Reservoir Simulation Tutorials: Barham Sabir Mahmood
Reservoir Simulation Tutorials: Barham Sabir Mahmood
GRID.
We want a grid that’s a little more interesting than those that are easily constructed
directly in the data file. The grid is normally generated from maps. Since that is not
covered in the course, we will use an alternative route, by the program FILL. You are
not required to learn or understand FILL, but for those who are interested, the input
file to FILL will be our starting point. (FILL can be a good alternative for those of
you who later may want to construct grids which are more complicated than those that
can be constructed inside Eclipse, but don’t need or want to go the full step to RMS or
FloGrid. May be relevant for some Master or PhD studies. If you don’t see this as
interesting or necessary, you can include the attached file EX3GRID.FILLED as a
black box.)
The reservoir slopes downward towards NW, with the four top corners at depths: NW:
2400m SMSL (Sub Mean Sea Level), NE: 2020 m SMSL, SE: 1880 m SMSL, SW:
2200m SMSL.
In addition, two points along the center north-south line are given: North center at
2120m SMSL, and South center at 1950m SMSL.
The grid doesn’t cover the entire envelope rectangle, in NW, a triangle 1,75km (EW)
x 2,5km (NS) is missing, in NE a triangle 200m (EW) x 2km (NS) is missing, and in
SE a triangle 2km (NS) with width the whole grid is missing.
In addition, the top of the grid is eroded, such that all blocks with I = 24-30 are
missing in the top layer, and all blocks with I=28-30 in layer 2.
Petrophysics:
All of the above is defined in the file EX3GRID.FILL. If you want to generate a grid
from this input you will need to install and run the program FILL on your PC.
Processing EX3GRID.FILL with FILL generates a file EX3GRID.FILLED (which is
also included in the files attached to this exercise).
The GRID section of your DATA file will consist of a single statement,
INCLUDE’ing this file.
Take a quick look at the file EX3GRID.FILLED, and confirm that it contains the grid
defined by keywords COORD and ZCORN as we should expect. Also look at
ACTNUM and some petrophysics.
Additional data:
The simulation shall be organized such that the oil is above the bubble point, 227 bars,
at all times, i.e. we consider dead oil. No free gas, i.e. Rs will be constant.
The oil-water contact is at 2250m SMSL.
Use a reference pressure of 320 bars at datum depth 1900m SMSL. Saturation
equilibration initialization shall be with cell center saturations.
Relative permeability:
Sw Krw Kro Pc
0.2 0 0.85 0
0.25 0.006 0.727 0
0.3 0.013 0.613 0
0.35 0.022 0.509 0
0.4 0.032 0.414 0
0.45 0.043 0.330 0
0.5 0.057 0.255 0
0.55 0.071 0.189 0
0.6 0.087 0.133 0
0.65 0.105 0.087 0
0.7 0.124 0.51 0
0.75 0.144 0.024 0
0.8 0.167 0.007 0
0.85 0.190 0 0
0.9 0.215 0 0
1 1 0 0
PVT-data:
Water has a volume factor of 1.024 at a pressure of 308.2 bars. At this pressure, water
compressibility is 4.64E-5 bars-1, and viscosity is 0.42 cP.
In Eclipse Office, the input data can be visualized graphically, e.g. relperm-curves.
Wells:
Define a single oil producer in grid block column 23, 38, completed in layers 1 – 5.
Use a well diameter of 0.31 m.
The initial oil rate will amount to approx. 2% of total STOIIP (Stock tank (i.e.
standard conditions) Oil Initially In Place) pr year. Remember to define bubble point
pressure as minimum allowed pressure. Set economic limits of min. oil rate 200
Sm3/day, max water cut 0.9. End run if these are violated.
Start date: 1.1.2018, define DATES or TSTEP each half year after that, for 10 years.
To get output that can be viewed in FloViz, you will need to use the RPTRST
keyword at the start of the SCHEDULE section (hasn’t been lectured yet.) Just copy it
from the Ex2.DATA file.
(RPTRST BASIC=5, FREQ=6 / )
The NOSIM keyword allows for data checking and reporting without running the
simulation.
You will probably not get any convergence error reports. But in case you do; are there
any easy fixes for that? (TUNING)