Hello
Running a simulation in Eclipse is not as much fun if it takes 8 hours to finish less than one year
My model is big but t is not a problem if I inject water only, or water and gas
The problem starts when I start to inject solvent, then the simulation will be slower than a snail, except the snail is faster than my simulation.
If anyone worked on injecting solvent in miscible WAG injection using Eclipse E100, would you please tell me what is the trick to run the simulation faster?
Best Regards
Hello Zainab,
You may set up the numerical terms in your model, such as:
minimum time step: 1E-10 days
maximum time step: 1000 days
the actual time steps for the first month after starting solvent injection to be daily.
Also, you can look at the E100 manual for other numerical terms setting.
Hope this help!
You may also consider using multi processors given each run and that depends on your PC capabilities. Sometimes, the run with single processor takes 8 hours; it takes around 2 hrs when you have 16 processors Or 3.5 hrs when you have 8 processors.
Zainab,
I don't remember how to implement that in Eclipse, but I know that in CMG.
If you want, I can put you in contact with a friend who used E300 for WAG-EOR applied on an Iraqi oil field. He did that during his PhD at Petronas.
https://www.researchgate.net/profile/Ahmed_Khalil_Al-Khazraji
Best,
You may also look at his related article at JNGSE:
10.1016/j.jngse.2017.02.020
This is one of the issues, I am dealing with Black oil E100 not E300
I feel I am the only person who is doing this simulation
Zainab,
I would call Schlumberger office in UK. They have an office in Aberdeen. An expert will be in touch with you and he can guide you step by step to solve your problem. They also can get access to your PC to show you how to fix the various problems.
- Badges
- Science topic