Advanced ECLIPSE User Course: Vertical Equilibrium Abingdon Technology Center Training

Advanced ECLIPSE User
Course
Vertical Equilibrium
Abingdon Technology Center Training
Schlumberger Information Solutions

19 March 2007
Copyright Notice
© 2005 Schlumberger. All rights reserved.
No part of this manual may be reproduced, stored in a retrieval system, or
translated in any form or by any means, electronic or mechanical, including
photocopying and recording, without the prior written permission of Schlumberger
Information Solutions, 5599 San Felipe, Suite 1700, Houston, TX 77056-2722.
Disclaimer
Use of this product is governed by the License Agreement. Schlumberger makes
no warranties, express, implied, or statutory, with respect to the product described
herein and disclaims without limitation any warranties of merchantability or fitness
for a particular purpose. Schlumberger reserves the right to revise the information
in this manual at any time without notice.
Trademark Information
Software application names used in this publication are trademarks of
Schlumberger. Certain other products and product names are trademarks or
registered trademarks of their respective companies or organizations.
Page 2 of 32 Advanced ECLIPSE User Course: Vertical Equilibrium 2

The Concept of VE
• Equilibrium is established in a cell instantaneously

• the phase hydrostatic potential is independent of depth within a cell
• no capillary transition zones
• the saturation distribution is a step function
Current
Water contact
displacing
oil Initial
contact
Connate water Critical oil
© 2005 Schlumberger Information Solutions. All rights reserved.
VE stands for Vertical Equilibrium. It describes a condition under which equilibrium is

established in a cell instantaneously when fluids move into the cell. With VE, there is no capillary
transition zone; the phase hydrostatic potential is the same at any depth within a cell; and the
phase saturation distribution is a step function.
In the above example, the cell was initially in equilibrium state with water below and oil above
the initial oil water contact in the cell. With VE, when water moves into the cell a new oil water
contact will be established instantaneously. Above the current oil water contact there is oil with
connate water. Between the current oil water contact and the initial oil water contact, there is
water with residual oil. In this case, water can flow to the neighbour cells through the side faces,
but can not flow through the top face because there is no mobile water near the top.
Compared to VE, the default dispersed flow assumes that all phases are distributed uniformly in a
cell. In the above example, with a dispersed flow model, water can move to the neighbour cells
through all the six faces provided that there are potential differences across the faces and the
phase saturation in the cell is above the critical saturation.
In a reservoir, if the vertical communication dominates the communication in horizontal
directions, with vertical fractures for example, the VE concept may provide a more accurate
description of fluid flow than the default dispersed flow option.

Conditions of VE
• Three forces that contribute to the potential gradient

• viscous force
• capillary force q
• gravity force h A
• Characteristic numbers
• viscous number: N =
887 q μ B
v
A
• capillary number: N c =
k v Δ Pc
h
• gravity number: N g = k v Δ ρ g cos θ

• ECLIPSE VE: Nc<Nv<Ng, Nc<<Ng
• The gravity dominates both capillary and viscous forces
There are three forces that contribute to the potential gradient in fluid flow through porous media,
namely, viscous force, capillary force and gravity force. The viscous force is mainly controlled by
the fluid viscosities, the capillary force by the capillary pressures between phases, and the gravity
force by the difference in fluid densities. Three characteristic numbers are defined, as shown in
the above slide, to measure these three forces - viscous number, capillary number and gravity
number. In the above equations, q is the flow rate, μ is the fluid viscosity, B, the formation
volume factor, kv, the vertical permeability, A, cross-sectional area, h, the thickness of the cell,
ΔPc, the capillary pressure difference, Δρ, the difference in fluid densities, θ, the dipping angle of
the cell. They are all in FIELD units.
The case in which Nv / Ng and Nv / Nc are small is given the name Vertical Equilibrium in
reservoir engineering. Coats (SPEJ, 1967) uses the ratio Nv / Nc to define a quality to help decide
whether VE prevails. He suggests a value of 50 or less to indicate that VE exists.
In ECLIPSE, in addition to Nv < Ng and Nv < Nc , it is also assumed that Nc<< Ng to have VE
conditions. This means that VE exists when gravity dominates both capillary and viscous forces.

Dispersed Flow vs. VE
• Dispersed flow: phases are distributed uniformly in a cell

• A phase can flow through its faces as long as it is mobile
• VE: step saturation distribution

• water may flow out of the bottom of the cell, but not out of the top
Current
Water contact
displacing
oil Initial
contact
Connate water Critical oil
In ECLIPSE, the default option is the dispersed flow. With this option, oil, water and gas are
assumed to be evenly distributed in a cell. An average saturation is used to look up the saturation
tables to get the relative permeability. A phase is mobile if its average saturation is larger than the
critical saturation and the phase can move to other cells through any face as long as there is a
potential gradient. If capillary pressure exists, a transition zone will develop between the gas zone
and the oil zone, and between the oil zone and the water zone.
With VE, however, oil, water, and gas are distributed in step functions with the less dense phase
above the denser phase depending on the position of the fluid contacts. There are no transition
zones. The relative permeability for a phase is calculated for each face based on the fraction of
the cell faces over which a mobile saturation exists. A phase can only move to its neighbor cells
through the face where the phase is mobile.
In the above example, with the default dispersed flow, water can move to its neighbor cells
through all the six faces provided that the average water saturation in the cell is above the critical
saturation and there are potential gradients between the cells. With VE however, water can only
move to its neighbor cells through the side faces and the bottom face, but not through the top face
even there is potential gradient because there is no mobile water over the top face.

Comparison Between Non-VE and VE
Above is shown a comparison in well water production between a VE run and a dispersed flow
run for the same model. As one can see, the two runs give very different results. In the VE run no
water is produced, while in the non-VE run, water breaks through in about 1150 days, and the
water production rate increases to 9700 STB/DAY at the end of the run. The reason for this will
be explained in the next slide.
If there is no water production observed from the history in this case, VE may better describe the
fluid flow in this reservoir. Of course one need to check whether the conditions of the reservoir
are closer to the VE conditions, i.e. the gravity force dominates the viscous force and capillary
force.

Comparison Between Non-VE and VE
VE1 – Non-VE case VE2 – VE case
The above shows the average water saturation distribution in a cross-section where the well is
completed, from the previous VE and Non-VE cases. Water moves towards the well much faster
in the dispersed flow run than in the VE run, therefore the non-VE run has much earlier water
breakthrough while the VE run does not have any water production.
This can be explained as follows. In the VE run, instantaneous equilibrium is established in each
cell, so water cannot flow to the next cell towards the producer P1 until the oil water contact
reaches the lowest point of the neighbour cell even though the average water saturation has
reached the critical saturation much earlier than when the water can move. In the dispersed flow
run, water moves to the next cell towards the producer as soon as the average water saturation
exceeds the critical saturation. This happens well before the oil water contact reaches the lowest
point of the next cell in the VE run.

Initialization
• item 9 of EQUIL is set to 0

• the fluid contacts are honored
accurately
• Arrays
• Depth of fluid contacts Sw=Swcon, Sg=Sgmax, So=1-Sw-Sg
• Tables
GOC
• Distinct Depth (at which one or
more corner points exist) Sw=Swcon, Sg=Sgcon, So=1-Sw-Sg
• Fractional volumes below DD

• Fractional face area below DD Sw=Swmax,
Sg=Sgcon,
OWC
So=1-Sw-Sg
For a VE run, during initialization the block center method (item 9 of keyword EQUIL is set to 0
in the SOLUTION section) is normally used to calculate the initial fluid in place. The initial
contact depths are exactly matched by the initial saturations, which yields a static initial solution.
This differs from non-VE runs where the block center fluid in place calculation may not give
accurate fluid in place when the centre of the block is near the oil water contact for large cells.
For a partial VE run where both rock curves and VE are used, which will be discussed later, an
non-zero value can be assigned to item 9 of keyword EQUIL, in which case a normal ECLIPSE
fine grid equilibration is performed using the rock curves.
ECLIPSE performs additional calculations for VE runs. For each active cell seven tables are
calculated and stored. These tables essentially are used to calculate the saturation functions.
In the first table, the eight corners of the cell (or those that are “distinct”) are sorted into order of
increasing depth. The fraction of the cell volume lying below each distinct depth is then
calculated (normalized to 1). This provides a table of fractional pore volume versus depth and, in
effect, relates the cell phase saturation to the locations of fluid contacts in the cell.
For each of the six faces(X+, X-, Y+, Y-, Z+, Z-) of the cell, the four ( or distinct) corner depths
are again sorted and the fractional face area below each distinct depth is calculated (normalized to
1). These tables will be used to calculate the relative permeability out of each of the six faces. To
speed up the convergence, flow out of the top of a cell is allowed to begin once the cell is 90%
displaced.

These internal tables can be written into the debug file by using keyword VEDEBUG in the
GRID section.

Fluid Distribution
Reg Sw Sg So
1 Swcon Sgmax 1-Sg-Sw
GOC
GAS Swcon 1-Sw-So Socrg
2
GOCc
OIL
3 Swcon Sgcon 1-Sw-Sg
OWCc
4 1-Sg-So Sgcon Socrw
WATER
OWC
5 Swmax Sgcon 1-Sw-Sg
During a VE run, the fluid distribution in a cell may have one or more of the following five
regions, as shown above:
Region 1: Above the highest gas oil contact. Gas has its maximum saturation, Sg=Sgmax. Water
has the connate saturation, Sw=Swcon. And the oil saturation is So= 1-Swcon-Sgmax.
Region 2: Below the highest gas oil contact, but above the current gas oil contact. This is the
region where gas has displaced oil, so the critical oil saturation with respect to gas remains
So=Socrg. Water has its connate saturation, Sw=Swcon. And the rest is the gas, Sg=1-Socrg-
Swcon.
Region 3: Below the current gas oil contact and above the current oil water contact. Water and
gas take their connate saturation values, and the rest is oil: Sw=Swcon, Sg=Sgcon, So=1-Swcon-
Sgcon.
Region 4: Below the current oil water contact, but above the lowest oil water contact. In this
region, water has displaced oil, so the critical oil saturation with respect to water remains, So=
Socrw. Gas takes its connate saturation, Sg=Sgcon, and the rest is water: Sw =1-Socrw -Sgco.
Region 5: Below the lowest oil water contact. Water takes its maximum saturation, Sw=Swmax.
Gas saturation is connate, Sg= Sgcon, and the rest is oil: So=1-Swmax-Sgcon.
All the endpoints are taken from the saturation tables or from the endpoints defined explicitly
with the endpoint scaling facility.

Solution of Flow Equations
KRW
• Calculates relative permeabilities

• based on the fractional area below the DD
• Calculates pseudo capillary pressure
• e.g. Pc=g(ρw-ρ0)(Zc-Zcentre) for O/W 0 1
• Zc - depth of current OWC; Zcentre - depth of cell centre Fractional area
• Solves equations as normal:

• Po, Sw, Sg, Rs, Rv
• Calculates current contacts (at the end of time step)
• based on the fractional volume below the DD
• S=average phase saturations
After the initialization, for each time step ECLIPSE solves the flow equations as normal. It is
only the way in which the relative permeability for each phase and the capillary pressure are
calculated that is different in the VE run from that in the dispersed flow run. In the VE run, the
relative permeability for each phase is calculated based on the table of the fractional areas below
the distinct depths for each cell face versus depth, and the endpoint relative permeability values.
The fraction of that face covered by the phase is obtained by looking up the pre-calculated table
using the current value for the fluid contact depth. The relative permeability is then equal to the
face fraction multiplied by the appropriate “end-point” relative permeability.
In VE, capillary pressure effect is accounted for to balance the extra pressure due to the
difference between the current fluid contact depths and the cell centre depth. The pseudo capillary
pressure is calculated by: Pc=g(ρw-ρ0)(Zc-Zcentre) for oil and water, where Zc is the depth of
current OWC and Zcentre is the depth of cell centre.
With relative permeability and capillary pressure being calculated as above, the flow equations
are solved for Po, Sw, Sg, Rs, Rv as normal.
The depths of the current fluid contacts can then be calculated from the average fluid saturations
based on the table of the fractional volumes below the distinct depths versus depth.
Repeat the above steps for the next time step.

Partial VE
• Using partial rock curve and

partial VE
• VEFRAC: defines the VE VE curve
relative permeability fraction
• VEFRACP: defines the VE Real curve
capillary pressure fraction Rock curve
• VEFRACV & VEFRACPV:
allow to define the above
fractions on a cell-by-cell basis
E.g. in the PROPS section:
VEFRAC
0.8/
VEFRACP
0.8/
ECLIPSE will use 80% VE and 20% rock curves for both Kr and Pc
In reality, a reservoir may behave between VE and dispersed flow. In other words, the saturation
curves of the reservoir may lie between the rock curves and the VE straight lines. ECLIPSE
provides a facility called partial VE to simulate this scenario. Basically, a fraction, say 0.8, can
be defined by keyword VEFRAC so that ECLIPSE uses 80% of VE relative permeability plus
20% of rock relative permeability to get the total relative permeability that is used in the flow
equations.
Similarly, one can use keyword VEFRACP to define a fraction for the capillary pressure
calculation.
For example,
VEFRAC
0.8 /
VEFRACP
0.8 /
tells ECLIPSE to calculate the relative permeability and capillary pressure by
Kr=0.8*Kr_ve+0.2*Kr_rock
PCr=0.8*PC_ve+0.2*PC_rock
Where Kr_ve, and PC_ve are the VE relative permeability and capillary pressure, and Kr_rock
and PC_rock are the the permeability and capillary pressure from the rock curves respectively.
Keywords VEFRACV and VEFRACPV allow to define such fractions on a cell by cell basis.

Residual Flow Model
oil water
• In the VE run, the residual oil
Kr
remains immobile which in some
of the cases may be slightly mobile
• To simulate the case: Sw
=
• additional saturation tables
are entered for the residual Kr
oil
oil and referred by RESIDNUM water

• The total relative permeability
Krtot=Krve+Krr (So) VE curves (SATNUM) Sw
• Krve: VE Kr
• Krr: Kr from the sat table (referred Kr
+
to by RESIDNUM)
• So: oil saturation oil
water
Residual curves (RESIDNUM) Sw
In a VE run, the residual oil remains immobile. But in some cases, the residual oil may be slightly
mobile, and over a period of time some of this oil will flow slowly, reducing the oil saturation to
a new residual value. The residual flow option in VE is designed to simulate the VE behavior, i.e.
straight line for the saturations above the residual saturation, together with the residual flow. To
implement the residual flow option, a second set of saturation tables is defined to simulate the
residual flow. These tables are referred by keyword RESIDNUM which, similar to SATNUM,
is defined in the REGIONS section. The RESIDNUM tables will typically have very small oil
relative permeability values, with zero relative permeability values for phases with no residual
flow, for the saturations below the original residual saturation. Any capillary pressure values
entered in the RESIDNUM saturation tables will be ignored.
The total relative permeability in a VE run with residual flow is calculated by:
Krtot=Krve+Krr (So)
Where
Krve - VE Kr
Krr - Kr from the RESIDNUM table
So: oil saturation
This is illustrated in the above diagram – the relative permeability curves for a residual flow
model are obtained by combining the VE curves with the endpoints from the SATNUM table,
and with the residual curves referred by RESIDNUM.

Faults and VE
• Dispersed flow:
• water can flow from A to B
and A to C
C
• VE:
• water can flow from A to B
but not from A to C
B A OWC
In a VE run, because the relative permeability is related to the fraction of the area that is covered
by a phase in each face of a cell, the flow across a fault can be different in a VE run from that in a
non-VE run. In the above example, block cell A is connected to cells B and C due to the fault. In
a dispersed flow model, water can move from block cell A to B, and A to C as well provided that
the potential in block A is larger than the potential in B and C, and that the average water
saturation in cell A is above the critical saturation. In a VE model however, water can only move
from A to B, but not from A to C because the OWC has not reached the lowest point of the
interface between A and C. If the fault is very conductive, the VE option is better than the
dispersed flow option.

Block Centre Geometry and VE
• No difference between dip and fault

• Water cannot flow from A to B
because the OWC is lower than the
bottom of cell B B
• Better use corner point grid
• OPTIONS OWC A
• switch 19 = 1: water can flow from
A to B. Kr is based on the full face
of A.
• switch 19 =2: ECLIPSE treats the
block centre grid as corner point
grid
For block centre geometry, the top face and the bottom face of a cell are flat. In this case an
unrealistic step function change in the relative permeability would be obtained when the fluid
contact reaches the top of a cell, which would cause difficulties in solving the non-linear flow
equations. To avoid this, the cell top is ‘spread’ slightly, over 10% of the cell depth, so that the
flow relative permeability rises from zero to its maximum value as the fluid contact traverses this
interval.
In block centre grid cells, a dip and a fault cannot be differentiated. The flow from one cell to its
neighbour cells in the uphill direction is similar to the flow across a fault as discussed in the
previous slide. In the above example, water cannot flow from A to B in a VE run if the water
level in cell A is below the bottom of cell B. It would be better to use corner point geometry for
dipping reservoirs. If switch 19 of keyword OPTIONS is set to 1, the full face of each cell will
be specified for flow. And if switch 19 of keyword OPTIONS is set to 2, the flow between A and
B will be calculated as if a corner point geometry is used. In this case, the pore volume and
transmissibilities are preserved, however the VE flow calculations are performed assuming the
cells to be part of a continuous layer.

Compressed VE
• The entire reservoir section is in vertical equilibrium

• It takes full account of the reservoir heterogeneity
• The individual layers are automatically merged in the z-direction
into compressed VE cells.
• Internal tables are used to account for the variation in rock
properties across the width of the merged layers
• The merging of layers can be only carried out over particular
sections of the reservoir
• e.g. a shale breaks, non-VE layers
In the standard VE, equilibrium is established instantaneously in each cell. In some cases, if the
vertical communication is much higher than the horizontal directions and the gravity forces
dominate the viscous and capillary forces throughout the entire reservoir, equilibrium may be
established over the entire vertical section of the reservoir. The compressed VE option in
ECLIPSE is designed to simulate this situation. With this option on, the entire reservoir is in
vertical equilibrium – the phases will be distributed with the less dense phases above the denser
phases.
This option still takes full account of the reservoir heterogeneity – one can have 3D model having
many geological layers with different properties for example. When this option is on, all the
layers in the model are merged automatically in the z direction into compressed VE cells. If there
are vertical flow barriers, the merging of layers is only carried out over particular sections of the
reservoir, rather than across the whole width. It would form several VE sections. Non-neighbour
connections are generated automatically between VE sections. ECLIPSE accounts for the
variation in rock properties across the width of the merged layers using internal tables. These
tables consist of a set of data items tabulated against distinct depths of the the corner points of the
original cells that make up the compressed VE cell.

Breaking the Compressed VE Cells
• A cell can be prevented from being merged with the active cell
below in a column if:
• 0 is given to a cell in the COLLAPSE keyword
• TRANZ=0: impermeable barrier
• PERMZ<= the critical value defined in CRITPERM
• An inactive cell with DZ>0 between the two active cells
• A cell has been refined using RADFIN or CARFIN
• The column will then be represented by two or more compressed
VE cells
When ECLIPSE compresses grid cells into the compressed VE cells, the following situations
prevent a cell from being merged with the active cell below in a column:
A value of 0 is assigned to a cell by keyword COLLAPSE in the GRID section
There is a vertical impermeable flow barrier with TRANZ=0
The vertical permeability is equal or less than a critical values defined by keyword CRITPERM.
If the CRITPERM keyword is absent, the critical permeability defaults to 0.
An inactive cell with a non-zero thickness DZ between two active cells
A cell has been refined using RADFIN or CARFIN. The compressed VE option can be used in
conjunction with Local Grid Refinement. Any cells that have been refined are not placed in
compressed VE merged cells, as a more detailed description of the reservoir is required in these
areas. Compressed VE cells will be generated adjacent to the refinement and will be
automatically connected by non-neighbour connections.
In other words, any of the above situations will prevent the cells from being merged into a single
collapsed VE cell. The column will then be represented by two or more compressed VE cells.
Non-neighbour connections will be generated wherever required.

Tabular Description of the Cells
• Data items vs. distinct depths (corner points of the original cells
that make up the compressed VE cells)
• Data items for the cell volume table
• VOLUMES below the distinct depths
• saturation end points
• relative permeabilities at the end points
• Data items for the cell face table
• AREAS below the distinct depths
• relative permeability at the end points
• The internal tables can be examined by using keyword VEDEBUG
in the GRID section
As discussed previously, internal tables are set up to describe the the variation in rock properties
across the compressed VE cells. In these tables, data items are tabulated against depth of distinct
corner points of the original cells that make up the compressed VE cells. There are tables for the
cell volume and tables for each face of the cell. The data items for the cell volume tables include
normalized fraction of volume that is below each distinct point, the saturation endpoints and the
corresponding relative permeability values. The data items for the cell face tables include
normalized fraction of area that is below each distinct point, the relative permeability values at
saturation endpoints.
The endpoints include:
SOCRW – critical oil saturation with water
SOCRG - critical oil saturation with gas
SWCR - critical water saturation
SGCR - critical gas saturation
SOWMX - oil saturation at maximum water gas
SOGMX - oil saturation at maximum water gas
SWCO - connate water saturation
SGCO - connate gas saturation
KRO1 - Kro at SWCO, SGCR
KRO2 - Kro at SWCO, SGCO

KRO3 - Kro at SWCR, SGCR
KRW1 - Krw at SOCRW, SGCO
KRW2 - Krw at SOWMX, SGCO
KRG1 - Krg at SOGMX, SWCO
KRG2 - Krg at SOCRG, SWCO
These internal tables can be written into the debug file by using keyword VEDEBUG in the
GRID section.

Using the VE Options (1)
• RUNSPEC section
VE
‘NOCOMP’/ standard VE
‘COMPRESS’/ compressed VE
To use the VE option, in the RUNSPEC section keyword VE is entered with an option of
‘NOCOMP’ for the standard VE, or an option of ‘COMPRESS’ for the compressed VE.
Example:
RUNSPEC
VE
‘COMPRESS’ / the compressed VE will be used
Or
VE
NOCOMP / the standard VE will be used. This is the default option.

• GRID section
• COLLAPSE: prevents a cell from being merged with the active cell below in
the compressed VE run
• CRITPERM: sets critical PERMZ value
• VEFIN: defines whether VE is used in the LGR (default - no)
• VEDEBUG: request output for internal tables for compressed VE
In the GRID section, cells can be given a value of 0 for array COLLAPSE so that they are not
merged with active cells to form the compressed VE cells. Keyword CRITPERM can also be
used to set a critical PERMZ value. If a cell has PERMZ that is below the critical value, the cell
will not be merged into the compressed VE cell. When a LGR is defined in the GRID section,
keyword VEFIN can be used to specify whether VE is used in the LGR. The default is that
dispersed flow is applied within the LGR even the global grid is using VE. Keyword VEDEGUG
can be used to request output of the internal tables that are used by VE.
Example:
GRID
EQUALS
COLLAPSE 0 1 20 1 25 2 2 / cells in the box (1-20, 1-25, 2-2) will not be
/ collapsed into the compressed VE cells
CRITPERM
0.1 / Cells whose PERMZ <= 0.1 will not be merged into compressed VE cells.
CARFIN
LGR01 1 2 2 3 1 3 4 6 6 1/
VEFIN
YES / VE will be applied within the LGR

ENDFIN
VEDEBUG
1 2 1 2 1 2 / the internal tables for cells in the box (1-2, 1-2, 1-2) will be written into the
debug file

• PROPS section
• VEFRAC, VEFRACP, VEFRACV, VEFRACPV: define VE weighting factors
for Kr and PC
• Saturation tables are defined as normal.
• Extra set of Saturation tables are defined for residual flow
• REGIONS section
• SATNUM: refers to the sat table for VE Kr
• RESIDNUM: refers the sat table for the extra Kr
In the PROPS section, the fractions with which VE is used to calculate the relative permeability
and capillary pressure are defined by keywords VEFRAC and VEFRACP. These fractions can
also be defined on cell by cell basis using keywords VEFRACV and VEFRACPV. Saturation
tables are defined as normal. The endpoints of these tables will be used by the VE, and the rock
curves will be used by the partial VE. Extra set of saturation tables need to be defined if residual
flow option is used.
In the REGIONS section, SATNUM is used to refer the saturation tables for the VE and partial
VE. RESIDNUM is used to refer the saturation tables that are used for the residual flow.
Example:
PROPS
SWOF
…… / table 1 for SATNUM
…… /table 2 for RESIDNUM
VEFRAC
0.8 / Kr=0.8*Kr_VE+0.2*Kr_rock
VEFRACP
0.7 / PC=0.7*PC_VE+0.3*PC_rock
REGIONS

EQUALS
SATNUM 1 / all the cells use SAT table 1 for the VE or the partial VE
RESIDNUM 2 / all the cells use SAT table 2 for residual flow
/

• SCHEDULE section
• COMPVE/ COMPVEL: defines well’s
top and bottom depths of the Level at which water will
connection so that partial penetration flow into the well
can be accounted for
• RPTSCHED (mnemonic WOC/GOC):
controls the output of the oil-
water/gas-oil contact depths at report Sw=Swc
steps
OWC
In the SCHEDULE section, COMPVE or COMPVEL (when wells are completed in LGRs) can
be used to reset the top and bottom depths of well connections, to allow for partial penetration
within a grid block when calculating the phase relative permeability values at the connection. The
skin factor can also be adjusted to allow for partial penetration.
In runs using the Vertical Equilibrium option, the relative permeability values at a well
connection depend upon the perforation's top and bottom depths and the fluid contact depths. By
default (without COMPVE keyword), the top and bottom depths of each connection are set to
the depths of the centre of the top and bottom faces of the connecting grid block, giving full
penetration of the grid block.
The following mnemonics in keyword RPTSCHED can be used to output fluid contact depths
and their variants: GOC(current gas oil contact depth), GOCDIFF(current GOC depth – initial
GOC depth), WOC(current oil water contact depth); WOCDIFF(current WOC depth – initial
WOC depth); WOCGOC(current WOC depth – current GOC depth).
Example:
SCHEDULE
RPTSCHED
GOC WOC / output current GOC and WOC at report steps
COMPVE

PROD1 4* 1 1.0 2010 2050 YES / sets for connections of well PROD1 the
/ saturation table number(1), vertical equilibrium fraction (1.0),
-- top-of perforation depth (2010), and bottom-of perforation depth(2050).
-- The skin factor and connection transmissibility factor are recalculated to
-- account for partial penetration.

Some Notes
• A VE run cannot restart from a non-VE run as info about the lowest and highest
contact depths is required
• In a VE run with corner point geometry where there is a gap between a pair of
adjacent active cells in a column, the contact depths cannot be resolved in the
dead space
• OPTIONS(52)= 2: bottom corner point depth = top corner point depth
• OPTIONS(4, 5): use critical saturation when calculating the contact depths
• prevent gas/water from being mobile when Scon<S<Scr
• but the VE equilibration is not stable, as Scon is always used in the
equilibration
A VE run cannot restart from a non-VE run because information about the lowest and highest
contact depths is required in a VE run, but not available in a non-VE restart file.
In a VE run with corner point geometry, if there is a gap between a pair of adjacent active cells in
a column, for example the pinched out cells due to MINPV, the contact depths cannot be
resolved in the dead space. To overcome this problem, setting switch 52 of keyword OPTIONS
to 2 results in that the bottom corner point depths for each cell are modified to be consistent with
the top depths of the next active cell below in the same column. If LGR is used, ECLIPSE always
modifies the corner point depths in local grids to avoid gaps between local and global grids where
there is a connection.
During the VE calculation, the connate saturation of a phase was always assumed to be present
throughout a cell. As soon as this is exceeded, a contact forms, and flow is generally possible
through the top or bottom of the cell. This will usually occur at a lower saturation than the critical
value. For example, the connate gas saturation is normally zero. If any gas saturation exists in a
cell it will be assumed to be concentrated at the top, and thus be mobile through the top of the
cell, even if it is below the critical gas saturation. This is a straightforward consequence of the
assumption of vertical equilibrium, but may not be appropriate in some cases. For example, if gas
appears in a cell through bubble point crossing, it would be evenly distributed through the cell,
and could not concentrate at the top until the critical, rather than the connate, saturation is

reached. Switches 4 and 5 of the OPTIONS keyword enable the critical water and gas saturation
values to replace the connate in the simulation if required. However, this may result in the VE
equilibration option not being static, as VE equilibration always uses connate saturations to
calculate the fluid contact depths.

References
• ECLIPSE Technical Description

• Vertical Equilibrium

Advanced ECLIPSE User Course: Vertical Equilibrium Abingdon Technology Center Training

Uploaded by

Copyright:

Available Formats

Advanced ECLIPSE User Course: Vertical Equilibrium Abingdon Technology Center Training

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Advanced ECLIPSE User Course: Vertical Equilibrium Abingdon Technology Center Training

Uploaded by

Copyright:

Available Formats

Advanced ECLIPSE User

Abingdon Technology Center Training

Schlumberger Information Solutions

Page 2 of 32 Advanced ECLIPSE User Course: Vertical Equilibrium 2

• Equilibrium is established in a cell instantaneously

Connate water Critical oil

© 2005 Schlumberger Information Solutions. All rights reserved.

VE stands for Vertical Equilibrium. It describes a condition under which equilibrium is

Page 3 of 32 Advanced ECLIPSE User Course: Vertical Equilibrium 3

• Three forces that contribute to the potential gradient

• gravity number: N g = k v Δ ρ g cos θ

© 2005 Schlumberger Information Solutions. All rights reserved.

Page 4 of 32 Advanced ECLIPSE User Course: Vertical Equilibrium 4

• Dispersed flow: phases are distributed uniformly in a cell

• VE: step saturation distribution

Connate water Critical oil

© 2005 Schlumberger Information Solutions. All rights reserved.

Page 5 of 32 Advanced ECLIPSE User Course: Vertical Equilibrium 5

© 2005 Schlumberger Information Solutions. All rights reserved.

Page 6 of 32 Advanced ECLIPSE User Course: Vertical Equilibrium 6

VE1 – Non-VE case VE2 – VE case

© 2005 Schlumberger Information Solutions. All rights reserved.

Page 7 of 32 Advanced ECLIPSE User Course: Vertical Equilibrium 7

• item 9 of EQUIL is set to 0

• Fractional volumes below DD

© 2005 Schlumberger Information Solutions. All rights reserved.

Page 8 of 32 Advanced ECLIPSE User Course: Vertical Equilibrium 8

Page 9 of 32 Advanced ECLIPSE User Course: Vertical Equilibrium 9

© 2005 Schlumberger Information Solutions. All rights reserved.

Page 10 of 32 Advanced ECLIPSE User Course: Vertical Equilibrium 10

• Calculates relative permeabilities

• Solves equations as normal:

© 2005 Schlumberger Information Solutions. All rights reserved.

Page 11 of 32 Advanced ECLIPSE User Course: Vertical Equilibrium 11

• Using partial rock curve and

© 2005 Schlumberger Information Solutions. All rights reserved.

Page 12 of 32 Advanced ECLIPSE User Course: Vertical Equilibrium 12

oil and referred by RESIDNUM water

Residual curves (RESIDNUM) Sw

© 2005 Schlumberger Information Solutions. All rights reserved.

Page 14 of 32 Advanced ECLIPSE User Course: Vertical Equilibrium 14

© 2005 Schlumberger Information Solutions. All rights reserved.

Page 16 of 32 Advanced ECLIPSE User Course: Vertical Equilibrium 16

• No difference between dip and fault

© 2005 Schlumberger Information Solutions. All rights reserved.

Page 17 of 32 Advanced ECLIPSE User Course: Vertical Equilibrium 17

• The entire reservoir section is in vertical equilibrium

© 2005 Schlumberger Information Solutions. All rights reserved.

Page 18 of 32 Advanced ECLIPSE User Course: Vertical Equilibrium 18

© 2005 Schlumberger Information Solutions. All rights reserved.

Page 20 of 32 Advanced ECLIPSE User Course: Vertical Equilibrium 20

© 2005 Schlumberger Information Solutions. All rights reserved.

Page 21 of 32 Advanced ECLIPSE User Course: Vertical Equilibrium 21

Page 22 of 32 Advanced ECLIPSE User Course: Vertical Equilibrium 22

© 2005 Schlumberger Information Solutions. All rights reserved.

Page 23 of 32 Advanced ECLIPSE User Course: Vertical Equilibrium 23

© 2005 Schlumberger Information Solutions. All rights reserved.

Page 24 of 32 Advanced ECLIPSE User Course: Vertical Equilibrium 24

Page 25 of 32 Advanced ECLIPSE User Course: Vertical Equilibrium 25