HUAN
-
Posts
13 -
Joined
-
Last visited
Posts posted by HUAN
-
-
Hi, Jia:
Appreciation for your help. But I still have some questions. Could you give me some ideas?
1. First, my ovjective is to construct a model without both the injection of external water at the left boundary and the export of mass at the right boundary. So, technologically, it is a strictly closed system without any interactions with the external system.
2. My first question is what the boundary condition 'normal' in the 'Flow' pane mean?
3. In X1t, to consutrct such a model, I need to do three steps: (1) setting the discharge as 0, (2) setting the left boundary condition as 'outlet', and (3) setting the right boundary as 'outlet'? Because in 'outlet' condition, diffusion and dispersion are precluded. Therefore, the mass will not cross such a boundary. If this is correct?
4. In X2t, I tried to set both the left and right boundaries as 'closed'. However, the simulation returns errors and does not run. Therefore, to my understanding, to construct such a model, I need to do three steps: (1) setting the left boundary as 'closed' or setting the discharge as '0', (2) setting the left boundary as 'outlet', (3) setting the right boundary as 'open', but the discharge as '0', and (4) setting the right boundary as 'outlet'. If this is correct.
Can you help me make sure if above methods are correct?
Cheers
Huan
-
Hi,
I am planning to run a reactive transport model with only diffusion. The ideal model is shown in below. The water in shale will be transported to sandstone by only diffusion. Therefore, I do not want an external water to flux in. So, I want to close the left boundary. However, I did not find a way to do so. Can we implement this model? If yes, by what way can I do do?
Appreciated.
Huan
-
Hi, Sir or Madam:
1. I am trying to construct a model to test a model in which there is no advection and only diffusion. Therefore, I set the discharge to 0 m/yr (see following figures).
2. However, results show that the fluid velocity is not 0. It also shows variations with evolving time.
3. Anyone has any ideas? Do I wrongly mix some conceptions?
Appreciation.
-
Hi, Jia:
Thank you for response. Following your suggestion, I will re-run the simulations.
Best Regards
Huan
-
Hi,
I just updated my old version of GWB (11) to the latest 2022 version. But, I cannot open my previous works using this 2022 GWB software.
Can you help me with this?
Huan
-
52 minutes ago, Jia Wang said:
Hello Huan,
Thank you for providing the information on your machine. With these specs, I think your computer isn't the primary issue here. Running a complicated Reactive Transport Model can often take a little bit of time to complete but if you find it to be too slow, it can be related to other issues. Here are a couple of things to check:
1) Check if one or more kinetic reactant reaction rates are too fast. If your reaction rates are really fast, the program will select really small time steps for your simulation. In that case, your run might take a lot more time. If you find that there are reactions that are occurring extremely fast relative to the time scale of your interest, you might consider using an equilibrium approach for those reactants.
2) Sometimes runs may take a long time if you are modeling complex biological reactions. In that case, you might want to adjust your model to use the reaction rate evaluated at the old time level, a fully explicit approach. This is controlled by the variable theta, which can range from 0 to 1, where 0 is the full weighting of the rate at the old time level and 1 is fully weighting at the new time level. By default, X1t, X2t, and React are set to an use an average of the kinetic rates at the old and new time level with theta set at 0.6. To change to a fully explicit model, go to Config --> Stepping -> theta = 0. You can find more information regarding theta in 4.1 (Setting kinetic reactions) in the GWB Reaction Modeling Guide.
If these suggestions didn't help with improving your run speed, please post your input file along with any custom databases so we can take a closer look.
Hope this helps,
Jia WangHi, Jia Wang:
GWB takes more than 99% of the CPU when I run X1t simulation (the attached figure). So I think CPU of our computer is out of date. This computer was bought in 2013. That is why I want to copy GWB to virtual machine or super computer in our university. Buying a new computer is also an option.
I will try your suggestions. But, can you tell me that if it is fesible/legal to copy GWB?
Huan
-
Hi, everyone:
Our group has a professional verison of GWB. But the simulation speed is too slow when I use X1t. For example, it may requires more than 3 days or more to finish one run. Sometimes, the simulation calculation just terminate at 30%, or even 90%. So, I think this maybe due to the simulation computer is too old.
So I want to copy the GWB software to the virtual machine or super computer in our university. Can anyone tell me that is it feasible?
If it is not feasible, I want to buy a new workstation. Can anyone give me some recommendations about the computer parameters?
Our current computer:
processor: intel(R) Xeon(R) CPU E5-1630 v3 @ 3.70GHz 3.70GHz
Installed memory (RAM): 32.0 GB
System type: 64-bit operating system, x64-based processor
Thanks for help.
Huan
-
Dear Jia Wang:
Deep appreciation. It works after adding the nucleus to the script. It is very useful.
What can I say? Maybe just thanks. But you solved a big problem for me. It has confused me for some time.
Best wishes for you.
Huan
-
Hi, Jia Wang:
Thank you very much for your quick response.
1. Sorry for missing my thermo dataset. I am trying to simulate the influence of CO2 on sandstone diagenesis. So, I give a value for fugacity of CO2 and therefore use pH to balance charge. I attached my dataset.
2. Now, I am testing kinetic precipitation rate law to constrain mineral precipitation, like quartz. But it did not work. Therefore, I just use feldspar (primary) and quartz (secondary) to test the kinetic precipitation rate law.
3. The dissolution rate is set according to the three-mechanism-rate in Palandri and Kharaka (2004):
The rate parameters are also derived from Palandri and Kharaka (2004):
4. I attached a React_output file. It shows that the quartz is supersaturated. The log (Q/K) is 1.0201. But it does not precipitate. So, I think maybe it is not related to dissolution rate of Albite.
5. I am thinking if the rate law is right or wrong. But I checked several times. I cannot think of any point is wrong. The rate law is :
rate = surface * ((acid) + (neutral) + (base))
Return rate
6. The nucleus density is also tested. I improved the nucleus density to as high as 10000 cm2/cm3, but it still did not work.
7. Questions:
I think we write same rate law for both mineral precipitation and dissolution, is it right? GWB determine if it dissolve or precipitate according to Q/K, is it right?
If a primary mineral might precipitate, should we set a value for nuclues density? For example, the rock initially has 5% kaolinite, 50%K-feldspar, and with injection of CO2, K-feldspar might dissolve into kaolinite. In this case, do we need to set nucleus density for kaolinite?
Appreciation for your help.
Best wishes
Huan
-
Hi, everyone:
I am trying to use kinetic rate law for mineral dissolution and precipitation.
Here, I give a simple example.
Initial mineral: Albite
Secondary mineral: Kaolinite, Quartz, Calcite
If I do not use kinetic rate law for precipitation of kaolinite, quartz and calcite, they would precipitate. However, if I use kinetic rate law, they do not precipitate.
Can anyone give me some advice? The rate law, nucleus density are set. I cannot figure out what is wrong. Very confusing.
Thank you very much.
Huan
-
1
-
Can we close the left boundary in X1t and X2t?
in The Geochemist's Workbench
Posted
Hi, Jia:
I appreciate your assistance. It works well now.
Cheers