Keyword Descriptions

The list of keywords given below include:-

Eclipse keywords used in the TransGen <Project>.TGDATA file (generated by the WizGen module) to determine what Eclipse data and how they are included in the TransGen calculations (via the ViewGen module) of permeability and transmissibility multipliers from the Fault Seal Potential measures (e.g. DIMENS, COORD, ZCORN, NTG, PERMX, PERMY).

TransGen specific keywords to enter information specific to the calculations performed by TransGen (e.g. TGAXES, TGFSP, TGMETRIC, TGNEWKEY,TGPLUGIN , TGSHALE) All TransGen-specific Keywords start with TG.

TransGen specific keywords to input/output fault trace properties when using the new fault drag and hierarchical zone effects functionality (i.e. TGTRACE, TGDRAG, TGTHROW, TGFZONE, TGXTRACE).

Eclipse keywords used in the TransGen <Project>.TGDATA file (generated by the WizGen module) to determine what Eclipse format data is output from TransGen to include modifications to the transmissibilities in the parent simulator model (i.e. EDITNNC, TRANX, TRANY, NNC).

The format necessary for inclusion of these keywords in a TransGen run is described below.

NOTE:- It is recommended that you use the WizGen module to add all the relevant keywords to the instruction file (<Project>.TGDATA) in the correct order and that you do not attempt to edit the TGDATA manually.

ACTNUM

Active Grid Block Identification

This is one of the Eclipse keywords which can be used in the Eclipse simulation model to control which cells are active/inactive (see also MINPV, MINPVV and PORO). If the ACTNUM keyword is included in the Eclipse model, it must also be included in the TransGen run file to ensure the same cells are active/inactive in both models.
The ACTNUM keyword is followed by one integer for every grid block in the current box.  A value of 1 indicates that the corresponding grid block is active, whilst a 0 indicates that it is inactive.  The data must be terminated by a slash (/).


Grid blocks are ordered with the X axis index cycling fastest, followed by the Y and Z axis indices.  Repeat counts may be used for repeated values (e.g. 12*0).  Note that spaces may not be inserted on either side of the asterisk.

ADD

Adds a constant to the specified array in current box

This is one of the Eclipse keywords which can be used to modify data arrays in the Eclipse simulation model (see also COPY, EQUALS and MULTIPLY). If the ADD keyword is included in the Eclipse parent model, it should also be included in the TransGen run file to ensure data compatibility between the two models.
The ADD keyword can be followed by any number of records, each of which is terminated by a slash (/).  The data is terminated by a null record (i.e. a record with no data before the terminating slash).  Each record consists of at least 2, and up to 8 items of data.

Item 1  The name of the array to be modified (enclosed in quotes).

Item 2  The constant to be added to the array specified by item 1.  The constant may be positive or negative, and may be real or integer.

Items 3-8 may be used to redefine the input box for this and subsequent operations within the current keyword. If items 3-8 are not defined (a slash is inserted after item 2), they default to the values which were used for the previous operation within the current keyword.  For the first operation in the keyword, the box defaults to the values set by the most recent BOX or ENDBOX keyword.  If there is no preceding BOX or ENDBOX in the current section, the box is taken to include the entire reservoir.

Item 3  IX1   First block to be modified on the X axis.

Item 4  IX2   Last block to be modified on the X axis.

Item 5  JY1   First block to be modified on the Y axis.

Item 6  JY2   Last block to be modified on the Y axis.

Item 7  KZ1   First block to be modified on the Z axis.

Item 8  KZ2   Last block to be modified on the Z axis.

The data must satisfy

        1 <= IX1 <= IX2 <= NDIVIX
        1 <= JY1 <= JY2 <= NDIVIY
        1 <= KZ1 <= KZ2 <= NDIVIZ

BOX

Re-defines the current input box

This Eclipse keyword when included in the parent simulator model should be followed by six integers which re-define the current input box.  Subsequent operations using COPY or MULTIPLY only alter grid blocks within the current input box.  Similarly, data read into an array (e.g. using PERMX or PORO) are assigned to the grid blocks in the current input box.

N.B. BOX remains active until it is disabled with ENDBOX or re-defined with another BOX command. If you forget to disable BOX you may restrict the region where later keywords are applied. BOX can also be redefined by commands which use a box as part of their input data (e.g. ADD, COPY). A box defined as part of another command also remains active until disabled or redefined.

The data should be terminated by a slash (/).  

Item 1  IX1   First block on X axis of the new input box.

Item 2  IX2   Last block on X axis of the new input box.

Item 3  JY1   First block on Y axis of the new input box.

Item 4  JY2   Last block on Y axis of the new input box.

Item 5  KZ1   First block on Z axis of the new input box.

Item 6  KZ2   Last block on Z axis of the new input box.

The data must satisfy

        1 <= IX1 <= IX2 <= NDIVIX
        1 <= JY1 <= JY2 <= NDIVIY
        1 <= KZ1 <= KZ2 <= NDIVI

N.B. BOX/ENDBOX inside the  should not be placed inside the '--<RELATIONS+>/--<RELATIONS->' delimitors that WizGen uses to define its sections.   The user must move the keyword(s) outside of the delimitors. WizGen will not output them inside these sections, but they may be added in the WizGen Project file editor, but at any other position in the file.  This does not place any restriction on functionality.

COORD

Co-ordinate lines

The data associated with this Eclipse keyword must be included in the TransGen run file to define the map position of the cell corners. Each co-ordinate line defines the positions for the grid block corner points for each (i, j) in the grid.  Given the depth of a particular grid block corner, and the associated co-ordinate line, the x and y co-ordinates of the corner point can be calculated.

A co-ordinate line is specified by two triplets of x, y and z co-ordinates, representing two distinct points on it.  If the (x,y) co-ordinates of the top and bottom points are identical, then the z co-ordinates of the points are not used.

UNITS: m (METRIC), ft (FIELD), cm (LAB)

DEFAULT:  <undefined>

The keyword line is followed by (NDIVIX+1)(NDIVIY+1) co-ordinate lines, each consisting of two point each consisting of 3 values: the x, y and z co-ordinates.  The current version of the program limits the number of reservoirs (or sets of co-ordinate lines) in the grid to one.  The last co-ordinate line is followed by a slash (/).

For example:-

COORD
   1000    2000    1000    3000    1000    1000
   1000    0           1000    1000     0          1000
   2000    0           1000    2000     0          1000
   3000    0           1000    3000     0          1000
   0           2000    1000    0            2000   1000
   1000    2000    1000    1000     2000   1000
   2000    2000    1000    2000     2000   1000
   3000    2000    1000    3000     2000   1000
   0           4000    1000    0           4000    1000
   1000    4000    1000    1000    4000    1000
   2000    4000    1000    2000    4000    1000
   3000    4000    1000    3000    4000    1000
   0           0           2000    0           0           2000
   1000    0           2000    1000    0           2000
   2000    0           2000    2000    0           2000
   3000    0           2000    3000    0           2000
   0           500      2000     0          500      2000
   1000    500      2000     1000   500      2000
   2000    500      2000     2000   500      2000
   3000    500      2000     3000   500      2000
   0          1000    2000      0         1000    2000
   1000   1000    2000     1000   1000    2000
   2000   1000    2000     2000   1000    2000
   3000   1000    2000     3000   1000    2000     /

COPY

Copies data from one array to another

This is one of the Eclipse keywords which can be used to modify data arrays in the Eclipse simulation model (see also ADD, EQUALS and MULTIPLY). If the COPY keyword is included in the Eclipse parent model, it should also be included in the TransGen run file to ensure data compatibility between the two models.
The COPY keyword may be followed by any number of records, each of which is terminated by a slash (/).  The data is terminated by a null record (i.e. a record with no data before the terminating slash).  Each record consists of at least 2, and up to 8 items of data.

Item 1  The name of the source array from which data is to be copied (enclosed in quotes).

Item 2  The name of the destination array (enclosed in quotes).  Often the destination array is altered.

Items 3-8 may be used to redefine the input box for this and subsequent operations within the current keyword. If items 3-8 are not defined (a slash is inserted after item 2), they default to the values which were used for the previous operation within the current keyword.  For the first operation in the keyword, the box defaults to the values set by the most recent BOX or ENDBOX keyword.  If there is no preceding BOX or ENDBOX in the current section, the box is taken to include the entire reservoir.

Item 3  IX1   First block to be modified on the X axis.

Item 4  IX2   Last block to be modified on the X axis.

Item 5  JY1   First block to be modified on the Y axis.

Item 6  JY2   Last block to be modified on the Y axis.

Item 7  KZ1   First block to be modified on the Z axis.

Item 8  KZ2   Last block to be modified on the Z axis.

The data must satisfy

        1 <= IX1 <= IX2 <= NDIVIX
        1 <= JY1 <= JY2 <= NDIVIY
        1 <= KZ1 <= KZ2 <= NDIVIZ

DIMENS

Specify the model dimensions

The data associated with this Eclipse keyword must be included in the TransGen run file to define the number of columns, rows and layers in the model. The keyword is followed by three integers which define the number of blocks along the X, Y and Z axes respectively. The data must be terminated by a slash (/).

NOTE:- The DIMENS keyword and the relevant associated data are automatically added to the <project_name>.TGDATA run file when the Columns, Rows and Layers in the current project are defined and saved via the Coordinate System page in WizGen in either Basic or Flexible Project mode.

DIMENS
IX  IY  IZ   /

Item 1  Number of grid blocks in the x direction
Item 2  Number of grid blocks in the y direction
Item 3  Number of grid blocks in the z or depth direction


DIMENS MUST be specified at the top of the TransGen run file.  Geometry and property keywords depend on the dimensions of model being known, and an error will be generated if the DIMENS (or SPECGRID) keyword is missing.

EDITNNC

Change a non-neighbour connection

The data associated with this EDITNNC Eclipse keyword can be output from TransGen providing a list of transmissibility multipliers for all faulted non-neighbour connections.  The faulted non-neighbour connections are those of the parent model (i.e. the model defined by the ZCORN geometry which is input to both Eclipse and TransGen). Data associated with this keyword are output to the file specified for EDITNNC on the Output - simulator input page in WizGen.  

The form of the keyword output by TransGen is as follows:-

EDITNNC
IX  IY  IZ  JX  JY  JZ  TRANM /
.
.
/

Each line following the EDITNNC keyword specifies a faulted non-neighbour connection and the associated transmissibility multiplier and is terminated with a slash (/).  After the last non-neighbour modification a single slash (/) terminates the list.

The arguments in each line are:

IX, IY, IZ  The co-ordinates of the first cell joined to the non-neighbour connection.

JX, JY, JZ  The co-ordinates of the second cell joined to the non-neighbour connection.

TRANM   The transmissibility multiplier for the non-neighbour connection.  The multiplier TRANM cannot be negative but can be zero.

The EDITNNC data together with the TRANX, TRANY data (and NNC data- if the new fault drag and hierarchical zone effects were incorporated in the TransGen model) output from a TransGen run should be included in the EDIT section of the Eclipse input file.

NOTE:- Prior to the TransGen version 3.2 release, there was a one-to-one correspondence between across- fault connections in the Eclipse simulator and in TransGen. Provided, as was usually the case, the fault rock permeability was lower than the harmonic average permeability of the two grid-blocks, the transmissibility multipliers output to the EDITNNC file were in the range 0.0 to 1.0.
However the current version of TransGen including fault drag and hierarchical zone effects can produce radically different sets of across- fault connections to those in the parent model, and this can result in:-

• Many Transmissibility multiplier values of 0.0 in the EDITNNC file. These are for connections that exist in the parent model, but are not formed following the user-defined geometrical modifications to the fault trace. A value of zero in the file is necessary for these connections; if they were omitted from the file, the simulator would use the default unfaulted transmissibility for these connections.


• Many Transmissibility multiplier values in the EDITNNC file considerably larger than 1.0. These values do not reflect calculation of very high fault rock permeability, but are generally generated for connections with much larger juxtaposition areas in the user-defined model than in the parent model.

END

Marks the end of the input

This Eclipse keyword has no data field.  It marks the end of the input read from the instruction file. When TransGen reads the END keyword it finishes reading and begins computation, even if END does not occur at the end of a file.

ENDBOX

Reset box to encompass the entire grid

The Eclipse ENDBOX keyword has no associated data. It causes the input box to be reset so that it encompasses the entire grid.  Thus, for an 11 by 19 by 4 grid, it has the same effect as:

BOX
1   11   1   19   1   4   /

EQUALS

Set array to a constant in current box

This is one of the Eclipse keywords which can be used to modify data arrays in the Eclipse simulation model (see also ADD, COPY and MULTIPLY). If the EQUALS keyword is included in the Eclipse parent model, it should also be included in the TransGen run file to ensure data compatibility between the two models.
The EQUALS keyword may be followed by any number of records, each of which is terminated by a slash (/).  The data is terminated by a null record (i.e. a record with no data before the terminating slash).  Each record consists of at least 2, and up to 8 items of data.

Item 1  The name of the array to be modified (enclosed in quotes).

Item 2  The constant to be added to the array specified by item 1.  The constant should be positive, and may be real or integer.

Items 3-8 may be used to redefine the input box for this and subsequent operations within the current keyword. If items 3-8 are not defined (a slash is inserted after item 2), they default to the values which were used for the previous operation within the current keyword.  For the first operation in the keyword, the box defaults to the values set by the most recent BOX or ENDBOX keyword.  If there is no preceding BOX or ENDBOX in the current section, the box is taken to include the entire reservoir.

Item 3  IX1   First block to be modified on the X axis.

Item 4  IX2   Last block to be modified on the X axis.

Item 5  JY1   First block to be modified on the Y axis.

Item 6  JY2   Last block to be modified on the Y axis.

Item 7  KZ1   First block to be modified on the Z axis.

Item 8  KZ2   Last block to be modified on the Z axis.

The data must satisfy

        1 <= IX1 <= IX2 <= NDIVIX
        1 <= JY1 <= JY2 <= NDIVIY
        1 <= KZ1 <= KZ2 <= NDIVIZ

FAULTS

Specify fault positions

This Eclipse keyword can be used to specify fault positions during input or to report fault positions in an output file. By default, TransGen locates faults in the model by searching for fault displacements implicit in the corner-point geometry (keywords COORD and ZCORN).

In previous versions of TransGen, the FAULTS keyword prompted ViewGen to import user-defined cell faces and to mark these faces as faulted within the model. If the FAULTS keyword was specified in the TGDATA file ViewGen did not attempt to automatically search for faults. Calculations were based solely on the imported faults.

New to TransGen 3.1, faults are always generated automatically by ViewGen. The FAULTS keyword can still be used to import user-defined faults for visualisation, but only those user-defined faults that coincide with faults found by ViewGen can be viewed.

There is no limit to number of fault data records, each record defining a segment of a fault, as follows:

Item 1  Fault Name    (Up to 8 characters, enclosed in inverted commas).

Item 2  IX1   Lower I-coordinate of cells along the fault.

Item 3  IX2   Upper I-coordinate of cells along the fault.

    IX1 must equal IX2 if the face (Item 8) is 'X' or 'I'.

Item 4  IY1   Lower J-coordinate of cells along the fault.

Item 5  IY2   Upper J-coordinate of cells along the fault.

    IY1 must equal IY2 if the face (Item 8) is 'Y' or 'J'.

Item 6  IZ1   Lower K-coordinate of cells along the fault.

Item 7  IZ2   Upper K-coordinate of cells along the fault.

    IZ1 must equal IZ2 if the face (Item 8) is 'Z' or 'K'.

Item 8  Face of the fault.

    This should be one of 'X', 'Y', 'Z', 'I', 'J', 'K', 'X-', 'Y-', 'Z-', 'I-', 'J-', 'K-'.

The set of records must end with a blank record, containing only a slash (/).

HINT:- The FAULTS (derived connection property) data can be output to file at the end of a TransGen run via the Output - derived and user-defined properties page of WizGen.

FIELD

Indicates that field units are to be used

i.e. all units of length are reported in feet.

This Eclipse keyword has no associated data.

The units of measurement (i.e. Metric, Field or Lab) in a TransGen project are set via the Units option on the Coordinate System page of WizGen.

INCLUDE

Name of data file to be included

The name of the data file to be included at the current position, enclosed in quotes, should be inserted on the line after the keyword.  The data should be terminated by a slash (/).

The data file may contain any valid TransGen keyword, and can INCLUDE other data files (such as cell or connection properties associated with User-defined keywords as in a Flexible WizGen project).  INCLUDE is normally used to limit the size of the main instruction file.

The format of the INCLUDE keyword in the TGDATA run file is determined by the Eclipse data files selected via the Included Data page of WizGen.
So, for example, the INCLUDE entries in the TGDATA file could be as follows:-

--<INCLUDES+>
INCLUDE
`/home/aeh/TGproject/<project_name>_INPUT/COORD.DATA' /
INCLUDE
`/home/aeh/TGproject/<project_name>_INPUT/ZCORN.DATA' /
INCLUDE
`/home/aeh/TGproject/<project_name>_INPUT/TGVS.DATA' /
INCLUDE
`/home/aeh/TGproject/<project_name>_INPUT/PERM.DATA' /
--<INCLUDES->

KRNUM

Directional keywords indexing each cell to relative permeability & capillary pressure tables

These six Eclipse KRNUM directional keywords (KRNUMX, KRNUMX- KRNUMY, KRNUMY-, KRNUMZ KRNUMZ-) and the indices defining which relative permeability and capillary pressure table to apply to each grid-block in the simulation model, are created at the end of the ViewGen run according to the divisions and priorities assigned on the Two phase flow - Groupings section in WizGen and output  to a file as specified in the Two phase flow - Output section in WizGen.

NOTE:- To implement the use of Two-phase fault rock properties in the Eclipse simulation model, a file containing the KRNUM keywords together the upscaled relative permeability and capillary pressure tables output (SWOF output file) from the Two-Phase Flow module in TransGen need to be included in the simulator.

The KRNUM data file replaces the original SATNUM data. For example, KRNUMY is used to define the SWOF table for flow from cell [I J K] to cell [I J+1 K], while KRNUMY- defines the table for flow from cell [I J K] to cell [I J-1 K]. For all unfaulted cell faces and for all cells in the two vertical directions (i.e. KRNUMZ and KRNUMZ-) the original SATNUM index is used.

In the 3*3*2 example illustrated above, 16 cell edges are faulted. Assuming the cells originally have the same relative permeability functions, then the original SATNUM file will be:-

SATNUM
18*1
/

If a separate pseudo-relative permeability function is calculated for each faulted cell face, then the new KRNUM file will be:-

KRNUMY
1 1 1 1 2 1 1 3 1
1 1 1 1 4 1 1 5 1
/
KRNUMY-
1 1 1 1 1 6 1 1 7
1 1 1 1 1 8 1 1 9
/
KRNUMY
10 11 1 1 1 1 1 1 1
12 13 1 1 1 1 1 1 1
/
KRNUMY-
1 1 1 14 15 1 1 1 1
1 1 1 16 17 1 1 1 1
/
KRNUMZ
18*1
/
KRNUMZ-
18*1
/

LAB

Indicate that lab units are to be used

i.e. all units of length are reported in centimetres

The Eclipse keyword has no associated data.

The units of measurement (i.e. Metric, Field or Lab) in a TransGen project are set via the Units option on the Coordinate System page of WizGen.

METRIC

Indicate that metric units are to be used

i.e. all units of length are reported in metres

The Eclipse keyword has no associated data.

The units of measurement (i.e. Metric, Field or Lab) in a TransGen project are set via the Units option on the Coordinate System page of WizGen.

MINPV

Sets a minimum pore volume any cell must have to be active

The keyword is used in the Eclipse model to declare a threshold pore volume which any cell must exceed or it will be made inactive. The MINPV keyword sets a threshold value for all the cells in the current model. An inactive cell does not contribute to the total volume of the system and is treated by default as a barrier. This is one of the Eclipse keywords which can be used in the Eclipse simulation model to control which cells are active/inactive (see also ACTNUM, MINPVV and PORO).  

If the keyword is not included in the Eclipse run file, the minimum pore volume automatically defaults to1.0E-6 (in the current units).

The keyword should be followed by a line containing the threshold pore volume, in the current units,  and terminated by a slash (/).  The value for a threshold pore volume should be a positive real number.

NOTE:- New to TransGen version 3.2, the Cell pore volume cut-off (set on the Miscellaneous Options page of WizGen) defines the MINPV keyword in the TransGen run file and can either be kept at the default setting of 1.0E-6 or changed to mirror a different MINPV value included in the Eclipse simulator run file.

MINPV
1.0e-06 /

Units: cubic metres (METRIC),  rb (FIELD) or cubic centimetres (LAB).

HINT:- MINPV only affects active cells, those which have been rendered inactive via the ACTNUM keyword will remain so, even if their pore volume exceeds the threshold set by MINPV.

MINPVV

(Advanced keyword)

Sets the minimum pore volume a number of cells must have to be active

The MINPVV keyword is used in the Eclipse model to declare a threshold pore volume which a cell must exceed or it will be made inactive.  One threshold value is input for each cell in the current BOX.  If the keyword is not included, a default value of 1.0 E-6 is set and is applied to all the cells.  An inactive cell does not contribute to the total volume of the system and is treated by default as a barrier. This is one of the Eclipse keywords which can be used in the Eclipse simulation model to control which cells are active/inactive (see also ACTNUM, MINPV and PORO).  

NOTE:- If the MINPVV keyword is included in the Eclipse parent model, it should be included in the TransGen run file to ensure TransGen identifies the same active/inactive cells.

The MINPVV keyword should be followed by a line containing one positive real number for every cell in the current box, specifying its threshold pressure in the current units,  and terminated by a slash (/).

For example:-

-----  IX1 - IX2  IY1 - IY2  IZ1 - IZ2

BOX

                1         8      2         2      2        3  /


MINPVV
1000.0 1000.0 1000.0 1000.0 1000.0 1000.0 1000.0 1000.0
   500.0   500.0   500.0   500.0   500.0   500.0   500.0   500.0  /

ENDBOX

Units: cubic metres (METRIC),  rb (FIELD) or cubic centimetres (LAB).

MINPVV only affects active cells, those which have been set inactive via the ACTNUM keyword will remain so even if their pore volume exceeds the threshold set by MINPVV.

MULTIPLY

Multiply array by a constant in current box

This is one of the Eclipse keywords which can be used to modify data arrays in the Eclipse simulation model (see also ADD, COPY and EQUALS). If the MULTIPLY keyword is included in the Eclipse parent model, it should also be included in the TransGen run file to ensure data compatibility between the two models.
The MULTIPLY keyword may be followed by any number of records, each of which is terminated by a slash (/).  The data is terminated by a null record (i.e. a record with no data before the terminating slash).  Each record consists of at least 2, and up to 8 items of data.

Item 1  The name of the array to be modified (enclosed in quotes).

Item 2  The constant by which the array specified in item 1 is to be multiplied.  The constant should not be negative, but may be real or integer.

Items 3-8 may be used to redefine the input box for this and subsequent operations within the current keyword. If items 3-8 are not defined (a slash is inserted after item 2), they default to the values which were used for the previous operation within the current keyword.  For the first operation in the keyword, the box defaults to the values set by the most recent BOX or ENDBOX keyword.  If there is no preceding BOX or ENDBOX in the current section, the box is taken to include the entire reservoir.

Item 3  IX1   First block to be modified on the X axis.

Item 4  IX2   Last block to be modified on the X axis.

Item 5  JY1   First block to be modified on the Y axis.

Item 6  JY2   Last block to be modified on the Y axis.

Item 7  KZ1   First block to be modified on the Z axis.

Item 8  KZ2   Last block to be modified on the Z axis.

The data must satisfy

        1 <= IX1 <= IX2 <= NDIVIX
        1 <= JY1 <= JY2 <= NDIVIY
        1 <= KZ1 <= KZ2 <= NDIVIZ

NOTE:- If this keyword is present in the Eclipse parent model run file, it should be included in the TransGen run file to ensure TransGen uses the same data as used in the parent model.

MULTX

X direction transmissibility multipliers

MULTX, MULTY and MULTZ are grid-block-based multipliers used by Eclipse. They have the same format as any other grid-block property (e.g. PERMX) and can be included in a TransGen run in exactly the same fashion.

NOTE:- If MULTX values from Eclipse are included in TransGen, they will influence the contents of the TransGen TRANX output file as the transmissibilities will be calculated as a function of both MULTX and fault rock.

A MULTX value assigned to a cell (I, J, K) is applied to the transmissibilities between this cell and all cells with which it forms connections in the X+ direction, i.e. those with the indices (I+1, J, K*), where K* can take any value. Hence, if no fault is present on the X+ side of cell (I, J, K), the multiplier is only applied to the single connection into cell (I+1, J, K), but if a fault is present, MULTX is applied to any connection (neighbour or non-neighbour) formed from (I,J,K) in this direction.

The keyword should be followed by one non-negative real number for every grid block in the current input box specifying the X direction transmissibility multipliers.  The data must be terminated by a slash (/).

UNITS:  mD (METRIC, FIELD or LAB)

Every MULTX value in the top plane (k = 1) must be specified in one way or another by the end of the instruction file.  Values in lower planes (K > 1) which are not specified, default to the plane above.

Grid blocks are ordered with the X axis index cycling fastest, followed by the Y and Z axis indices.  Repeat counts may be used for repeated values (e.g. 115*208.4).  Note that spaces may not be inserted on either side of the asterisk.

MULTY

Y direction transmissibility multipliers

MULTX, MULTY and MULTZ are grid-block-based multipliers used by Eclipse. They have the same format as any other grid-block property (e.g. PERMX) and can be included in a TransGen run in exactly the same fashion.

NOTE:- If MULTY values from Eclipse are included in TransGen, they will influence the contents of the TransGen TRANY output file as the transmissibilities will be calculated as a function of both MULTY and fault rock.

A MULTY value assigned to a cell (I, J, K) is applied to the transmissibilities between this cell and all cells with which it forms connections in the Y+ direction, i.e. those with the indices (I, J+1, K*), where K* can take any value. Hence, if no fault is present on the Y+ side of cell (I, J, K), the multiplier is only applied to the single connection into cell (I, J+1, K), but if a fault is present, MULTY is applied to any connection (neighbour or non-neighbour) formed from (I,J,K) in this direction.

The keyword should be followed by one non-negative real number for every grid block in the current input box specifying the Y direction transmissibility multipliers.  The data must be terminated by a slash (/).

UNITS:  mD (METRIC, FIELD or LAB)

Every MULTY value in the top plane (k = 1) must be specified in one way or another by the end of the instruction file.  Values in lower planes (K > 1) which are not specified default to the plane above.

Grid blocks are ordered with the X axis index cycling fastest, followed by the Y and Z axis indices.  Repeat counts may be used for repeated values (e.g. 115*208.4).  Note that spaces may not be inserted on either side of the asterisk.

MULTZ

Z direction transmissibility multipliers

MULTX, MULTY and MULTZ are grid-block-based multipliers used by Eclipse. They have the same format as any other grid-block property (e.g. PERMX) and can be included in a TransGen run in exactly the same fashion.

MULTZ is analogous to MULTX and MULTY, and apply to connections in the Z+ direction.

The keyword should be followed by one non-negative real number for every grid block in the current input box specifying the Z direction transmissibility multipliers.  The data must be terminated by a slash (/).

UNITS:  mD (METRIC, FIELD or LAB)

Every MULTZ value in the top plane (k = 1) must be specified in one way or another by the end of the instruction file.  Values in lower planes (K > 1) which are not specified default to the plane above.

Grid blocks are ordered with the X axis index cycling fastest, followed by the Y and Z axis indices.  Repeat counts may be used for repeated values (e.g. 115*208.4).  Note that spaces may not be inserted on either side of the asterisk.

NNC

Explicit entry of non-neighbour connections

The output of data associated with the NNC Eclipse keyword is new to the TransGen version 3.2 release, allowing the output of transmissibility data for non-neighbour connections that do not exist in the geometry of the parent model. These NNC connections are generated when using WizGen in Flexible project mode with the new fault drag and hierarchical zone features included in the run file. Data associated with this keyword are output to a file specified for NCC on the Output - simulator input page in WizGen.

The form of the keyword output is as follows:-

NNC
  IX  IY  IZ  JX  JY  JZ  TRAN /
  .
  .
/


Where [IX, IY, IZ] [JX  JY  JZ] is a non-neighbour connection as defined by the coordinates (X, Y, Z) of the two cells (I and J) joined at the non-neighbour connection that does not exist in the geometry of the parent model and TRAN is transmissibility of the non-neighbour connection.

These non-neighbour connections are common on traces modified using the new functionality.
The output file containing the NNC data together with the EDITNNC, TRANX and TRANY files can then included in the EDIT section of the Eclipse input file to allow the implicit representation of more complex geometry than is represented explicitly in the simulation model.

NOTE:- Two forms of NNC connections are produced by the new functionality:-

• across- fault connections in which JX = IX+1 and IY = JY, or in which IX = JX and JY = IY+1. These connections are generated either because of throw changes on the trace or because of the inclusion of a fault zone


• same-stack connections in which IX = JX and IY = JY. These connections can only be generated by fault zones (See Processing modified traces).

NTG

Net to Gross Ratio

The data associated with this Eclipse keyword are needed in TransGen to define the shale content of the cells. Net to Gross is a measure of the net thickness of good reservoir (i.e. sand) to gross interval thickness with values ranging from 1 for best sands to 0 for non-reservoir (i.e. 100% shale). The keyword is followed by one non-negative real number between 1.0 and 0.0 for every grid block in the current input box.  The values specified are used to convert from gross to net thickness, and act as multipliers of grid block pore volumes and transmissibilities in the X and Y directions.  The data must be terminated by a slash (/).

Any NTG values which are not specified when the end of the instruction file is reached default to 1.0.

Grid blocks whose pore volume is zero are treated by TransGen as inactive.  Inactive blocks can be unambiguously identified by setting either PORO or NTG to zero in inactive blocks.  It is also possible to use the ACTNUM keyword.

Grid blocks are ordered with the X axis index cycling fastest, followed by the Y and Z axis indices.  Repeat counts may be used for repeated values (e.g. 115*0).  Note that spaces may not be inserted on either side of the asterisk.


For example:-

-------- IX1-IX2 JY1-JY2 KZ1-KZ2

BOX

6   11   4   9   2   3    /

NTG
   6*0.4   6*0.4   6*0.48   6*0.51   6*0.7   6*0.72
   6*0.4   6*0.4   6*0.48   6*0.51   6*0.7   6*0.72   /

PERMX

X direction absolute permeabilities

The data associated with this (and the PERMY) Eclipse keyword are needed in TransGen to define the cell permeabilities. The keyword is followed by one non-negative real number for every grid block in the current input box specifying the X direction absolute permeability.  The data must be terminated by a slash (/).

UNITS:  mD (METRIC, FIELD or LAB)

Every PERMX value in the top plane (k = 1) must be specified in one way or another by the end of the instruction file.  Values in lower planes (K > 1) which are not specified default to the plane above.

Grid blocks are ordered with the X axis index cycling fastest, followed by the Y and Z axis indices.  Repeat counts may be used for repeated values (e.g. 115*208.4).  Note that spaces may not be inserted on either side of the asterisk.

For example:-

-------- IX1-IX2 JY1-JY2 KZ1-KZ2

BOX

5   16   3   8   1   1   /

PERMX  
  100   1500   10*60  
    90   1500    10*60
    80   1500    10*65
    70   1500    10*70
    60   1500    10*75
    50   1500    10*75   /

PERMY

Y direction absolute permeabilities

The data associated with this (and the PERMX) Eclipse keyword are needed in TransGen to define the cell permeabilities. The keyword is followed by one non-negative real number for every grid block in the current input box specifying the Y direction absolute permeability.  The data must be terminated by a slash (/).

UNITS:  mD (METRIC, FIELD or LAB)

Every PERMY value in the top plane (k = 1) must be specified in one way or another by the end of the instruction file.  Values in lower planes (K > 1) which are not specified default to the plane above.

Grid blocks are ordered with the X axis index cycling fastest, followed by the Y and Z axis indices.  Repeat counts may be used for repeated values (e.g. 115*208.4).  Note that spaces may not be inserted on either side of the asterisk.

For example:-

-------- IX1-IX2 JY1-JY2 KZ1-KZ2

BOX

5   16   3   8   1   1   /

PERMY  
   100   1500   10*60  
     90    1500   10*60
     80    1500   10*65
     70    1500   10*70
     60    1500   10*75
     50    1500   10*75   /

PERMZ

Z direction absolute permeabilities

The data associated with this Eclipse keyword can be included in TransGen. The keyword is be followed by one non-negative real number for every grid block in the current input box specifying the Z direction absolute permeability.  The data must be terminated by a slash (/).

UNITS:  mD (METRIC, FIELD or LAB)

Every PERMZ value in the top plane (k = 1) must be specified in one way or another by the end of the instruction file.  Values in lower planes (K > 1) which are not specified default to the plane above.

Grid blocks are ordered with the X axis index cycling fastest, followed by the Y and Z axis indices.  Repeat counts may be used for repeated values (e.g. 115*208.4).  Note that spaces may not be inserted on either side of the asterisk.

PORO

Grid block porosities

This Eclipse keyword when included in the Eclipse simulation model defines the porosities of every cell in the current input box. It is also one of the Eclipse keywords used in the Eclipse simulation model to control which cells are active/inactive (see also ACTNUM, MINPV and MINPVV), i.e. the cells whose pore volume is zero are treated by Eclipse as inactive. If data associated with the PORO keyword are included in the Eclipse parent model, they should also be included in the TransGen run file.
The PORO keyword should be followed by one non-negative real number for every grid block in the current input box specifying the grid block porosities. The data must be terminated by a slash (/).


Grid blocks are ordered with the X axis index cycling fastest, followed by the Y and Z axis indices.  Repeat counts may be used for repeated values (e.g. 115*0.217).  Note that spaces may not be inserted on either side of the asterisk.

For example:-

-------- IX1-IX2 JY1-JY2 KZ1-KZ2

BOX

5   16   3   8   2   3   /

PORO
  0.16   0.14   0.12   0.1   0.08   0.09   0.1   0.11   0.12   0.13   0.14   0.14
  0.16   0.14   0.12   0.1   0.09   0.09   0.1   0.11   0.12   0.13   0.14   0.14
  0.15   0.14   0.12   0.1   0.10   0.09   0.1   0.11   0.12   0.13   0.14   0.14
  0.15   0.14   0.12   0.1   0.10   0.09   0.1   0.11   0.12   0.13   0.14   0.14
  0.14   0.14   0.12   0.1   0.09   0.09   0.1   0.11   0.12   0.13   0.14   0.14
  0.14   0.14   0.12   0.1   0.08   0.09   0.1   0.11   0.12   0.13   0.14   0.14
  0.18   0.15   0.12   0.1   0.08   0.09   0.1   0.11   0.12   0.13   0.15   0.15
  0.18   0.15   0.12   0.1   0.09   0.09   0.1   0.11   0.12   0.13   0.15   0.15
  0.15   0.15   0.12   0.1   0.10   0.09   0.1   0.11   0.12   0.13   0.15   0.15
  0.15   0.15   0.12   0.1   0.10   0.09   0.1   0.11   0.12   0.13   0.15   0.15
  0.15   0.15   0.12   0.1   0.09   0.09   0.1   0.11   0.12   0.13   0.15   0.15
  0.15   0.15   0.12   0.1   0.08   0.09   0.1   0.11   0.12   0.13   0.15   0.15   /

NOTE:- In the absence of PORO data, TransGen sets the porosities to 1.0.

RPTRST

Controls on output to the RESTART file

The Eclipse RPTRST should be followed by a list of mnemonics which control the output of data to the Restart file. The list should be terminated by a slash (/). One of the Eclipse output controlled by this keyword is FLOWS, i.e. the output of interblock flows (including non-neighbour connection flows and flows between global and local grids).

The Eclipse Restart file can be used in TransGen as a source of across-fault flow rates when including two-phase fault rock calculations in the run file (see Using WizGen in Flexible project mode)

SATNUM

Index Grid referencing the saturation function data

The Eclipse SATNUM keyword indexes each grid-block in the model to a set of relative permeability and capillary pressure curves as defined by the SWOF keyword. The keyword takes only integer values, but in all other respects is identical to other grid-block properties recognised by TransGen.

NOTE:- The data associated with the SWOF and SATNUM keywords derived from an Eclipse reservoir model MUST be input into TransGen in order to implement the Two-phase flow functionality.

Examples:-

SATNUM
2 2 2 2 2 2 2 1
1 1 1 6 6 6 6 6
...
1 1 1 1 1 1 1 1
2 2 2 2 2 2 1 1
1 6 6 6 6 6 6 6
/

or

SATNUM
128*1
256*2
/

The SATNUM keyword is essential for running the TransGen Two-phase fault-rock calculations and should be loaded via the Included Data page of WizGen in Flexible project mode.
SATNUM is a recognised TransGen keyword and therefore it does not need to be specified as a user-defined grid-block property.

SPECGRID

Specify the model dimensions, number of reservoirs and coordinate system

This Eclipse keyword is followed by four integers and a character;  which define the number of grid-blocks along the X, Y and Z axes respectively, the number of reservoirs, and a flag which indicates whether the Cylindrical or Cartesian coordinates are used.  Following the keyword, there is a single record containing four integers and a character, and the data must be terminated by a slash (/).

SPECGRID
10 20  12 1 F

Item 1  Number of grid blocks in the x direction - for Cartesian geometries
Item 2  Number of grid blocks in the y direction - for Cartesian geometries
Item 3  Number of grid blocks in the z or depth direction - for Cartesian geometries
Item 4 The number of reservoirs, each with its own coordinate system.  For TransGen this must only be one.
Item 5 A single T or F indicating Cylindrical or Cartesian coordinates.  TransGen only accepts Cartesian coordinates.

SPECGRID should be specified at the top of the instruction file.  Geometry and property keywords depend on the dimensions of model being known, and an error will be generated if a SPECGRID or DIMENS keyword is missing.  TransGen only accepts a single reservoir in a 8 cornerpoint cartesian geometry.

SWOF

Defines water & oil relative permeability & water-oil capillary pressure as a function of water saturation

The Eclipse SWOF keyword defines the water relative permeability, oil relative permeability and water-oil capillary pressures as a function of water saturation. The keyword should be followed by as many tables as the highest integer in the SATNUM grid, each terminated by a slash (/). The order of the tables correspond to the SATNUM indexing.

NOTE:- The data associated with the SWOF and SATNUM keywords derived from an Eclipse reservoir model MUST be input into TransGen in order to implement the Two-phase flow functionality.

Each table consists of 4 columns of data:-

Column 1. Water saturation

Values should be between 0 and 1 and should increase monotonically down the column. The first value in the column is interpreted as the connate water saturation.

Column 2. Water relative permeability

Values should be between 0 and 1 and should be level or increase monotonically down the column. The first value in the column must be zero.

Column 3. Oil relative permeability

Values should be between 0 and 1 and should be level or decrease down the column. The last value in the column must be zero.

Column 4. Capillary pressure

Values should be level or decrease down the column. Units are bars (METRIC), psi (FIELD) or atm (LAB).

Example:-

SWOF
0.3817 0.0 0.85 2.95990
0.5053 0.000001 0.8228 1.45830
0.5913 0.00001 0.4985 0.44050
0.6297 0.1131 0.0199 0.05020
0.6633 0.1907 0.0072 0.02620
0.7001 0.3001 0.0000 0.0000
/ end table 1
0.1458 0.0 0.85 2.96180
0.3308 0.000001 0.8496 1.53730
0.5 0.00001 0.4164 .33570
0.5423 0.008 0.3247 .04620
0.5588 0.0477 0.045 .02110
0.7 0.3 0.0000 0.0000
/ end table 2

The tables associated with the SWOF keyword are essential for running the TransGen Two-phase functionality and the file containing this data must be loaded via the Two phase flow - INPUT page of WizGen in Flexible project mode. In the current version of TransGen, all SWOF tables should be contained in the same file.

NOTE:- SWOF is the only saturation function keyword recognised and supported by the current version of TransGen. Other Eclipse keywords for defining two or three phase relative permeability or capillary pressure functions (e.g. SGFN, SWFN, SOF2, SOF3, SOF32D, SGOF, SLGOF) are not supported.

NOTE:- Endpoint scaling is not supported by the current version of TransGen.

TGAXES

(new TransGen keyword)

Define the ordering of the cells in the input files and direction of the Y axis for the COORD section

This TransGen keyword allows the import of data ordered in a different manner from the Eclipse default.  Data for TransGen must be ordered in row order  (columns cycle fastest),  but the origin (Row 1,  Column 1) may be placed in any corner of the model.  Similarly, the direction in which the x and y coordinates are ordered, defining the position of the COORD lines,  may be in local model coordinates (Y increases 'downwards') or in geographic coordinates (Y increases 'upwards' to the north).  The keyword should be followed by a record containing two integer numbers and terminated by a slash (/).  

For example: the eclipse default is:-

TGAXES
1  0  /

Item 1 IORDER specifies the position of the origin and the direction of the rows and columns
           IORDER =1 (Row 1,  Column 1)  in 'top left' .  Rows are east-west
                           This is a special case for the Eclipse default, the only permitted value ILOCAL is 0
           IORDER =2 (Row 1,  Column 1)  in bottom left.  Rows are east-west
           IORDER =3 (Row 1,  Column 1)  in top left.  Rows are east-west
                            IORDER=3 ILOCAL = 0 is also the Eclipse default.
                            IORDER=3, ILOCAL  =1 is a frequent RMS ordering
           IORDER = 4 (Row 1,  Column 1)  in bottom right  Rows are east-west
           IORDER = 5 (Row 1,  Column 1)  in top right  Rows are east-west

Item 2 ILOCALspecifies the direction in which Y increases.  It may have two values:
       0 - local coordinate system. Y increases 'downwards'
       1 - Geographic coordinate system. Y increases 'upwards' ie to the north in map view.

TGDRAG

(new TransGen version 3.2 keyword)

To manually define throw modifications

When using the new Include fault drag and hierarchical zone effects functionality, this TransGen keyword allows manual definition of throw modifications on all fault traces in the model (either included explicitly in the geometry of the parent model or included as user-defined faults using TGTHROW). The keyword is designed to allow a portion of the throw of a fault to be accommodated with local normal drag (the yellow areas in the cartoon below), but can also be used to include uncertainty in fault throws in the model.

NOTE:- Probabilistic and/or deterministic fault throw modification can also be made using the equations set via the Drag applied to fault traces page in the WizGen or by user-defined criteria specified in the DRAG plugin.

The keyword can be followed by as many definitions as needed on separate lines, each line and the file are terminated by a slash. The format of each line is:

I J `Direction' Drag_ratio /

Drag_ratio can take any positive or negative value (or zero) and defines the ratio between the throw of the trace after drag has been applied to the throw on the trace reflected in the original model geometry. Hence a value of 0.5 will halve the throw, a value of 0.0 will get rid of the fault throw altogether, a value of 1.0 will leave the throw unaltered, a value of 2.0 will double the throw and a value of  1.0 will switch the sense of throw of the fault. Based on the input values, revised drag ratios are calculated and applied on a layer-by- layer basis, using similar considerations to those discussed in relation to the TGTHROW keyword to ensure coherent geometries.

Example

The following TGDRAG definition applied to the parent model shown in Figure a, will result in the geometry shown in Figure b.

NOTE:- In this example an integer value is written for each line in the file to the right of the slash - anything to the right of a slash is ignored while the file is being read and these are only placed in the file for reference, to highlight portions of model in which the values of TGDRAG result in different behaviour  - these are highlighted in Fig. b. For the traces labelled 1, values of TGDRAG are in the range 0 to 1 and the fault throw is progressively lowered. For the traces labelled 2, Drag ratios are in the range 1.0 to 2.0, and the fault throws are increase by up to 100%. All traces associated with the splay fault labelled 3 have TGDRAG values set to -1.0 and the sense of throw of this fault is reversed. The traces labelled 4 have drag ratios around 1.0 adding fairly random variability to the fault throws.

Like the example discussed with reference to TGTHROW, the display shown in Figure b is not obtained by ViewGen and is shown for illustrative purposes only.

One or more files containing the TGDRAG keyword and others (e.g. TGFZONE, TGTHROW and/or TGTRACE) can be included in the TransGen run via the Included Data page of WizGen, along with all cell and (if used) user-defined connection properties, when running in Flexible project mode with the Include fault drag and hiererchical zone effects option selected.

NOTE:- If TGDRAG is used to modify the throw on user-defined fault traces, the keyword MUST be loaded into TransGen AFTER the TGTHROW definition.

HINT:- The TGDRAG data can be output to file at the end of a TransGen run via the Output - derived and user-defined properties page of WizGen.

TGEVS

This TransGen keyword has been made redundant in TransGen 3 and has been replaced by the effective Vshale settings in the new TGFSP keyword. The TGEVS keyword will be recognised by version 3, but TransGen stops with an error report.

Method of effective vshale calculation in TransGen version 3

In TransGen version 3, how the effective vshale property is calculated and used for a particular fault seal potential measure is set as part of the new TGFSP keyword (instead of the now defunct TGEVS keyword).
When using WizGen in "Basic project" mode, the effective vshale can be calculated from net-to-gross only, vshale only or a combination of both (depending on the data associated with the NTG and TGVS keywords included in the current TransGen run and the Effective vshale computation method set on the Fault Rock Properties page of WizGen).
When using WizGen in "Flexible project" mode, the effective vshale can be calculated similarly or based on a user property depending on the Effective vshale setting applied via the Fault Seal Potential Variables page of WizGen.

Net-to-gross only - TransGen considers the non-net region to be shale and takes the shale content to equal 1 minus the Net-to-Gross value (where Net-to-Gross is the ratio of the net thickness of good reservoir, i.e. sand to gross interval thickness).

                                                                              eVs = (1 - NTG)

Vshale only - TransGen takes Vshale as the shale content.

                                                                                eVs = TGVS

NGT and VShale - TransGen assumes the non-net region to be pure shale and takes the Effective Vshale content of a grid block to be:-

                                                                      eVs = (1 - NTG) + (TGVS x NTG)

User Property - Transgen takes the user-defined Effective Vshale grid values assigned to a user-defined keyword. These values should all ideally lie between 0.0 and 1.0. If the user-defined Effective Vshale grid contains values less than 0.0 or greater than 1.0, these values will still be included in the FSP calculation, but the results cannot be guaranteed meaningful. Meaningful FSP values are assumed to have a range from zero to infinity and all numbers entering the summation are either zero or positive.

TGFSP

(new TransGen version 3.2 keyword)

To allow the calculation of fault seal potential (FSP) measures

NOTE:- The now redundant TGSGRM and TGEVS keywords are replaced by settings in the TGFSP keyword.

The keyword is followed by up to 5 Fault Seal Potential (FSP) definitions each on separate lines, each terminated by a "/". The keyword is then terminated by another "/".
The keyword and associated FSP definitions are created automatically by settings saved on the Fault Rock Properties page of WizGen in "Basic project" mode or on the Fault Seal Potential Variables page of WizGen for a "Flexible project".

Each FSP definition consists of 12 settings separated by spaces, i.e:-

TGFSP
name  l  m  n  p  shale_switch  shale_string  distance_switch  combine_switch  evshale_switch  evshale_string  plunge_correction  /
/

where:-

name is the FSP variable name, which is case-sensitive (e.g. 'sgr').

exponent l is the power to which "throw" is raised.

exponent m is the power to which "thickness" is raised.

exponent n is the power to which "distance" is raised.

exponent p is the power to which "eVshale" is raised.

shale_switch  1 = named layers; 2 = eVshale cutoff; 3 = import from user-defined keyword; 4 = everything is shale.

shale_string if shale_switch = 1, shale_string = a list of layers (e.g. '3-5, 7, 9')

                               if shale_switch = 2, shale_string = a cutoff value (e.g. '0.7')
                               if shale_switch = 3, shale_string = a user-defined keyword (e.g. 'shale_1')
                               if shale_switch = 4, shale_string ie empty (e.g. ' ')

distance_switch  1 = centre of beds; 2 = far side of beds; 3 = near side of beds.

combine_switch  1 = max layers then sum; 2 = sum then max layers; 3 = max; 4 = min; 5 = mean; 6 = sum then average; 7 = sum in footwall; 8 = sum in hangingwall

evshale_switch 1 = ntg; 2 = tgvs; 3 = ntg and tgvs; 4 = import from user-defined keyword

evshale_string if evshale_switch = 1, 2 or 3, evshale_string is empty (e.g. ' ')

                                    if ev_switch + 4, evshale_string = a user-defined keyword name (e.g. 'evshale_1')

plunge_correction  0 = calculated on vertical projection; 1 = calculated in 3D

For example:-

TGFSP
`sgr'    -1 1 0 1 4 ' ' 1 6 1 ' ' 1 /
'csp_yielding'    0 2 -1 0 1 '8-13, 26-29, 34, 38, 45-47' 1 1 1 ' ' 1 /
'csp_fulljames'   0 2 -1 0 2 '0.5' 2 2 3 ' ' 1 /
/   

The name and the two strings must be included in single quotes ' '.  The contents of the strings depend on previously-defined switches.

If particular exponents are set to 0.0, the options for subsequent settings are irrelevant to the calculation. Nonetheless, valid options MUST be supplied. So, for example, if exponent n is set to 0.0 (as it would be in the SGR calculation), a distance_switch must still be defined. Similarly, if exponent p is set to 0.0, the effective Vshale of each cell is not included in the calculation, but still a valid evshale_switch must be applied. If not, ViewGen will issue an error message and stop.

TGFZONE

(new TransGen version 3.2 keyword)

To include user-defined fault zone(s) in a TransGen run

When using the new Include fault drag and hierarchical zone effects functionality, this TransGen keyword can be used to include large fault zones visible seismically, but too small for explicit representation in the parent flow model. The TGFZONE keyword can be followed by as many definitions as needed on separate lines, each line and the file are terminated by a slash. The format of each line is:-

I J `Direction' type length width r1 r2 r3 r4 damage_para damage_perp /

I and J are the indexes of the cell stacks to which the fault trace is adjacent.

Direction can be either `DIR-X' or `DIR-Y'. The single quotation marks are essential.

Type must be set 1 (there is no functionality associated with type in this version of TransGen).

Length and width are the components in the XY plane of the length of the fault zone component parallel and perpendicular to the trace. Units are meters (metric), feet (Field), cm (lab)

r1, r2, r3 and r4 are throw ratios assigned to the four corners of the ramp.

Damage_para and damage_perp are permeability multipliers used in the ramp.

These parameters are defined fully in the section on Fault zone properties.

For example:-

TGFZONE
--I J direction type length width r1 r2 r3 r4 kmult_para kmult_perp
9 3 'DIR_X' 1 50 20 1.0 1.0 0.0 0.0 1.0 1.0 /
12 15 'DIR_Y' 1 14 30 0.7 0.8 0.3 0.2 0.75 0.25 /
/

HINT:- The TGFZONE keyword should only be used to include large deterministic fault zones visible seismically, but too small for explicit representation in the flow model. Smaller, sub-seismic fault zone structure is better modelled stochastically using the WizGen tool (see using the Assign hierarchical zones option on the Hierarchical fault zone definition page in WizGen).

One or more files containing this new keyword and others (e.g. TGTHROW, TGDRAG and/or TGTRACE) can be included in a TransGen run via the Included Data page of WizGen, along with all cell and (if used) user-defined connection properties, when running a project in Flexible project mode with the Include fault drag and hiererchical zone effects option selected.

HINT:- The TGFZONE data can be output to file at the end of a TransGen run via the Output - derived and user-defined properties page of WizGen.

TGKDT

This TransGen keyword has been made redundant in TransGen 3 and has been replaced by the PERM plugin either created automatically by WizGen in "Basic project" mode (see Plugins generated by WizGen in Basic project mode) or by the user via the User-defined plugins page of WizGen (in "Flexible project" mode).  The keyword (which was used in TransGen 2 to include permeability modifier versus depth data) will be recognised by version 3, but TransGen stops with an error report.

TGKSE

This TransGen keyword has been made redundant in TransGen 3 and has been replaced by the PERM plugin either created automatically by WizGen in "Basic project" mode (see Plugins generated by WizGen in Basic project mode) or by the user via the User-defined plugins page of WizGen (in "Flexible project" mode).  The keyword (which was used in TransGen 2 to input the constants in the equation defining the relationship between shale gouge ratio and permeability) will be recognised by version 3, but TransGen stops with an error report.

TGKST

This TransGen keyword has been made redundant in TransGen 3 and has been replaced by the PERM plugin either created automatically by WizGen in "Basic project" mode (see Plugins generated by WizGen in Basic project mode) or by the user via the User-defined plugins page of WizGen (in "Flexible project" mode).  The keyword (which was used in TransGen 2 to input a shale gouge ratio vs permeability lookup table) will be recognised by version 3, but TransGen stops with an error report.

TGMETRIC

The TransGen TGMETRIC keyword (new to the version 3 release) acts as a binary switch in the TGDATA  file; it is either included, in which case it is applied (default) or excluded and therefore not applied. With the TGMETRIC keyword included in the TGDATA file, all calculated fault displacement, fault rock thickness and all Fault Seal Potential measures are reported (graphically and in output files) and stored internally within TransGen using metres as the unit of length.

When using WizGen in "Basic project" mode, the TGMETRIC keyword is ALWAYS included in the TGDATA files.

However, with WizGen in "Flexible project" mode, the user has the option to NOT include the TGMETRIC keyword by selecting the FSPs in native units option on the Miscellaneous Options page. With the TGMETRIC keyword missing from the TGDATA runfile, all fault displacement, fault rock thickness and all calculated FSP measures are reported in the units of length appropriate to the Units specified on Coordinate System page of WizGen, i.e. feet if the Units are FIELD or cm if the Units are LAB. Inclusion or exclusion of the TGMETRIC keywordmakes absolutely no difference if the Units in the TGDATA run file are METRIC.

Depth is not affected by the keyword which is always displayed and stored internally in the units of length appropriate to the TGDATA file.

The purpose of the TGMETRIC keyword is to ensure plugins are transportable between different UNITS conventions. Clay Smear Potential (CSP) for example has the dimensions of length. If the UNITS in TGDATA are set to FIELD, then CSP will be reported in metres if TGMETRIC is present and in feet if it is not. With the TGMETRIC keyword included (i.e. activated), this means a PERM plugin with a particular sealing threshold applied to CSP will not need to be changed if the same plugin is used in a reservoir model of another field which has been built using a different UNITS convention.

If TGMETRIC is not included (in Flexible project mode only), many plugins will be specific to the UNITS convention for which they have been written.

Inclusion/exclusion of the TGMETRIC keyword does not influence the transmissibility calculation: dimensions of fault thickness are always converted back to the specified units system before transmissibilities are calculated ensuring consistency of units in the TRANX and TRANY files.

TGMINTR

Sets a minimum unfaulted transmissibility any connection must have to be output

This TransGen keyword is provided for the user to declare a threshold unfaulted transmissibility value that any connection must exceed or it will be excluded from the output files.  The keyword together with the default value of 1.0 E-6 (i.e. the threshold Eclipse uses for this property) is automatically included in the TGDATA run file using the Unfaulted transmissibility cutoff setting on the Miscellaneous Options page of WizGen in either "Basic" or "Flexible project" mode. The default value should only be changed in exceptional circumstances and should never to set to a value greater than 1.0 E-6 if Eclipse input is being generated.

The keyword is followed by a line containing the threshold value, terminated by a slash (/) The value for a minimu transmissibility multiplier cutoff should be a positive real  number.

TGMINTR
1.0e-06  /

Units: dimensionless

HINT:- The TGMINTR setting only affects active connections, so for example those which have been set inactive via the TGREMNNC keyword will remain so even if their unfaulted transmissibilities exceeds the threshold set by TGMINTR.

TGNEWKEY

New keyword, added to TransGen version 3, to allow the inclusion of user-defined grid-block (i.e. cell) and connection properties. It is automatically added to the TDDATA run file when any new Cell property and/or Connection property keywords are defined via the User-defined keywords page when using WizGen in "Flexible Project" mode.

The TGNEWKEY keyword is followed by as many lines as there are new properties, terminated by a slash (/).

Each line has the form:

`name' property_type/

where property_type is 1 for grid-block (cell) properties and 2 for connection properties.

The name can be up to 32 characters long and may contain underscores(_) and numbers as well as letters. However, the first character of the name must be a letter. The letters can be either upper or lower case, but if used in plugins, the property name must be referenced exactly as input as the C++ macro language is case sensitive.

TGNEWKEY
`cell_property1' 1 /
`cell_property2' 1 /
`cell_property3' 1 /
`connection_property1' 2 /
`connection_property2' 2 /
/

User-defined keywords can be used in plugins, in the TGFSP Keyword and visualised in the graphics interface. If a user-defined keyword referenced via theTGFSP keyword or in a User-defined plugin is not defined using TGNEWKEY, the Transgen run will fail and issue an error message.

Unlike recognized grid-block properties (e.g. PERMX, ATCNUM), the values of user-defined grid-block properties do not need to be included explicitly in the TransGen run through instructions in the TGDATA file (although they can be). They may also be calculated in the CELLPROP plugin (see Example). If the values of the property are not included explicitly, they are assigned an initial value of 0.0.

TGNNC

This keyword has been made redundant in TransGen 3. It will be recognised by version 3, but TransGen stops with an error report.

You can create a similar user-defined non-neighbour connection property in TransGen version 3 when using WizGen in "Flexible project" mode by adding it as a User-defined keyword and including an appropriate data file via the Included Data page.

Input a user-defined non-neighbour connection property

This keyword can be used to input a user-defined property onto fault surfaces for visualisation purposes.  The form of the keyword is as follows:

TGNNC
IX  IY  IZ  JX  JY  JZ  User-Value /
.
.
/

Each line following the TGNNC keyword specifies a connection (neighbour or non-neighour and is terminated with a slash (/).  After the last non-neighbour connection a single slash (/) terminates the list.

The arguments in each line are:

IX, IY, IZ  The co-ordinates of the first cell joined to the non-neighbour connection.

JX, JY, JZ  The co-ordinates of the second cell joined to the non-neighbour connection.

User-Value  A real number representing the value of some user-defined property.

The user-defined property is not used in any computation.  It can be selected for visualisation from within the graphics viewer.

TGNOCALC

(new TransGen keyword)

Prevent calculation of fault transmissibility multipliers

This keyword does not take any data. If present, TransGen builds the Eclipse model geometry, reads property data, but does not perform any calculations. Inclusion of this keyword in the TGDATA run file allows the model to be visualised quickly.

This keyword is automatically included in the TGDATA run file when the Do not perform calculation option is toggled "on" in the Miscellaneous Options page of WizGen.

TGPLUGIN

New keyword added to TransGen version 3 allowing flexible calculations of effective Vshale, fault permeability, fault thickness and across-fault connection area for inclusion in transmissibility calculations.

The form of the TGPLUGIN keyword is a set of strings enclosed by single quotes (one string per line) that associate a plugin name with a full C++ source filename. The source file contains the plugin code. For example:-

TGPLUGIN
`CELLPROP=/home/aeh/TGproject/TGproject_INPUT/.plugins/cellproperties.cpp'
`THICK=/home/aeh/TGproject/TGproject_INPUT/.plugins/thickness.cpp'
`PERM=/home/aeh/TGproject/TGproject_INPUT/.plugins/permeability.cpp'
`AREA=/home/aeh/TGproject/TGproject_INPUT/.plugins/area.cpp'
/

A plugin is a macro, written in C++, that manipulates the values of cell and connection properties. There are four types of plugin which may be used during the TransGen calculations, two of which are essential, i.e. Fault thickness and fault permeability MUST be calculated via the THICK and PERM plugins. The CELLPROP and AREA plugins are not essential.

NOTE:- When using WizGen Light, the THICK and PERM plugins are automatically created and saved to the project's <project_name>_INPUT/.plugins directory as _AUTO_THICK_PLUGIN.cpp and _AUTO_PERM_PLUGIN.cpp respectively and added as strings below the TGPLUGIN keyword in the TGDATA run file when new Fault Rock Properties settings are saved. When using WizGen Heavy, all plugins have to be created via the User-defined plugins page of WizGen Heavy.

TGREMNNC

(new TransGen keyword)

Disable non-neighbour connections within the model

This TransGen keyword is supplied for compatibility with future releases of TransGen. However, due to small discrepancies in numerical accuracy between TransGen and Eclipse, TransGen occasionally specifies more non-neighbour connections than Eclipse finds. A warning detailing which NNCs are extra is generated during Eclipse input, and can be safely ignored. If desired, the TGREMNNC keyword can be used with the list given in the Eclipse warning to remove these NNCs from the TransGen output. The form of the keyword is as follows:

TGREMNNC
IX  IY  IZ  JX  JY  JZ   /
.
.
/

Each line following the TGREMNNC keyword specifies a non-neighbour connection and is terminated with a slash (/).  After the last non-neighbour connection a single slash (/) terminates the list.

The arguments in each line are:

IX, IY, IZ  The co-ordinates of the first cell joined to the non-neighbour connection.

JX, JY, JZ  The co-ordinates of the second cell joined to the non-neighbour connection.

TGRPT

(new TransGen keyword)

Controls on output from TransGen

This TransGen keyword is generated when output file(s) are specified for one or more properties on the Output - simulator input page of WizGen.This keyword is followed by several arguments that control what the program outputs and where the output is written. TransGen can output the following keywords (and associated data) for later inclusion in an Eclipse simulation: EDITNNC, TRANX, TRANY, NNC (or NNC MULT, TX REPL, TY REPL for inclusion in a MORE simulation). The parameter GRAPHICS will be specified with the Enable 3D graphics viewer after calculation has completed option toggled "on" (default setting).

The keyword and associated string(s) are automatically added to the TGDATA run file by setting and saving the Output - simulator input page appropriately in WizGen.

TGRPT
'EDITNNC=editnnc.data'
'TRANX=tranx.data'
'TRANY=trany.data'
'GRAPHICS'
/

The character string following the '=' sign is the name of the output file into which results will be written. The GRAPHICS parameter is a simple switch and does not require a file name. The parameters can be written in upper, lower or mixed case. The data must be terminated by a slash (/).

HINT:- Additional outputs of derived and/or user-defined properties are controlled via TGXRPT keyword.

TGSGRM

This TransGen keyword has been made redundant in TransGen 3 and has been replaced by an option in the TGFSP keyword. TGSGRM will be recognised by version 3, but TransGen stops with an error report.

Across-fault Shale Gouge Ratio combination method

This keyword specifies how SGR values calculated at identical positions but on different sides of a fault surface are combined to give the same SGR value to both vertices. The form of the keyword is as follows:

TGSGRM
method
/

where method is one of 'hangingwall', 'footwall' or 'average' (note: the parameter must be enclosed in quotes). If 'footwall' is specified the SGR value from the up-thrown side of the fault is used in both vertices, 'hangingwall' causes the down-thrown SGR value to be used. The default is 'average', which means the arithmetic average of the two SGR values is used in both vertices.

N.b. where footwall layers are eroded it is recommended that 'hangingwall' is used.

Different methods can be defined for different volumes of the reservoir using the BOX/ENDBOX keywords.

TGSHALE

(new TransGen version 3.1 keyword)

Shale Data

This TransGen keyword and the associated data only needs to be incorporated in the TGDATA run file (by adding the relevant file on the Included Data page of WizGen) if Clay Smear Potential (in a "Basic project") or any Fault Seal Potential measure (in a "Flexible project") are calculated using the TGSHALE keyword for Shale definition:-

see using the Defined shale cells option for Shale definition under CSP options when using either Only CSP or CSP then SGR to calculate fault permeability on the Fault Rock Properties page of WizGen in "Basic project" mode.

OR

see using Shale definition based on the TGSHALE keyword on the Fault Seal Potential Variables page of WizGen in "Flexible project" mode.

The data associated with this keyword determines if a cell should be regarded as shale or not. The keyword should be followed by one non-negative real number of either 1.0 or 0.0 for each cell in the current project where 1.0 represents a shale cell and 0.0 a non-shale cell. Only cells defined as shales are subsequently used to calculate the fault seal potential measure. The data must be terminated by a slash (/).

Grid blocks are ordered with the X axis index cycling fastest, followed by the Y and Z axis indices.  Repeat counts may be used for repeated values (e.g. 115*0).  Note that spaces may not be inserted on either side of the asterisk.

TGSTRLNE

(new TransGen development keyword)

Allow the display of streamlines from 3DSL

The TranGen keyword is followed by an integer specifying the number of streamlines followed by a slash (/), and block of data for each streamline.   The wells will be displayed when wells are selected.  This keyword is under development and the current limited capability does not reflect what will be available in the future.

Each streamline  data block consists of a header record containing an integer specifying the number of nodes in the streamline; a record for each node with X, Y, Z and time of flight, and a final data block terminator consisting of a slash (/). flowed --

TGSTRLNE
997 /
-- STREAMLINE=10 NNODES=282 N_SOURCES/SINKS=2 SLTYPE=1
-- NODE=2 WELL=WELL1  LAYER=11
-- NODE=282 WELL=WELL2  LAYER=1
-- X-COORD      Y-COORD      Z-COORD (m)  TOF(days)    BLOCKINDEX
282 /
0.2450000E+04 0.1850000E+04 0.3671200E+04 0.0000000E+00
0.2475225E+04 0.1900000E+04 0.3676210E+04 0.1157407E-04
0.2500000E+04 0.1983561E+04 0.3684583E+04 0.2664420E+03
0.2519080E+04 0.1999988E+04 0.3691049E+04 0.4226107E+03
0.2561148E+04 0.2100012E+04 0.3694734E+04 0.1388542E+04
0.2570230E+04 0.2200023E+04 0.3700764E+04 0.2385265E+04
...

272 records omitted
...
0.3497699E+04 0.3900000E+04 0.3331850E+04 0.2136167E+07
0.3399813E+04 0.3805615E+04 0.3345406E+04 0.2138803E+07
0.3300188E+04 0.3815180E+04 0.3354544E+04 0.2139157E+07
0.3250000E+04 0.3850000E+04 0.3356975E+04 0.2139157E+07
/

TGTDE

This keyword has been made redundant in TransGen 3 and has been replaced by the THICK plugin either automatically by WizGen in "Basic Project" mode (see Plugins generated by WizGen in Basic project mode) or created by the user via the User-defined plugins page of WizGen (in "Flexible Project" mode).  The TGTDE keyword will be recognised by version 3, but TransGen stops with an error report.

Co-efficients of equation relating displacement to thickness

The equation used to calculate fault thickness as a function of fault displacement is:

where THICKf and Df are the thickness and displacement at a particular connection vertex.  The constants a and b are the two data items in the TGTDE keyword.  The data must be terminated with a slash (/). If TGTDE is not specified in the run file, the constants a and b default to 0.00588 (1/170) and 1 respectively (following Manzocchi et al 1999).

TGTDT

This keyword has been made redundant in TransGen 3 and has been replaced by the THICK plugin either automatically by WizGen in "Basic Project" mode (see Plugins generated by WizGen in Basic project mode) or created by the user via the User-defined plugins page of WizGen (in "Flexible Project" mode). The TGTDT keyword will be recognised by version 3, but TransGen stops with an error report.

Displacement to thickness lookup table

This keyword can be used instead of TGTDE and allows input of a table of paired displacement and thickness values. The thickness of a faulted connection vertex is calculated from this table by linear interpolation. The data is supplied as an even number of paired real values of displacement and thickness. The data must be terminated with a slash (/). e.g.

TGTDT
0.1 0.001 1. 0.01 10. 0.1 100. 1.0 etc /

TGTHROW

(new TransGen version 3.2 keyword)

To define User-defined fault throws

When using the new Include fault drag and hierarchical zone effects functionality, this TransGen keyword can be used to:-

define the locations and throws of  fault trace(s) that do not exist explicitly in the model geometry

modify the throw of a trace with an explicit throw in the model geometry. This is not recommended - the TGDRAG keyword, which is associated with a greater range of functionality, is designed to do this.

The keyword can be followed by as many definitions as needed on separate lines, each line and the file are terminated by a slash. The format of each line is:-

I J `Direction' Throw
I and J are the indexes of the cell stacks to which the fault trace is adjacent.
Direction can be either `DIR-X' or 'DIR-Y'. The single quotation marks are essential. If the direction is `DIR-X`, the trace lies between the cell stacks [I,J] and [I+1,J], and if it is `DIR-Y', it lies between cell stacks [I,J] and [I,J+1]. Note that traces cannot be defined on the X- and Y- faces of a cell stack - this must be achieved by assigning the trace to the X+ and Y+ faces of the adjacent cell stack.

Throw can take both positive and negative values. A negative throw value signifies that the cell stack closer to the model origin lies on the upthrown (footwall) side of the fault, while a positive throw value signifies that it lies on the downthrown (hangingwall) side of the fault. Units are the native units of the TransGen run (meters for metric, feet for field and cm for lab).
For example the line:-

12 32 `DIR-X' -23.0 /

places a user-defined trace between cell stack [12 32] and [13 32] with the depths of the faces of the former stack being raised by 12.5m and of the latter stack lowered by 12.5 m. If the same trace is specified twice, the two throws are summed. See Processing modified traces (in the section on The revised ViewGen workflow) for details of how the input data are processed to produce a coherent geometrical model.

Faults defined using the TGTHROW keyword are processed independently of faults included explicitly in the model geometry. However if a user-defined trace specified with TGTHROW overlies an existing trace in the parent model, ViewGen issues a warning and lists the trace in the TGPRT file. Only the throw specified in the TGTHROW keyword is used to construct the 2D sub-resolution fault throw grid.

Example

Application of the following listed TGTHROW include file to an initially unfaulted 20 by 20 cell model (Fig a) will result in the coherent sub-resolution fault geometry shown in Figure b.

NOTE:- Fig b is shown purely as an example of the operation of the TGTHROW keyword - this display will not be obtained in ViewGen since the parent model (Fig a) remains unfaulted (see Visualising fault drag and/or fault zone data for further discussion).

One or more files containing the TGTHROW keyword and others (e.g. TGFZONE, TGDRAG and/or TGTRACE) can be included in the TransGen run via the Included Data page of WizGen, along with all cell and (if used) user-defined connection properties, when running a "Flexible project" with the Include fault drag and hiererchical zone effects option selected.

HINT:- The TGTHROW data can be output to file at the end of a TransGen run via the Output - derived and user-defined properties page of WizGen.

TGTRACE

(new TransGen version 3.2 keyword)

Allows the inclusion of user-defined trace properties

When using the new Include fault drag and hierarchical zone effects functionality, this TransGen keyword allows the inclusion of to 10 user-defined trace properties for each fault trace (system or user-defined) in the model. These properties can then be used in DRAG and/or FZONE plugins (see The set of trace properties available for use in the DRAG & FZONE plugins).

The keyword can be followed by as many definitions as needed on separate lines, each line and the file are terminated by a slash. The format of each line is:

I J `Direction' `i1 i2 i3 i4 i5' `d1 d2 d3 d4 d5' /

[i1 to i5] are five user-defined properties with integer values and [d1 to d5] are five user-defined properties with non- integer values (floating point numbers). These user-defined trace properties are entirely optional and fewer than five values of each can be specified. For example the line:

12 32 `DIR-X' `11 22 13' `12.322423 0.0001' /

assigns to the trace three user-defined integer values and two user-defined floating point values, while the line:

12 32 `DIR-X' `' `12.322423  /

uses only one user-defined value which is a floating point number.

One or more files containing the TGTRACE keyword and others (e.g. TGFZONE, TGTHROW and/or TGDRAG) can be included in the TransGen run via the Included Data page of WizGen, along with all cell and (if used) user-defined connection properties, when running a project in Flexible mode with the Include fault drag and hiererchical zone effects option selected.

HINT:- The TGTRACE data can be output to file at the end of a TransGen run via the Output - derived and user-defined properties page of WizGen.

TGVOLERR

(new TransGen keyword)

Sets a minimum cell volume error tolerance

This TransGen keyword sets a precision limit for rejecting badly constructed cells. It is automatically included in the TransGen run file using the Cell volume error tolerance, Lower limits setting on the Miscellaneous Options page of WizGen (default value 1.0e-06).  A grid-block is flagged as inactive by TransGen if any component tetrahedra have a volume of less than minus this value, or the total cell volume is less than this volume. Changing it from a value of 1.0e-06 should only be done in special circumstances.

NOTE:- This Cell volume error tolerance cutoff is different from the Eclipse limit on minimum cell pore volumes set by the MINPV keyword which TransGen also recognises (as set by the Cell pore volume cutoff on the same page of WizGen).

The keyword should be followed by a record setting the limit, followed by a slash.  

TGVOLERR
1.0e-06 /

Units: cubic metres (METRIC),  cubic feet (FIELD) or cubic centimetres (LAB).

TGVS

(new TransGen keyword)

Vshale content

To use Effective vshale values based on either Vshale only or NTG and Vshale (see Effective vshale computation method on the