using this equation:

$DENSITY
4 1.000000e+003 0 0.2 ; C0 is 0. beta_C (drho_dC) is 0.2

but this looks for me on the first sight more like a concentration related behaviour, also on the website it is described like that.

In addition viscosity should change too with temperature.

I’m new to OGS and I have to say I didnt find a lot about this stuff in the web.

Best regards,

MB