Water cut delay

[Tejaswy]

Hello,
I am doing a test on Water coning in a Oil reservoir. Now the water should cone and it is coning but I have having a very early water cut. I was hoping to have the water cut after a couple of years of production rather than from day 1
I was hoping you guys could help me out.


PS we cannot change the grid dimensions, Perms and Poro. The rest we are free to edit.

[vinomarky]

Can you explain why you can't change perms? Does that extend to MULTZ modifiers? Putting low vertical perm region/layer(s) below your perfs is probably easiest way to delay watercut increase

If you cant change that, then reduce your offtake rate and/or try moving your perfs up (or water down)

Failing that, increase water density and/or reduce hydrocarbon density

Main levers you have are vertical perm, distance from water, production rate, mobility ratio and density difference

[Tejaswy]

I increased the water density to 64, there was a slight delay but it was only for 30 days, I still did have some water production in the first 30 days but i think that is negligible 0.08
Changed the well contact from OWC to almost the tip of the reservoir 3020 feet with increased water density....same as above
Production rate is 1000 bbls/day that is pretty standard for the well I think.
"mobility ratio and density difference " I have dead oil ....I will get back to you on this. I will look at the props section

[vinomarky]

Implicit in changing mobility ratios of course is not only visc changes, but also rel perm changes....

You have not explained why you can't alter the vertical perm

[vinomarky]

So I just took a look at your deck

First observation, you have a kv/kh ratio of 0.28 which is quite a bit higher than I would have thought would be realistic for 10m vertical blocks of 70mD rock. Kv/kh of 0.1, 0.05 or less is more in line with reality unless you have massive clean sand units.

Second observation, you have only 4 layers of 10m thickness (very coarse), your well is completed in the upper three layers, and your WOC is at the very top of the 4th layer...... and you didn't expect instant water? What would you think would happen in reality if someone perforated right to the WOC?

Even dropping the WOC to the bottom of the 4th layer (to give yourself some standoff) I'd still expect quick simulated WC increase due to numerical dispersion in the coarse blocks (ie a single block separating your well from the water)

If you really cant change the grid dimensions/granularity, and if you cant reduce PERMZ or apply a MULTZ then rel perm is about you only credible lever left, but bear in mind that while you may be creating a scenario which yields the sorts of delays you are after, it most likely would not properly reflect reality.

You make no mention as to whether this is a real scenario or not, why you are unable to change perms and whether that extends to vertical perms (an easy way to reduce coning)

[Tejaswy]

Thanks for your reply Vino, This is not a real world scenario. This is a problem of water coning made up for by my guide. Having said that I am pretty sure in the future course of Action we will reduce the PermZ but as of now I am stuck with it.

The plan after getting a good production rate,a few years of production and then use action keywords to convert the well into a injector well and pump in liquid of higher density to suppress the cone.

Second observation, you have only 4 layers of 10m thickness (very coarse), your well is completed in the upper three layers, and your WOC is at the very top of the 4th layer...... and you didn't expect instant water? What would you think would happen in reality if someone perforated right to the WOC?

I had moved the perforation to layer 2 and tried production, that din't really work either

I will look into the relative permeability curves.

[vinomarky]

If you plan on converting it to a water injector, and you reduce the rel perm to water to a point where water has difficulty flowing, then you may run into problems when starting to inject water

You may want to specify just the cells beneath and immediate proximity to the well as having the heavily retarded water rel perms (such as below for example), and the rest of the model with your original rel perm table - you define these different regions with the SATNUM keyword

If you have the flexibility, you could always do a MULTZ Box adjustment in the SCHEDULE section - example for you deck might be;
(insert below into SCHEDULE section before production starts)

----------------------------------------
BOX
20 30 20 30 3 4 /

MULTZ
242*0.01 /
-----------------------------------


And an example retarded water rel perm table;
SWOF
-- TABLE 2
-- Sw Krw Kro Pcow
-- Fraction Fraction Fraction psi
0.200 0.00000 1.00000 0.00000
0.235 0.00000 0.92595 0.00000
0.270 0.00000 0.85381 0.00000
0.305 0.00000 0.78366 0.00000
0.340 0.00000 0.71554 0.00000
0.375 0.00000 0.64952 0.00000
0.410 0.00000 0.58566 0.00000
0.445 0.00000 0.52405 0.00000
0.480 0.00001 0.46476 0.00000
0.515 0.00003 0.40789 0.00000
0.550 0.00010 0.35355 0.00000
0.585 0.00025 0.30187 0.00000
0.620 0.00060 0.25298 0.00000
0.655 0.00135 0.20706 0.00000
0.690 0.00282 0.16432 0.00000
0.725 0.00563 0.12500 0.00000
0.760 0.01074 0.08944 0.00000
0.795 0.01969 0.05809 0.00000
0.830 0.03487 0.03162 0.00000
0.865 0.05987 0.01118 0.00000
0.900 0.10000 0.00000 0.00000
1.000 1.00000 0.00000 0.00000 /

[Tejaswy]

Thanks vino,
I just increased the density of water to 64, put the well way at the top of layer 2 and used Implict. I have a watercut of 0.05 at 2 years. That's fair enough I think. i wonder what my guide will think of it

[Tejaswy]

I will have to ask my guide on the MultiZ command.Thanks for that. You have been great help to me

[Petro_Freak]

If not MULTIZ, why don't you try work on your drawdown ? Just an observation. Also like vino pointed out aptly, the Ky/Kz ratio is high. 0.07-0.1 seems better.

Hope this helps,

Cheers

[Tejaswy]

Thanks for the help. I reduced the perms and the watercut is significantly low now.
Although there is one thing that I was pointed out to.

The movement of the aquifer.
Should it not move in a "^" shape rather than the awkward pattern it is moving ?

Can you guys help ?

[vinomarky]

Re-post your newly modified deck to ensure we are on the same wavelength