Production Allocation Method

[aanzola]

Hi guys!
I new in the page so I hope you can help me.

I am looking for a method to allocate production of wells when you just have the total production of the wells (for instance 15 wells) and the Well Head Pressure of each well.

I am looking for a way to calculate a factor that can be used to allocate the production using just those parameters (total production of the wells and the WHP of each well)

Do any of you know a way (practical one) to make the allocation in this case ?

Regards...

Alejandro

[06pg22]

from total production you mean Cumulative production or Production rate?

[aanzola]

from total production you mean Cumulative production or Production rate?

I mean Production rate. The production rate of the 15 wells. I am looking for a kind of numeric methods or something quickly to do even when the accuracy of the resullt dont be the best...

Regards... Alejandro

[kader_007]

Sponsor,
the best way to carry on a production allocation is to have first the well performance of all your wells (15 in your case), this is done through the Prosper (Petex software).

You have to provide the software with some data obviously, that concerne PVT, reservoir pressure and temerature (GOR, Water cut, salinity etc) that way you can generate for each well an IPR (inflow performance relationship) which will give you the variation of your bottomm hole flowing pressure versus rate.

To generate the aforesaid IPR you'll have also to choose the model you would like to predict with, for instance "PI entry" will request you to enter the Productivity Index of your well (PI) in addition to the Reservoir pressure, temperature, GOR and water cut; an other useful model would be "Darcy" through its "enter skin by hand" menu you can enter the skin if you have it (from well test interpretation) or enter a value of 0 (optimistic) to generate the curve.

the second step, would be to enter your completion caracteristics, well deviation, etc..under the menu "Equipement Data" that way you'l generate the lift curves or in other words, the flux equations through the tubing, called VLP (for vertical Lift Performance) under a Reservoir simulation model, they will be called VLC (C stands for Curve).
The VLP curves than, are generated entering the top node pressure (the well head) so the software can interpolate the points up to the Well Head..keep in mind that this method is called Nodal Analysis, that's why we use the words "top node2 etc.

Having the VLP ready, you'll have to plot the intersection between the 2 curves IPR/VLP, the intersection point will give you, your actual working parameters in terms of pressure and rate.

Doing this for all your wells, you'll get a good allocation.

Given also that you run some well test campaign (to assess the skin), some static pressure survey (to have a pressure profile in the well, and assess the contacts). dynamic surveys to compute the Productivity Indexes and separator tests to confirm what your well performance analysis is telling you. And doing it frequently!

A good practice would be to implement MultiPhase Flow Meters (MPFM) devices around your gathering systems, so you can check at the gathering level your rates (isolating a well line, and routing it through the MPFM).

Hope it'll help

[aanzola]

Thanks Kader!
I got what you mean but the main problem that I have is I do not have enough information for do this... I can assume that but I am looking for(if this exist ) a kind of method to this a numeric calculation for this even when it can not be accuracy... Do you have any idea how to use the Well Head Pressure of each well and the total production rate (for instance 15 wells) to make a estimated allocation of the production of each well.

Best regards... Alejandro

[kader_007]

aanzola,
sorry but this is the only way you can get to the "right" figures.

if you simplify your reservoir to a tank like system, that would mean that all your well have the same characteristics (petrophisical, PVT, etc) which is hardely likely to be the reality of complicated reservoirs.

If you assume, that all your well started in the same date, that they have the same porosity, the same permeability, the same hydrocarbon and water saturations, the same pressures with no depletion in any of them or the same depletion in all, than you can just divide your total production by the number of wells you have (asuming also you are keeping the same choke opening in all of the them)...you see that nowhere around the world you'll find such an easy manageable reservoir.

My advice is to do nothing unless you have the right data, at least pressure surveys for all the wells (static and dynamic), see what the original MDTs where saying in terms of gradients, contacts and pressure, try to report all the MDT points for all the wells to a Datum that you'll choose a little above the Free water level (FWL) or the Oil water contact (OWC) if your FWL is not clearly identified (bear in mind that FWL is different from OWC).
Try to figure out the behaviour of your system first, in terms of depletion (the points of the same gradient in the same well, will shift together to lower values in term of pressure).

Again if you don't have the right data, try to create them (not inventing them) but by implementing at least a separater test for each well for 12 or 24 hours if you can afford it.

good luck!

cheers
-Kader

[vinomarky]

I'd strongly advise GETTING more data

You may not have the data historically, but if you can arrange to have some orifice plates installed and/or conduct some allocation tests in your wells it will vastly aid the accuracy

The problem you face is that even if you characterize your VLP curves etc, you still need to make assumptions about the local reservoir pressure and kh around each of your wells - both of which are usually quite different between wells

Perhaps you can go to logs and (gu)estimate kh per well, then assume average reservoir pressure per well is the same - armed with this info you can draw an EWAG (Engineered Wild Ass Guess) as to an appropriate production allocation based upon WHP's. Start with WHP and observed fluid cuts, assume a rate (if friction is significant), from VLP curves calc FBHP, from (gu)estimated kh and Pr calc flow rate from Darcy (iterate if friction is significant). Because the absolute value is not as important as the relative values between your wells this would probably be a reasonable method to use - I am assuming here that your wells are flowing naturally. If you are pumping them then go to your pump parameters for allocation.

Another thing to check is that usually wells are occasionally shutin for periods. While they generally are not shutin for long enough for a true buildup, if you choose a standard shutin period (ie 12 hours/24 hours etc) that is often met, then plot the WHP of your shutin wells over time (you may well have to make estimates of the fluid cuts in the tubing) you may glean more information about connectivity, kh and pressure support in each of your well - your better kh wells will build up more rapidly, and your higher Pr wells will extrapolate a sqrt time plot to a higher point.