OGS Richards_Flow

Shao_Hua · October 8, 2014, 8:21am

Dear colleagues,

I am running a case using Richards_flow. A coupled 2D and 3D mesh should simulate fault zone and rock mass.

Now is my problem (see attached file):

  The newest version V5.5 is quite problematically concerning the calculated inflow rate and distribution (page 2).

  The V5.4 supplies also questionable velocity distribution (page 3).

  If I run only 3D without 2D fault. The pressure and velocity distribution look acceptable (page 4), similar to those from Groundwater_flow (page 1).

Does anyone have experience on it and can give me some suggestions.

Thanks in advance.

Hua

OGS_1.pdf (324 KB)

Georg_Kosakowski · October 8, 2014, 9:35am

Hi Hua,
I had a similar problem last year with the liquid flow module. What are your numerical settings and the permeability difference.
Increasing the accuracy of the linear solver might help. You might look up the discussion on the mailing list from last year.
Greetings, Georg

···

-----Original Message-----
From: ogs-users@googlegroups.com on behalf of Shao, Hua
Sent: Wed 10/8/2014 10:21 AM
To: ogs-users@googlegroups.com; ogs-devs@googlegroups.com; ogs6@googlegroups.com; Mailingliste opengeosys@ufz.de
Subject: [ogs-users] OGS Richards_Flow

Dear colleagues,

I am running a case using Richards_flow. A coupled 2D and 3D mesh should simulate fault zone and rock mass.

Now is my problem (see attached file):

1. The newest version V5.5 is quite problematically concerning the calculated inflow rate and distribution (page 2).

2. The V5.4 supplies also questionable velocity distribution (page 3).

3. If I run only 3D without 2D fault. The pressure and velocity distribution look acceptable (page 4), similar to those from Groundwater_flow (page 1).

Does anyone have experience on it and can give me some suggestions.
Thanks in advance.
Hua

--
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 ogs-users+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Norihiro_Watanabe2 · October 8, 2014, 11:44am

Hi Hua,

How did you set the fault permeability in your MMP file? Giving permeability in local coordinates of the fault (i.e. permeability along the fault) is supported only in Liquid flow and Groundwater flow. For other processes such as Richards flow, you need to provide a permeability tensor in global coordinates.

You can also modify the code to support it in Richards flow. The coordinate transformation for liquid and groundwater flow is done in CalCoefLaplace() in fem_ele_std.cpp. You need to copy lines below "if (dim > MediaProp->geo_dimension)" to the Richards flow part.

Best,
Nori

···

On 10/08/2014 10:21 AM, Shao, Hua wrote:

Dear colleagues,

I am running a case using Richards_flow. A coupled 2D and 3D mesh should
simulate fault zone and rock mass.

Now is my problem (see attached file):

1.The newest version V5.5 is quite problematically concerning the
calculated inflow rate and distribution (page 2).

2.The V5.4 supplies also questionable velocity distribution (page 3).

3.If I run only 3D without 2D fault. The pressure and velocity
distribution look acceptable (page 4), similar to those from
Groundwater_flow (page 1).

Does anyone have experience on it and can give me some suggestions.

Thanks in advance.

Hua

--
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 ogs-users+unsubscribe@googlegroups.com
<mailto:ogs-users+unsubscribe@googlegroups.com>.
For more options, visit https://groups.google.com/d/optout\.

--
Norihiro Watanabe, Dr.-Ing.
Department of Environmental Informatics (ENVINF)
Wissenschaftler

Helmholtz-Zentrum für Umweltforschung GmbH - UFZ
Helmholtz Centre for Environmental Research GmbH - UFZ
Permoserstraße 15 / 04318 Leipzig / Germany

norihiro.watanabe@ufz.de / http://www.ufz.de
Telefon +49 341 235 1806

Sitz der Gesellschaft: Leipzig
Registergericht: Amtsgericht Leipzig, Handelsregister Nr. B 4703
Vorsitzender des Aufsichtsrats: MinDirig Wilfried Kraus
Wissenschaftlicher Geschäftsführer: Prof. Dr. Georg Teutsch
Administrativer Geschäftsführer: Dr. Heike Graßmann

wenqing.wang · October 8, 2014, 11:48am

Right.

···

On 08.10.2014 13:44, Norihiro Watanabe wrote:

Hi Hua,

How did you set the fault permeability in your MMP file? Giving permeability in local coordinates of the fault (i.e. permeability along the fault) is supported only in Liquid flow and Groundwater flow. For other processes such as Richards flow, you need to provide a permeability tensor in global coordinates.

You can also modify the code to support it in Richards flow. The coordinate transformation for liquid and groundwater flow is done in CalCoefLaplace() in fem_ele_std.cpp. You need to copy lines below "if (dim > MediaProp->geo_dimension)" to the Richards flow part.

Best,
Nori

On 10/08/2014 10:21 AM, Shao, Hua wrote:

Dear colleagues,

I am running a case using Richards_flow. A coupled 2D and 3D mesh should
simulate fault zone and rock mass.

Now is my problem (see attached file):

1.The newest version V5.5 is quite problematically concerning the
calculated inflow rate and distribution (page 2).

2.The V5.4 supplies also questionable velocity distribution (page 3).

3.If I run only 3D without 2D fault. The pressure and velocity
distribution look acceptable (page 4), similar to those from
Groundwater_flow (page 1).

Does anyone have experience on it and can give me some suggestions.

Thanks in advance.

Hua

--
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 ogs-users+unsubscribe@googlegroups.com
<mailto:ogs-users+unsubscribe@googlegroups.com>.
For more options, visit https://groups.google.com/d/optout\.

Shao_Hua · October 24, 2014, 8:46am

Thank you Nori & Wenqing. I changed the dimension to 3 for my fault in the Richards flow model, the velocity is now correct. The question is then my definition about area will be used or ignored?

I have now another oscillation problem with mass transport for my big 3D fault/matrix model (110 K elements with all tri, tet, quad, prima elements, which length varied fro, 1 meter to 50 meters). I have varied delta t, dispersion length ... without success.

Have any one (CAU-colleague) idea, which parameter can be used for the stable solution in the mass transport?
Hua

···

-----Ursprüngliche Nachricht-----
Von: ogs-users@googlegroups.com [mailto:ogs-users@googlegroups.com] Im Auftrag von Wenqing Wang
Gesendet: Mittwoch, 8. Oktober 2014 13:48
An: ogs-users@googlegroups.com
Betreff: Re: [ogs-users] OGS Richards_Flow

Right.
On 08.10.2014 13:44, Norihiro Watanabe wrote:

Hi Hua,

How did you set the fault permeability in your MMP file? Giving
permeability in local coordinates of the fault (i.e. permeability
along the fault) is supported only in Liquid flow and Groundwater
flow. For other processes such as Richards flow, you need to provide a
permeability tensor in global coordinates.

You can also modify the code to support it in Richards flow. The
coordinate transformation for liquid and groundwater flow is done in
CalCoefLaplace() in fem_ele_std.cpp. You need to copy lines below "if
(dim > MediaProp->geo_dimension)" to the Richards flow part.

Best,
Nori

On 10/08/2014 10:21 AM, Shao, Hua wrote:

Dear colleagues,

I am running a case using Richards_flow. A coupled 2D and 3D mesh
should simulate fault zone and rock mass.

Now is my problem (see attached file):

1.The newest version V5.5 is quite problematically concerning the
calculated inflow rate and distribution (page 2).

2.The V5.4 supplies also questionable velocity distribution (page 3).

3.If I run only 3D without 2D fault. The pressure and velocity
distribution look acceptable (page 4), similar to those from
Groundwater_flow (page 1).

Does anyone have experience on it and can give me some suggestions.

Thanks in advance.

Hua

--
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 ogs-users+unsubscribe@googlegroups.com
<mailto:ogs-users+unsubscribe@googlegroups.com>.
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 ogs-users+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/d/optout\.

wenqing.wang · October 24, 2014, 9:38am

The area value given in mmp is used in calculation.

···

On 24.10.2014 10:46, Shao, Hua wrote:

Thank you Nori & Wenqing. I changed the dimension to 3 for my fault in the Richards flow model, the velocity is now correct. The question is then my definition about area will be used or ignored?

I have now another oscillation problem with mass transport for my big 3D fault/matrix model (110 K elements with all tri, tet, quad, prima elements, which length varied fro, 1 meter to 50 meters). I have varied delta t, dispersion length ... without success.

Have any one (CAU-colleague) idea, which parameter can be used for the stable solution in the mass transport?
Hua

-----Ursprüngliche Nachricht-----
Von: ogs-users@googlegroups.com [mailto:ogs-users@googlegroups.com] Im Auftrag von Wenqing Wang
Gesendet: Mittwoch, 8. Oktober 2014 13:48
An: ogs-users@googlegroups.com
Betreff: Re: [ogs-users] OGS Richards_Flow

Right.
On 08.10.2014 13:44, Norihiro Watanabe wrote:

Hi Hua,

How did you set the fault permeability in your MMP file? Giving
permeability in local coordinates of the fault (i.e. permeability
along the fault) is supported only in Liquid flow and Groundwater
flow. For other processes such as Richards flow, you need to provide a
permeability tensor in global coordinates.

You can also modify the code to support it in Richards flow. The
coordinate transformation for liquid and groundwater flow is done in
CalCoefLaplace() in fem_ele_std.cpp. You need to copy lines below "if
(dim > MediaProp->geo_dimension)" to the Richards flow part.

Best,
Nori

On 10/08/2014 10:21 AM, Shao, Hua wrote:

Dear colleagues,

I am running a case using Richards_flow. A coupled 2D and 3D mesh
should simulate fault zone and rock mass.

Now is my problem (see attached file):

1.The newest version V5.5 is quite problematically concerning the
calculated inflow rate and distribution (page 2).

2.The V5.4 supplies also questionable velocity distribution (page 3).

3.If I run only 3D without 2D fault. The pressure and velocity
distribution look acceptable (page 4), similar to those from
Groundwater_flow (page 1).

Does anyone have experience on it and can give me some suggestions.

Thanks in advance.

Hua

--
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 ogs-users+unsubscribe@googlegroups.com
<mailto:ogs-users+unsubscribe@googlegroups.com>.
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 ogs-users+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/d/optout\.