Tutorial Saphir 32

Overview

Table of contents • A00-1

Saphir V.3.20 - © KAPPA Engineering 1990-2004

Tutorials

Saphir guided interpretations B01 B02 B03 B04 B05 B06 B07 B08 B09 B10 B11 B12 B13

• Saphir guided interpretation # 1:Basic features • Saphir guided interpretation interpretat ion # 2: Multi model • Saphir guided interpretation interpretat ion # 3: Multi gauge, multi period • Saphir guided interpretation interpretat ion # 4: Multi well, 2-D Map • Saphir guided interpretation interpretation # 5: Gas test, rate dependant dependant skin • Saphir guided interpretation interpretat ion # 6: QA / QC • Saphir guided interpretation interpretat ion # 7: 2-D Map Numerical model • Saphir guided interpretation interpretat ion # 8: 2-D Map Numerical model (2) • Saphir guided interpretation interpretat ion # 9: Material balance • Saphir guided interpretation interpretat ion # 10: Real time • Saphir guided interpretation interpretation # 11: Multilayer Analysis with Static Rates • Saphir guided interpretation interpretation # 12: Multilayer Analysis with Transient Rates • Saphir guided interpretation # 13: Numerical Multiphase

Additional examples BX01 BX02 BX03 BX04

• Additional Saphir example # 1: 2 Φ PSS • Additional Additional Saphir example # 2: 2: Horizontal Horizontal well • Additional Saphir example # 3: Rate grouping • Additional Additional Saphir example # 4: Tidal Tidal effects

Doc V3.20.00

Installation and Learning Guide

Saphir guided interpretation #1 • B01-1

Saphir V.3.20 - © KAPPA Engineering Engineering 1990-2003

Doc V3.20.00

B01 – Saphir guided interpretation interpretation #1: Basic features

This chapter is an introduction to the basic features offered by Saphir. This first interpretation can be followed in all levels of Saphir except Saphir Reader.

B01.1 • Before starting Saphir To follow this guided interpretation, you should have: •

Saphir installed on your system

•

The files Sapb01.rat and Sapb01.asc, which by default are copied during installation to your hard disk, and should be located in the program example directory directory

B01.2 • Starting a Saphir session In the File menu of the Saphir main screen choose “New”, the initialization “main options” dialog box is opened:

Fig B01.1 Initialization Initialization main options dialog box The initial analysis type can be selected here: Standard, Numerical or Multi-layer analysis. This screen allows the following inputs: •

Test Type: Standard or Interference.

•

Fluid Type: Oil, Gas, Water

•

Rates:

Oil, Gas, Water

The well and reservoir related parameter such as: •

Net drained vertical thickness, h

•

Well radius, rw

•

Average porosity




Doc V3.20.00

The user can also specify: •

Available rate data for loading, defined by the user.

•

The reference date and time to calculate the elapsed time from data vs date/time of day (TOD) files. files.

The user can have access to the “units” dialog by selecting the tab “Units”

Fig B01.2 Initialization Units dialog box The default units at installation are Oil Field. Alternative units can be chosen by clicking on the “Load” button (SI or Hydro systems). The system of units can be customised by selecting the desired unit for each parameter. The “Other Units” are Emeraude units not used by Saphir, but a single common system of units can be shared between the two applications. A customised customised units units file can be saved saved and reloaded reloaded whenever whenever required. required. Click on

to input the test information in the table:

Fig B01.3 Initialization Information dialog box Note: Additional information information can be input by clicking “Add”.

Go back to the Main Option dialog by pressing on the tab. Click on “Next” to accept the displayed values and the default settings. The PVT dialog is displayed:




Doc V3.20.00

Fig B01.4 Fluid PVT parameters dialog This dialog allows the input of PVT parameters when available. If PVT parameters are not available, checking one or more of the PVT parameters will activate the Calculate option

to access the Kappa PVT facility.

Click on “Create” to accept these default parameter values and to access to the “interpretation” control panel page. Save the current file as “Sapb01” with the “Save” option in the “File” menu, in the sub-directory “example” under your installation directory.

B01.3 • Saphir Main Screen The Saphir main screen, Fig. B01.5, consists of: •

A control panel similar to Microsoft Outlook TM on the left of the screen. The first page to appear is ”Interpretation”. Options not available at this stage are greyed out. The next logical default option is highlighted by a red frame.

•

The menu bar at the top of the screen gives access to standard Windows facilities.

•

The toolbars.

•

The screen pages corresponding to various aspects of the analysis; QA/QC, Edit Data, Edit Rate, 2-D Map and Analysis, can be selected by clicking on the relevant tab.

•

Process messages are displayed at the bottom of the screen.

Fig B01.5 Saphir main screen; Interpretation panel




Doc V3.20.00

B01.4 • Loading a rate history file As no file has been loaded at this stage, the only options enabled in the control panel are Load Q (rates) and Load P (pressures). The default, framed in red, is to load the rates. Click on

to load the rate. The first dialog asks for the origin of the data.

Fig B01.6 Origin of the data Keep “Ascii file” as the input origin, and click OK. Select the file Sapb01.rat with the browser, the load dialog is displayed:

Fig B01.7 Load Rate Dialog The main options in the load dialog are: •

Type of rate information: surface (default) or downhole.

•

Type of time information: duration (default) or time at start.

•

Definition of time shift.

•

Re-selection of the input file.

•

Definition of the file format.

The format can be defined as “field” (series of characters separated by tab, spaces or user defined), or “column” (selecting the index of the characters in the line). The default (also called free format) is indeed “field” mode, where the first field contains the (decimal) time information and the second the value, in this case the rate. Accept the default setting and click on “Load”.




Doc V3.20.00

B01.5 • Loading a pressure history file The rate history is now displayed in a new plot and the control panel the default option is now on “Load P”:

Fig B01.8 Main screen after rate load

Click on

and the next dialog box asks for the origin of the data.

Fig B01.9 Data origin dialog Keep “Ascii file” as the input origin and click ‘OK’ to confirm; select the file Sapb01.asc with the browser. The load dialog is displayed:




Doc V3.20.00

Fig B01.10 Load pressure dialog The program suggests loading the file Sapb01.asc in free format with no time shift, no window (i.e. no limit in the time and pressure ranges) and no filter. These items are illustrated in the second guided interpretation and in the relevant pages of the Reference Manual. Click on

to load the pressure history using the default settings. The pressures are now displayed on the

history plot:

Fig B01.11 Job history plot




Doc V3.20.00

B01.6 • Extracting a group In the control panel the “Extract dP” option is now enabled and suggested as default: i.e. select pressure data from a group where the diagnostic will be performed. Click on

to call the “Extract dP” dialog:

Fig B01.12 Extract Delta P dialog Saphir is not restricted to extracting DeltaP for a single constant rate and an analysis can be made on a “group” of rates. By default, when the rate history is loaded, Saphir creates groups that can be: •

A production group: A sequence of flow periods of strictly positive rates.

•

An injection group: A sequence of flow periods of strictly negative rates

•

A build-up: A zero rate period after a production

•

A fall-off: A zero rate period after an injection

The dialog Fig B01.13 allows the selection of the gauge (in this example there is only one) and the group (buildup #1 or production #1). Click on ‘OK’ to confirm the choice. The DeltaP parameters dialog is displayed:

Fig B01.13 DeltaP parameters dialog The default settings suggest extracting the last zero flow group from the pressure history, with a filter of 100 points per cycle and a derivative smoothing of 0.1. Please note that these default settings may be modified in the “Settings” page. A value of initial pressure (P at dt=0) is suggested: it is the closest existing pressure point to the time when the rate changes. If not, as in this case, the first pressure recorded in the build-up period is selected. Click on ‘OK’ to accept the proposed defaults values.




Doc V3.20.00

A log-log plot of the pressure change and the derivative is displayed at the same time as a semi-log plot of pressure change against superposition time and the job history plot, Fig. B01.15. Also note that the active group extracted is highlighted on the history plot. These "Automatic" plots are always displayed after a group has been extracted. Their scales may be changed but they cannot be deleted, as they are necessary for the rest of the interpretation process. The match lines have been drawn on the log-log plot, the pressure match is set where the derivative starts to stabilize and the 45° line for the time match has been set on the early time (wellbore storage) data. The default option on the control panel is now ‘Model’.

Fig B01.14 Main screen after extracting deltaP It can be seen that at early time, the derivative follows a unit slope, but with a difference between the pressure and its derivative, which is likely to come from a wrong estimation of the initial pressure. The difference is 3 psi out of a total pressure change of several hundred psi. The true value of p@dt=0 should be 3087 psia. We shall override the default option and correct deltaP to get a better pressure curve.




Doc V3.20.00

B01.7 • Active plot options

Activate the log-log plot by clicking on it and make the plot full page by doubleclicking on the title bar. Click on the right mouse button on the plot: the pop-up menu shown opposite is displayed. The equivalent toolbar specific to this plot is shown below. Note: This is only available when the log-log plot is full screen.

B01.7.1 • Correcting a flow period Click on the right mouse button in the log-log plot area and select “Set” and P@dt=0 in the pop-up menu. The following dialog is displayed:

Fig B01.15 Input the correct last flowing pressure: 3087 psia. Click OK. The early time data are now corrected in the log-log plot and we are ready to proceed with the analysis:

Fig B01.16 Log-Log after correction




Doc V3.20.00

B01.7.2 • Drawing a semi-log straight line Click on the right mouse button to display the pop-up menu and select the option “line” and then “regression”. On the plot click twice to indicate the time limits between which the derivative exhibits IARF stabilisation. Saphir displays a yellow line corresponding to the semi-log straight line slope. At the same time, the IARF straight line is drawn on the superposition plot:

Fig B01.17 IARF straight line The semi-log straight line can also be drawn with the same procedure on the superposition plot. Clicking on the right mouse button on the maximized log-log or the semi-log plot, then on results or the results button in the toolbar shows the results corresponding to the straight line. Results can also be accessed by double clicking in the small “status” window at the top right of the screen.

B01.8 • The file menu This is a good time to save the file and see how it is stored: Click on

to save the interpretation at this stage.

File menu: The available options, detailed in the reference manual are: •

New

To start a new interpretation

•

Open

To open a previous interpretation

•

Close

To close the current interpretation but remain in Saphir

•

Save

Save file to the current name and directory

•

Save as

Continue the interpretation with a new name



Saphir V.3.20 - © KAPPA Engineering 1990-2003 •

Previous Version

Reverts to the latest saved version

•

Print

Print plots, results and listing

•

Print Preview

To view on the screen the report final apperance

•

Print set up

Choose the printer, the page format etc.

•

Exit

Close Saphir

The session named Sapb01 is stored in a file named Sapb01.ks3 in the active directory.

Doc V3.20.00




Doc V3.20.00

B01.9 • Model generation The “Model” option

is now the default on the control panel.

Clicking on “Shift” causes a “flash” sign to appear on the interpretation page icons of the control panel. This calls the “automatic” equivalent of this option. Click on

to generate the automatic model.

The automatic model assumes a homogeneous infinite behaviour. The kh and C being calculated from the current pressure and time matches, Saphir estimates a value of skin coherent with the pressure change and generates the model. Using this model and the flow-rate history, the program then back calculates the initial pressure Pi from p@dt = 0. The multirate model is generated and displayed with the data for comparison on the Log-log, Semi-log and History plots, Fig. B01.18.

Fig B01.18 Main screen after the automatic model generation

Click on the model button

to call the model dialog box:

Fig B01.19 Model dialog box




Doc V3.20.00

Since the automatic model has been generated, Saphir is holding a value for the Skin factor and the ‘Generate’ option is enabled. The model dialog box is presented as a spreadsheet. Each parameter value can be edited in the corresponding box. Three drop boxes select the desired model by determining the well conditions, the reservoir type and the boundaries. On the lower left hand side, boxes can be checked to enable special model conditions such as rate dependent skin and changing wellbore storage. The pseudo time and show p-average can also be enabled. The Pi value can be specified and imposed instead of using the back simulated value from the model. In case of a multiwell configuration the user can add the interference from other wells. On the bottom the

button opens a dialog whereby the user can extend or shorten the simulation

calculation.

B01.10 • Matching the model to the data Activate the log-log plot by double clicking on the header. Click on the right mouse button and in the pop-up menu select “results” to see simultaneously a short version of the results and the facilities of the active plot. To modify the match, click on the plot holding the mouse button down. The model can then be dragged across the screen to the best fit with the data. The results calculated from the match are simultaneously updated.

Fig B01.20 Adjusting the match Double click on the plot header to go back to the Main Menu. The default is ‘Improve’. Click on Shift/Model to generate the automatic model again in order get a model consistent with the new pressure and time matches. The Improve option invokes a non-linear regression routine which iteratively matches the data to generated models, changing in this case the values of C, k and S, until the best fit is obtained. In more complex models, boundary distances, interporosity flow coefficients and all relevant parameters are taken into account. Individual parameters may be left variable or fixed.




Click on

Doc V3.20.00

to call the improve option. By default all parameters are set as variable, Fig B01.22.

Fig B01.21 Improve dialog box The parameters are selected constant or variable in the regression by setting the corresponding check box. Clicking on

will call the following dialog:

Fig B01.22 Fitted data selection The “ + ” (add) and “ - ” (delete) buttons allow regression points to be added or removed. Points are deleted by defining a box, as for a Zoom-in. When the second corner of the box is selected, and the mouse released, the points inside the corresponding time range are deleted. Additional points can be added by selecting “+” and clicking at the time when the point is to be added. Click on “OK” to accept the default set of points. The program returns to the Improve dialog. Click on

to start the regression process.

The program starts the non-linear regression, (Fig B01.23) changing all allowed regression parameters to reduce the standard deviation between the simulated pressures and the data for the selected points. The parameters of the previous iteration are displayed on the table showing their position between the limits set for the regression.




Doc V3.20.00

Fig B01.23 Regression progress dialog Double click on the plot header to return to the main screen.

B01.11 • Horner plot Since this test was a build-up following a single production period a Horner plot is a valid alternative to the superposition plot. In the Control panel select the page “Interpretation 2”. Click on the icon

to access the flexible plot facility:

Fig B01.24 Flexible plot dialog At the top of the dialog box the user can choose in the combo box predefined plots (MDH, Horner, Square root, Tandem root, user defined). To practice we will create a user defined plot. The default is the superposition (multirate) plot for the X axis, and for the Y axis, pressure values on a linear scale. To create a Horner plot, select Build-up, Saphir then suggests the equivalent production time tp, which can be edited. Default is cumulative production divided by the last flow-rate. Click OK to confirm your choice. The plot appears at the bottom right side of the screen showing the pressure data and the model. It also shows the corresponding derivative (slope of the Horner curve) to allow the user to visualise stabilized IARF. Activate this plot by a double click on the header. The plot is displayed full screen.




Doc V3.20.00

Clicking on the right mouse button displays a pop-up menu which allows the drawing of a straight line (see above) and then shows the main results calculated from the straight line. In the same pop-up menu the user can select “results”, it displays a table indicating the results of the straight line.

Fig B01.25 Horner and derivative plot Note: When exiting this Saphir session save the Sapb01 file to use it later in another guided session. What next? •

Restart the same interpretation from nothing without using this documentation.

•

Try the same facilities on a data set of your own.

•

Go to the second guided interpretation, to use more control panel functions and additional options. The second guided interpretation assumes that you are familiar with the basic facilities described in this first introduction.

•

Alternatively, File, Exit and you are back to the Windows menu.




Doc V3.20.00

B02 – Saphir guided interpretation #2: Multi model

This guided interpretation covers the use of the Ascii file format option, time shift, correction of time errors, the manual improve and the multi-model/analysis options. This guided session will use the data sets Sapb02.rat and Sapb02.asc stored in the Examples directory during the installation. It is assumed that the user is familiar with Saphir to a level demonstrated in the guided session #1; therefore the previously discussed functions will not be covered in detail in this session.

B02.1 • Starting a New Interpretation Main well test events: The test was started on 25th April at 7:00 AM. After two false starts the well was finally flowed for over 20 hours before being closed in for 24 hours on the 26th April around 6:30 PM. Start Saphir, in the menu “File” select “New” and go through the initialization procedure. Define the reference date as: 25 April 2002 00:00:00. Accept the other default values. When initialization is over, click “Create >>”. This moves Saphir to the main interpretation screen with the “Outlook” type control panel The “what’s next” default (red square) is on the

(rate) button.

Go to the “File” menu to save as Sapb02 in the example subdirectory.

B02.2 • Loading the Rate History with the Format option Click on

to start the load process, use the Ascii file Sapb02.rat. Choose the file, and the content is shown

in the table in the Saphir load data dialog. Scrolling down in the column mode, observe that the rate data is shown as ‘time at start’. Therefore, change the setting of the time selection droplist from “duration” to “time at start”. Staying in “Column mode” and scroll right and down to display the table with the data. Using the cursor, highlight the columns (click and drag on one row only) corresponding to the time values and select “hour:min” in the menu that automatically appears when the mouse button is released. Proceed in the same way for the rate data columns and select “Liquid rate”:




Doc V3.20.00

Fig B02.1 The ‘Load’ button in now enabled. At this point it is possible to click on ‘Load’ and the data load would proceed, however do not do this at this point. We will investigate an alternative format. Change to “Field mode” and scroll right and down to display the data. You can define the time format by clicking on the time field header and selecting “hour:min” and the rate format by clicking on the rate field header and select “Liquid rate”. The default unit for “Liquid rate” is STB/D, but this can be customized in the drop list that appears when clicking on the unit (STB/D) button When the format is correct all the valid lines are shown in white and the “Load” button is enabled:

Fig B02.2




Doc V3.20.00

Click on the “Load” button and you will observe the process bar in progress during the load. A warning is displayed:

Fig B02.3 This warning is given when the program finds the time values not to be strictly increasing and asks the user to confirm if this corresponds to an error to be ignored or a change to the next day. In this case, 24 hours have to be added to the time value. As the acquisition was made over several days, select “Apply to all” and the 24 hours jump would be systematically applied.

Fig B02.4 At the end of the load, not knowing the duration of the last flow period Saphir will ask for this duration. In this example, accept the default value 24 hours and click OK.

B02.3 • Loading the Pressure History with the FORMAT option Select

which is now the default. Accept the default ASCII in the load data dialog.

The file to load is “Sapb02.asc”, if you are not in the default directory (..\Examples) find it with the Windows™ browser. In the format definition window, Field mode is the default. Click on the column header button above the values of time and choose Hour:Min:Sec. Proceed in the same way with the pressure field: click on the pressure column header button and in the pop up menu select “Pressure”. The default unit is psia, but this can be changed by using the pop up list available with a click on the button indicating psia.




Doc V3.20.00

Fig B02.5 The background color changes from gray to white indicating that the format is properly defined. The pressure recording started on the 26th whereas the file reference date is the 25th. This can be corrected with the “Time shift” button. Click on “Time shift”. The document and the gauge reference date/time are displayed. Because the defined th

format is in “Time of day”, only the date is enabled for the gauge reference. Change the date to the 26 , and note that the Time Shift is automatically updated to 24 hrs. Validate with OK. Note that when you are back to the load data dialog, the “Time shift” button now has a green check mark.

Fig B02.5a Click on

to execute the load and again you will get the warning:

Fig B02.5b




Doc V3.20.00

As previously, answer “Apply to all”. The Job History is displayed, Fig B02.6, but the pressures are not synchronized with the rate file.

Fig B02.6 Job history plot

The default option in the control panel is now “Extract dP”. Click on

to call the “Extract dP” dialog, accept

the default value and load the last flow period, in this example the build-up.

Fig B02.7 Automatic plots, unsynchronized data




Doc V3.20.00

The appearance of the pressure and the derivative at early time is typical of a time error (pressure/rate not synchronized) and it is usually the rate history that is incorrect. This error is common since the rate information is taken from the well test report (usually hand written) whereas the pressure data comes from the gauge or from an acquisition system with a relatively more accurate clock time, therefore the start time of the build-up from the gauge and the well test report do not match exactly. We are now going to adjust the rate history to the exact time at which the pressure break (shut-in) is observed.

B02.4 • Adjusting the rate history Select the Edit rates option tab. The Well test history plot is displayed: •

On the left hand side there is a spreadsheet with the rate data.

•

On the top right hand side the pressure history plot.

•

In the middle the schematic of flow period groups

•

On the bottom right hand side the rate history plot

Fig B02.8 Job history plot in Edit rate aspect Create a pencil thin zoom-in (on the pressure plot) of the first few pressure points of the Build Up #3. Do this once or twice, by defining a window by click and drag. You should now see that what first appeared as the first build-up point is, in fact, a number of points very close together and they are still in the drawdown, you can see that P @ dt=0 does not correspond to where the rate history goes to zero, this need adjusting, Fig B02.9. Zoom in again if necessary. Note that on the Y-axis, the Zoom works on the pressure scale only, and not on the rate scale.




Doc V3.20.00

Fig B02.9 Zoom in to view end of flow period To synchronize the rate history shut-in time and the pressure history: •

Click on the

button (move rate to closest pressure).

•

Click and drag the dashed yellow vertical line on the rate plot to where the last flowing pressure point on the pressure plot should be.

The rate and the pressure histories are now synchronized.

Fig B02.10 Rate history synchronized




Doc V3.20.00

The Rate group schematic is also updated. Click on the “Analysis 1” tab to go back to the interpretation screen. With the Shift key pressed, select “Extract dP”, which executes an automatic “Extract dP” for the final flow period; the build-up: The log-log plot now has the correct shape at early time and we can proceed with the interpretation process.

Fig B02.11 Extract dP with corrected history

B02.5 • Basic Interpretation - Log-log analysis The build-up can be interpreted using various analytical models, first we will consider the diagnostic to show a response rom f a homogeneous reservoir with an outer boundary detected some 3 hours after shut-in. The following is very close to the first guided interpretation. The program has set the pressure match on the derivative data between 1 and some 3 hours, which is consistent with a homogeneous reservoir with a boundary effect. Click on

(shift/

) to call the automatic model option. The program generates the automatic model as

explained in the guided interpretation #1 and displays the updated plots (homogeneous infinite acting). The option improve

is now the default in the Control panel.




Doc V3.20.00

B02.6 • Improving the analysis "manually" We are going to improve the match of the early part of the data before considering the boundary. Click on the improve button

in the control panel.

Fig B02.12 Manual Improve; variable selection The program suggests the model parameters, C, S and k as variables. Leave all parameters variable then click on

. The log-log plot is displayed with the points selected for regression marked by white inverted

triangles. Delete the last three points, as we are not interested in the late time boundary effect at this stage. Select

in the menu bar. Then define a box (click & drag), exactly as in the zoom option.

Fig B02.13 Deleting the last 3 regression points The points inside the selected box are deleted. If you make a mistake click on •

Click on “OK” to go back to the “improve” dialog.

•

Click on “Run” to proceed with the regression.

to recover the original points.

We did not call the automatic improve, because in this case the program would have regressed on all default points including the last points where the boundary is observed. This would have changed the pressure match and reduced the quality of the early time fit (time match). When the regression is finished the three automatic plots are displayed.




Doc V3.20.00

Fig B02.14 Improved infinite acting model

B02.7 • Adding a boundary effect Select the model button

in the control panel.

The model dialog box is then displayed. In the “Boundaries” drop list select “One Fault”:

Fig B02.15 Model dialog and Boundaries menu A default distance to the sealing fault is proposed. To help in the estimation of a more realistic distance it is recommended to use the “pick option” button

:

The log-log plot is displayed and you are asked to pick the time when the derivative data leaves the previ ously established level of radial flow, at about 3 hours. Accept the distance calculated by the program and generate the model.




Doc V3.20.00

Fig B02.16 Boundary match before regression We see on the log-log plot that the distance was over estimated as the limit effect occurs too late. Accept the default manual improve Click on

, in the dialog then uncheck C and k to fix them as constants

to choose the regression points, click on

to recover the original points.

Click “OK” to go back to the previous display. Click “Run” and the regression program adjusts the distance to the boundary (Fig B02.17).

Fig B02.17 Boundary match after regression




Doc V3.20.00

Double click on the log-log plot header to display it full page, click on the right mouse button in the plot area and select “Results” in the pop-up menu to display the results dialog box or select the third button in the log-log plot menu bar:

Fig B02.18 Results dialog box The data stops before the second stabilization is reached. A semi-log analysis may now be attempted.

B02.8 • Semi-log analysis Activate the log-log plot by a double click on the header, right mouse click in the plot area and choose “Line” in the pop up menu (or use the ninth button in the log-log plot menu bar). Click twice to determine the time limits of the first stabilization corresponding to the radial flow. Saphir automatically switches to the semi-log plot (superposition) and displays the semi-log straight line corresponding to the line that was identified on the log-log plot. Two large vertical tick marks indicate the interval of the regressed data points, a smaller tick mark indicates the value at dt=1 hour, and p* (intercept) is indicated by a horizontal tick on each y-axis. The result window comes up automatically and shows the results both from the model match and the straight-line slope. A semi-log boundary (single fault) distance analysis can also be performed. Click on the “Interpretation 2” header in the left main control panel (towards the bottom) to call the special interpretation facilities. Click on

to call the “Flexible plot” facility:




Doc V3.20.00

Fig B02.19 Flexible plot dialog Choose linear pressure versus superposition time (Multi-rate). The superposition plot is displayed. The derivative plot can be displayed by selecting “Show/Derivative” in the menu by cllicking on the right button on the plot. The derivative plot will help you to recognize the IARF period. The principle is to match the first IARF (before the fault is detected) with the first radial line. kh and skin are calculated from this line. The second radial will correspond to the “Semi-IARF” after the fault is detected. Theoretically the second radial flow straight line will have a slope twice the 1st radial if there is only one single fault. If the slope is greater than twice the 1st radial, one could use this method, assuming intersecting faults, to give the ratio of the angle between the faults. Based on kh and the time at which both lines intersect, the distance to the fault is calculated. Click on the right mouse button, select “line”, “Combined”, and “Faults”, Fig B02.20:

Fig B02.20 Combined – Faults analysis




Doc V3.20.00

The “Line” pop-up menu now includes specific options for the first and second radial flow. Define the first radial by regression through the points corresponding to the first Infinite Acting Radial Flow on the derivative. In the same way define the second line (using Draw) through the last few points, which correspond to the start of semi-radial flow. You can also use the ninth button in the flexible plot menu bar.

Click on the right mouse button and select “Results” to call up the results, Fig B02.21, pull down the results dialog box in order to see the results of both straight lines. These results are automatically updated when the straight lines are modified. The slope of the line through the last points should be twice that of the first line so we will adjust it to this value. Click on the right button and in the pop-up: “2nd radial” “Modify”. Click near the center of the plot and hold the mouse button down. Then drag the mouse horizontally to adjust the slope, without affecting the height, until the slope is twice that of line 1 and then drag the mouse vertically to adjust the height, without affecting the slope, until the second line is at a tangent to the end of the model curve. If the “semi-IARF” was more established the second line would go through the last data points.

Fig B02.21 Specialized boundary analysis




Doc V3.20.00

B02.9 • Multiple interpretations On the right hand side of the tab bar select the tab “New”; it opens the new interpretation dialog box:

Fig B02.22 New Analysis Select “Nothing” to start from, this means that the new interpretation will be based on nothing as if we started from scratch, then click OK. Click Extract dP

on the Interpretation page of the Control Panel to call the Delta P dialog and proceed as

in before. The second model will be a 2-porosity transient, thus the pressure match has to be adjusted in order for this diagnostic to make sense. Activate the log-log plot by double clicking on the header and adjust the match by clicking and dragging.

Fig B02.23 Log-log match adjustment




Go back to the main screen. Click on automatic model

Doc V3.20.00

(shift/

) to call the automatic model option to

generate a homogeneous model fitting the early time data (Fig B02.24).

Fig B02.24 Homogeneous model match

Click on

to access the model catalog and select the 2-porosity transient (sphere) reservoir model in the

“reservoir” drop list box (Fig B02.25).

Fig B02.25 2-porosity parameters dialog




Click on the

button to determine interactively the lambda and omega values.

The log-log plot is activated. Click on the time corresponding to the change of slope (fig B02.26). Note: all the graphic zoom tool box is available for the user commodity.

Fig B02.26 2-porosity parameters graphic evaluation Click “Generate”. The main screen shows the log-log, semi-log and Cartesian plots. Click on Shift

(automatic improve) to adjust the match (Fig B02.27).

Fig B02.27 Final match with 2-porosity 2-porosity model

Doc V3.20.00




Doc V3.20.00

Click on “New” and choose the option to start from “Analysis #1”. The Automatic plots are displayed with model #1. Click on model

to access the model catalogue. Select a radial composite reservoir model and change the

boundary condition to “infinite”.

Fig B02.28 Model Dialog for Radial composite parameters

Click on the pick option Click on the next pick option

to determine interactively interactively the heterogeneity distance. to determine interactively interactively on the log-log plot the mobility and diffusivity diffusivity ratio

value. You are asked to pick the second stabilization level. Click “Generate”. Click on on Shift/ Shift/

(automat (automatic ic improv improve) e) to adjust adjust the the match. match. You You can can see on the the screen screen the the paramet parameters ers value value

progression progression which results in the refined match illustrated in Fig B02.29.

Fig B02.29 Composite model match after improve




Doc V3.20.00

You can continue to open as many analyses as required and go back to a previous diagnostic at any time by clicking on the corresponding tab. When the log-log, semi-log, semi-log, or history plot is maximized, maximized, you can toggle the “Compare analyses” mode on. Double-click on the history plot title bar. When the plot is maximized click on the

button in the toolbar or use the right click method to access the menu to display the Active only, All, or

List:

Fig B02.30 History plot showing “All” analyses Return to the main screen and save the session.

Installation and Learning Guide Saphir V.3.20 - © KAPPA Engineering 1990-2003

Saphir guided interpretation #2 • B02-20 Doc V3.20.00




Doc V3.20.00

B03 – Saphir guided interpretation #3: Multi gauge, multi period This guided interpretation covers the use of pressure data from multiple gauges, the comparative quality control of these data sets, the criteria that can be used to select the most reliable gauge and multiple flow period analysis. This guided session will use the data sets stored in the files Sapb03a.asc, Sapb03b.asc, Sapb03c.asc, Sapb03d.asc and Sapb03.rat stored in the examples directory which was installed during the setup of Saphir. In order to make the demonstration clear and straightforward, the data is synthetic (simulated) and the gauge acquisition problems were created to simulate some common field problems. It is assumed that the user is conversant with the functionality taught in the two previous guided sessions.

B03.1 • Starting the New Interpretation This oil well test comprises of a 26 hours clean up period followed by a 51 hours build-up, then the well was produced on increasing chokes for 26, 32, 39, 38 hours respectively and then closed in for an 85 hour final build-up. The data was acquired with two gauges (1 and 2) set at 9870 ft and two gauges (3 and 4) set at 9888 ft. This data is stored in files Sapb03a.asc, Sapb03b.asc, Sapb03c.asc, Sapb03d.asc respectively. Select “New” in the “File” menu and go through the initialization procedure, accepting the default values for the different parameters. When the initialization is over, click “Create >>”. The main interpretation page is opened by default with its “Outlook” type control panel. Click on the “QA/QC” tab to open the QA/QC page. Go to the “File” menu to save as Sapb03 in the example sub directory.

B03.2 • Loading the data sets Click on

in the QA/QC control panel page. Choose to load from an Ascii file (default) and in the file browser

select the file Sapb03a.asc.

Fig B03.1




Doc V3.20.00

Click “Load” and observe the loading process. During the load the following dialog appears:

Fig B03.2 Select “Apply to all” and OK. The first pressure data set is displayed:

Fig B03.3 Follow the same procedure for the other files. All the four pressure data sets are plotted on the screen:




Doc V3.20.00

Fig B03.4 To activate the legends click on

. The active and reference gauges are as defined by the droplists . The reference and active gauges will also be indicated

in the displayed legends.

B03.3 • Quality control options

B03.3.1 • Creating the difference plot

Since more than one pressure data set has been loaded the icon

“Difference” is enabled. Click on this

icon. The following dialog asks the user for the channel to use as a reference for the difference plot and the number of data points used to calculate it. Choose one of the lower gauges as reference and choose “Ref. Gauge – Gauges”. This is in order to keep the established convention of “Differential Pressure Analysis” and facilitate the recognition of wellbore phenomena such as a change in fluid densities in the wellbore.

Fig B03.5




Doc V3.20.00

Fig B03.6 The difference plot is displayed. You can click on the difference plot to activate it and use the zoom facility as many times as necessary to magnify and investigate the difference curves various responses. The above plot clearly shows that the orange curve is not synchronized in time with the others which is normal as it is rare that different pressure gauges have the same gauge time. In order to smooth the curves the QA/QC module of Saphir can easily do this synchronization in various ways: automatic through regression, using the mouse or input manually the shift. We will here show the automatic shift. It is important to choose a common event on all the data channels; the beginning of a build-up is usually fine. Zoom in a few times on the event, see below.

Fig B03.7 It can be seen that Sapb03b is not synchronized with the other gauges, which is the reason we see a number of peaks in the difference channel above.




Doc V3.20.00

We will now proceed with the automatic synchronization. Activate Sapb03b in the drop list and click on

, “Automatic Shift” and the Sapb03b is automatically

synchronized with the reference channel. As a reminder all the tool bar menus are also accessible with a right mouse click within the plot area.

The result after synchronizing can be seen as follows:

Fig B03.8 B03.3.2 • Data comparison and diagnosis The pressure difference between the gauge measurements permits diagnosis of the gauge performance and quality, possible problems can be identified and enables the interpreter to choose the most appropriate gauge to use for the analysis. In addition the engineer can choose which part of the data is valid for transient analysis to avoid misinterpretations. Details on differential analysis are given in the guided QA/QC session and in the reference manual. The plot above is shown after time synchronization between all the gauges and the reference gauge. The difference of gauge number 2 is not stable thus it is conceivable that the pressure gauge has a drift problem that already at this stage eliminates this gauge from the analysis. When looking at the lower gauges (3 and 4), the pressure difference observed (around 7 psi) is consistent with the 18 ft depth difference in an oil gradient. Gauge 3, however, is subject to an unexplained shift between 130 and 235 hours, which makes it unreliable. It is therefore to be rejected. Gauges 1 and 4 are consistent and do not present any apparent abnormal behavior.




Doc V3.20.00

The choice of gauge 4 is therefore self-explanatory as this is the gauge closest to sand face and minimum pressure corrections in a relative unknown fluid environment will minimize errors in the interpreted average drainage area pressure.

B03.4 • Loading the flow periods In the tab bar select “Analysis 1” to display the interpretation screen and the main control panel. In the “Gauge list” drop list in the toolbar select Sapb03d. B03.4.1 • loading the rate history

Click

to load the rate history from the file Sapb03.rat.

The screen now shows the pressure and rate history:

Fig B03.9 B03.4.2 • Refining the rate history Click on the

tab to access the “rate view/edit” window:




Doc V3.20.00

Fig B03.10

In the toolbar, click on the first button

The icon

to bring up the “Actions” drop list menu:

opens the “refine rates” function. This allows the user to refine a rate history by splitting a flow

period and by calculating the corresponding flow rate values in agreement (in time and pressure values) with the pressure history. To experiment, in the toolbar click on the second button

to bring up the “merge rates” dialog, and using

the from/to pick options, merge the flow periods from 3 to 6 and accept the average value.

Fig B03.11




Click on “refine rates

Doc V3.20.00

and select with the cursor the new production period #2, from start to finish.

Fig B03.12 Validate by “yes” the dialog “Keep new (continuous line) rate history?”.

Fig B03.13 Vertical red lines shows where Saphir chose the rate changes from its pressure history analysis. The recalculated rate history is quite similar to the original one and demonstrates the efficiency of this facility in cases like this: increasing rate values and clear pressure behavior as long as the cumulative production is known.




Doc V3.20.00

Fig B03.14 B03.4.3 • Loading one flow period

Click the “Analysis 1” tab and then

to call the “Extract dP” dialog, accept the default value and load the

last flow period in this case the last build-up. The main screen displays the automatic plots with the selected gauge data:

Fig B03.15




Doc V3.20.00

B03.4.4 • Loading all flow periods In the toolbar, open the “Group list” popup menu:

Choose “List”. The “Group selection” dialog is displayed, select “all” groups:

The program will display the "Delta P parameter" dialog for each group not already loaded.

Fig B03.16 Accept the default values for all of them. The plots in the main screen show the curve for all the groups:




Doc V3.20.00

Fig B03.17

Click on

to show the legends of the groups.

On the main screen we now have the pressure data from Sapb03d for all periods of the job history, the log-log plot and the superposition plot. On the superposition plot the various flow periods pressure data is normalized with respect to the reference rate.




Doc V3.20.00

B03.4.5 • Switching from one flow period to another The user can decide in the “Group list” pop-up the desired pressure group to be plotted. As an example you can select build-up #2 in the “Group list” pop-up and then only this group is plotted in the various graphs.

B03.4.6 • Switching to another pressure set In the “Gauge list” droplist the user can select another pressure data set (1,2,3 or 4). They can be compared if only one group is extracted. In the following “Gauge Selection” dialog the user can specify “List” and then the pressure data sets to display:

Fig B03.18 Select “All” and click OK: If the flow periods of the pressure data set were not already loaded the user is requested to input the load information in the “delta P parameters” dialog as many times as needed. Maximize the log-log plot full screen by double clicking on the plot header:

Fig B03.19




Doc V3.20.00

Note: in order to avoid confusion the plot cannot be multi-flow period and multi-gauge at the same time.

B03.5 • Flow period analysis Go back to the main screen (double click on the plot header). Select the most representative flow period with the most reliable gauge (build-up #2, Sapb03d) Click on

to generate the automatic model:

Fig B03.20

B03.6 • Comparison with the other flow periods The match with the current model can be checked on the other groups and flow periods. Switch from one rate group to another by selecting them in the “Group list” pop-up menu. The model is generated each time according to the rate history corresponding to the selected flow period and displayed in each type of plot. Since this pressure data set has been simulated the model match is correct for all of the flow periods but in reality this facility can reveal an inconsistent diagnosis from a single flow period that did not reach or exhibit a pressure behavior appearing in another. For instance the Production period #2, which is the merge of 4 drawdowns, has a longer duration than any other group and nevertheless shows, in spite of small “peaks” due to the grouping of the several flow periods, continuity without any unexpected limit effect. It demonstrates that the first diagnosis can be extrapolated from the 75 to the 125 hours of the Production group #2:




Doc V3.20.00

Fig B03.21 The model can also be compared to the complete set of groups by selecting “all” in the “group selection” dialog:

Fig B03.22




Doc V3.20.00

B03.7 • Comparison with the other pressure sets In a similar way, using the “Gauge list”, the user can superimpose the diagnostics made with the pressure data acquired by one gauge on the pressure acquired by another.

B03.8 • Multiple analyses The diagnostic is usually non unique, so other models can be independently considered without loosing any results from the first analysis. Click on the page tab “New” in order to initialize a new analysis and proceed as instructed in the Guided Session #2. Each new analysis provides the same facilities for switching from one flow period to another and from one gauge to another without needing to reload either a pressure set nor a flow period or delta P. The data to be analyzed is kept in the state it was left in the previous analysis. In the “group list” select production #1. Call the model dialog and add a sealing fault and using the ‘pick option’ define the effect of the fault at the very end on this drawdown (20 hr). This corresponds to a distance of approximately 2500 ft. No effect can be seen on this flow period.

Fig B03.23 Switch in the “group list” to Production #2. We can observe that the model with a sealing fault at 2500 ft shows a slight deviation at the end of the derivative that does not show on the data of this more extended period:




Doc V3.20.00

Fig B03.24 This demonstrates that a model acceptable with one group can be rejected by looking at another. It also shows the significance of grouping, to one flow period, several contiguous production periods.




Doc V3.20.00

B04 – Saphir guided interpretation #4: Multi well, 2-D Map

This guided interpretation covers the use of the multiple well function “2-D Map”. This option allows the user to take into account the possible pressure interference created by the production or injection of other wells in a connected part of the reservoir. This guided session will use the data sets stored in the files Sapb04.asc, Sapb04.rat, Sapb04a.rat and Sapb04b.rat stored in the Examples directory during the installation. This guided session is also an example of how pressure interference can lead to an erroneous diagnostics. In order to make the demonstration clear, the data is, once again, synthetic (simulated). It is assumed that the user is fluent with the use of the functionality taught in the three previous guided sessions.

B04.1 • Starting the New Interpretation This oil well test comprises a 24 hour production period, a 48 hour build-up and then a 24 hour production period with a reduced rate followed by a 72 hour build-up. The well test was performed on a well in a producing reservoir. Select “New” in the “File” menu and go through the initialization procedure, accepting the default values for the various parameters. When the initialization is over, click “Create >>”. We are now in the main interpretation page with the main control panel. Go to “File” menu to save as Sapb04 in the example sub directory.

B04.2 • Loading the well test data Click

to start the load rate process, load the file Sapb04.rat and accept all the default values as indicated

in guided session #1. Click

to start the load pressure process, load the file Sapb04.asc and proceed as before.

B04.3 • Analyzing the build-up#1 data Click

to call the “Extract dP” process, select “list” and choose to load all the flow periods, accept the

proposed P @ dt=0 for each flow period and all the defaults values.




Doc V3.20.00

B04.3.1 • Analyzing the first build-up In the “Group” drop list select build-up #1:

Fig B04.1

Click on

to generate the automatic homogeneous infinite model. Refine the match by dragging the match

and re-execute the automatic model.

Fig B04.2 The infinite homogeneous model matches with the pressure data and derivative for 4 hours, but the data exhibits

an

increasing

derivative

behavior

that

can

be

diagnosed

as

a

possible

limit

effect.




Select

Doc V3.20.00

in the control panel to call the model dialog and add a sealing fault. Use the pick option and

indicate the start of the limit effect around 5 hours. Click on generate. Click on Shift

(automatic improve) to

adjust the match. The result is a quite satisfactory match, as shown below.

Fig B04.3 Click on the right mouse button in the area plot and select “results” in the pop-up menu to show the parameter values. The result of this analysis is the presence of a sealing fault at some 400 ft. B04.3.1 • Analyzing the last build-up In the “Group” drop list select Build-up #2: the current model is not so good:

Fig B04.4




Doc V3.20.00

Click on the “New” tab to open a new analysis, adjust the pressure match in order to adjust it to the final derivative stabilization and click on

to generate the automatic homogeneous infinite model:

Fig B04.5 The build-up pressure derivative exhibits a shape that suggests a heterogeneous system behavior. Select

in the control panel to call the model dialog and select the dual porosity PSS reservoir model. Use

the pick option to indicate the shape and the position of the transition. Click on “Generate”. Click on Shift

(automatic improve) to adjust the match. The result is a good agreement:

Fig B04.6




Doc V3.20.00

The conclusion is that both build-ups can be interpreted but with different models and reservoir parameters, attempts could be made to match both with the same homogeneous (or heterogeneous) model and then to explain the reasons why one, or the other, does not match perfectly.

B04.2 • Taking into account other wells with the 2-D Map B04.2.1 • Reservoir production history When the well was tested, the reservoir was not at rest and a few operations were performed on other wells near the tested one: One well, 1000ft east, was opened one hour after the tested well to produce 400 BPD for 24 hours and then closed in for a long duration build-up. A third well, 500 ft north, 500 ft west was produced for 4 hours at 300 BPD two hours before the second buildup.

B04.2.2 • Adding other wells to the reservoir map We are going to add to the model the influence of these other wells using the 2-D Map facility. Click on the “2-D Map” tab to activate the reservoir map page. The map shows the active well at the origin of the axes (coordinates 0,0). Click on the right mouse button in the map area. Select “Create Well’ in the pop-up menu, the cursor becomes a yellow spot representing the new well. Drag the new well and drop it at the correct coordinates: X= 1000 ft, Y= 0 ft, the coordinates are displayed at the bottom of the screen in the information bar. Repeat the operation to add a third well and drop it at: X= -500 ft, Y= 500 ft

Fig B04.7




Doc V3.20.00

Positioning these wells at their exact coordinates is not easy. In order to save time we recommend you position approximately, then to double click on the well. The “New well” dialog box is displayed where you can input the exact coordinates values:

Fig B04.8 This dialog also allows loading of the well rate history by clicking on the “Production” tab. Proceed with the rate load procedure and load Sapb04a.rat for the Well#1. Click “OK” to validate your choice. Repeat the same operation with the Well#2 and load Sapb04b.rat.

B04.2.3 • Generating the multiwell model Click on the “New” tab to initiate Analysis 3. Click on

in the control panel to call the model dialog, select a homogeneous infinite model and check “Add

other wells”, then click on “generate”. Click on Shift

(automatic improve) to adjust the match.

Display all the build-ups and activate the log-log plot:

Fig B04.9




Doc V3.20.00

The same homogeneous infinite acting model matches with both build-ups. Displaying all the flow periods would show a complete agreement demonstrating that the apparent limit or heterogeneity effects were due to the influence of interfering wells.






Doc V3.20.00

B05 – Saphir guided interpretation # 5: Gas test This guided interpretation is a modified isochronal test on a gas well. The well was flowed for 10-hour periods at 2000, 4000 and 6000 Mscf/d, each flow period being followed by a build-up of 10 hours. After another 10-hour flow at 8000 Mscf/d the well was choked back to 7500 Mscf/d and flowed until stabilization before being closed for a final build-up of 24 hours. It is assumed that the user has followed the Guided Interpretations # 1 to # 4.

B05.1 • Starting a new interpretation Start a new Saphir gas project and follow the default path to enter the required PVT, well and reservoir characteristics. The file Sapb05.asc was copied during the program installation to the Example directory, and is the only file that is needed to start this interpretation. The file is the pressure file and we will proceed to load the file as usual through the control panel “Interpretation” Load P. We should have the resulting pressure plot as shown below.

Fig B05.1 We have not loaded a flow-rate and one can observe that Saphir has assigned a “zero” flow-rate to the duration of the pressure history. In this case we have no flow-rate file, thus we will have to build the history from a written well test report. This is the case for many well test interpretations and aim to illustrate the procedure adopted in Saphir.




Doc V3.20.00

B05.2 • Defining the flow history

Choose the tab

.

In the Edit Rates page, “Split a rate” Select the Split icon

and” Actions” “Add a rate”

are enabled in the toolbar.

and position the cursor at the end of the first drawdown and click.

Fig B05.2 In the dialog you define the flow rate before and after the split. Enter 2000 Mscf/d before and 0 Mscf/d after the split. The rate synchronizing buttons

“Move rate to cursor position” and

“Move rate to nearest

pressure” are now enabled in the toolbar. Then make another split at the end of the first build-up, 0 Mscf/d before and 4000 Mscf/d after the split. Rates

Continue until the flow rate history is complete. Follow the schedule to the right.

Mscf/d Zoom in and adjust the start of all flow-periods using the “Move rate to nearest pressure”

. The

2000

figure below illustrates the final pressure and flow rate history before extracting the delta P of the last

0

build-up.

4000 0 6000 0 8000 7500 0

Fig B05.3




Doc V3.20.00

B05.3 • Interpretation

Extract Delta P

for the final build-up and generate the automatic model

.

A good log-log and semi-log match is obtained but the simulation is not consistent with the data during the first four drawdowns. The automatic model generates the entire response with a constant skin equal to the skin during the build-up that has been analyzed. The discrepancy in the simulation is indicative of rate dependent skin.

Fig B05.4

The rate dependant skin parameters can be established by making a semi-log analysis of all the build-ups and/or drawdowns. We will proceed and analyze all the build-ups. Choose the List option from the Group drop list:

Select all the build-ups.




Doc V3.20.00

Fig B05.5

Maximize the log-log plot and, using the button “Line/Single”

, click once for the start of the regression

interval and again for the end of the interval. The line will be regressed on the reference group, which is in this case “build-up # 4”, the log-log plot automatically shifts to the semi-log (superposition) plot and the results and the semi-log straight line are displayed.

Fig B05.6 One could proceed to “draw” the other straight lines on the semi-log plot but it is recommended that the lines be drawn on the log-log plot to be sure that the IARF period is chosen correctly. Thus we switch to the log-log




plot

Doc V3.20.00

. Choose another reference group through the button

, and regress the straight lines on each

group. Most of the options in the plot menu bars are available through a pop-up menu, which is available by a right click in the plot.

Fig B05.7 The Skin vs Rate plot is automatically generated as soon as two straight lines have been drawn. When additional straight line are drawn, it is necessary to update the line of Skin vs Rate by pressing the right mouse button on the plot and selecting “line/reset” in the drop menu. When the “Skin vs. Q” plot has been deleted it is possible to recall it by pressing

in the “Interpretation

(2) control panel. The resulting plot is shown maximized below with the results. We have now four points in the relationship S’ = S0 + Dq and the rate dependency can be established.

Fig B05.8




Doc V3.20.00

The results can be input in the current model if rate dependant skin is selected.

Fig B05.9 The resulting history match is illustrated as follows:

Fig B05.10 It is possible to regress on the history (simulation) plot. Select “Improve”

. Check the radio button

“Improve on simulation”, we can now improve on both the rate dependant skin relationship and initial pressure.

Fig B05.11




Doc V3.20.00

In case the transient analysis of each flow period does not result and a satisfactory Skin versus rate plot cannot be constructed (which is often the fact with real data), then we can approach the problem of rate dependency as follows: We guess a value for dS/dQ, let say, 0.002 Mscd/d-1; this value is typically a small number. We analyze the final build up to get a value for the skin using the automatic model option. We call the model catalog and choose rate dependant skin, input the guessed value for dS/dQ (Fig B05.12) and use the calculator to adjust the skin at zero (no turbulence).

Fig B05.12

Click on the calculator

and choose the button

to adjust the skin to the skin found

during the analysis of the final build-up (S = 11.5576 in this case), then click on

to update the

model skin at no turbulence to the resulting value using the relationship S’ = S0 + Dq.

Fig B05.13 Finally the values that we are going to use to calculate the model has been updated and we can generate the model.

Fig B05.14




Doc V3.20.00

We can then run the regression on the Simulation again and a good match is obtained without having to generate the Skin versus rate plot.

B05.4 • Inflow performance relationship (IPR) The modified isochronal test is carried out to determine the inflow performance and absolute open flow potential of a well. For gas tests, in addition to the Darcy and m(P) relationships, multi-rate IPRs are incorporated in Saphir for the C and n method and Jones’, also known as LIT or a & b. The IPRs are accessed by the “IPR/AOF” button

in the “Interpretation (2)” control panel. Select Vertical

well IPR and LIT. The IPR dialog is then displayed.

Fig B05.15 Set the type of test in the top left of the dialog, to Modified isochronal. Then select the pick option in the test points dialog at the top right. The flow-rate history plot is opened and you should click in each of the drawdown periods, but not in the Extended flow period at this stage.




Doc V3.20.00

Fig B05.16 Click OK to leave the dialog and the flow-rate/pressure test points table is displayed. From the “picks”, the program has retained the flow-rate during the period, the lowest flowing pressure recorded during the drawdown, Pwf, and the highest shut-in pressure, Pws, before the drawdown. When the Test points table is in the

mode, the pressure data and flowrates may be edited,

inserted or deleted.

Fig B05.17 The flowing pressure for the Extended flow period is now required. Go to the Extended period dialog in the bottom right and select pick, the flow-rate history is displayed again and the pick should be in the Extended flow period. The program retains the flow-rate and the last flowing pressure. Finally the average pressure must be entered and since this is an infinite system, a Pi of 6000 psi should be used.




Doc V3.20.00

Fig B05.18 The IPR is plotted with its “natural” scale, in this case (Pavg2-Pwf2)/Q versus Q. By selecting “Change scale”

in toolbar, the IPR can be changed to a linear plot of Pwf versus Q

(deliverability curve).

Fig B05.19




Doc V3.20.00

B06 – Saphir guided interpretation QA/QC

B06.1 • Introduction The following guided session will use files installed during the Saphir installation in the Examples directory. The files Sapb06U.asc and Sapb06L.asc will be used. An Excel Spreadsheet that is useful for the analysis is furnished under the file name Sapb06.xls. The pressure files are the same as those published in one of the examples in SPE 24288, referred to in this manual. The files are data couples of elapsed time and corresponding pressure. It is assumed that the user has at least been through the Guided Saphir Interpretations #1 and #2. Launch Saphir and start a new project using the default path. Standard test, oil and default PVT, well and reservoir parameters. Click on the QA/QC tab in the control panel and proceed to load the file Sapb06U.asc. Click on The gauge attributes should always be completed fully, it is particularly important that a proper gauge name is given as this will identify the data when various channels are displayed in the plot window. In the example below the channel name is “Upper”, however it is good practice to use the serial number of the pressure gauge, even though this is repeated.

Fig B06.1 Load the “Lower Gauge”, Sapb06L.asc, the plot window should resemble Fig B06.2




Doc V3.20.00

It is the “Difference Channel” that will be an indicator if further investigation will be performed, and the general rule is: If one anomaly has been noted, then it is highly likely that others are present but then may be masked by coarse time scales. Generate the difference, Lower Pressure - Upper Pressure is the normal convention, but Saphir 3 can generate any difference (see below). There is also a choice of how many points to generate in the channel, usually a high number is best since this gives the differences a higher definition.

Fig B06.2

Fig B06.3

Fig B06.4




Doc V3.20.00

The top plot is the pressure plot and the difference plot below is the “child plot”. Each plot can be activated by clicking on the vertical header of the plot to the left in the screen. Currently the pressure plot is activated. Using the drop list for “Reference” and “Active” the active and the reference gauge can be swapped. In order to better understand the sequence of events of this field example, it is necessary to highlight a few important facts: 1.

The shut-in was made with a mechanical “Downhole Shut-in Valve”, set in a completion nipple some 30 meters above sandface.

2.

The pressure gauges are high accuracy memory recorders (HP’s) and were suspended in tandem below the shut-in valve.

3.

The complete tool string was run on slickline wireline.

4.

Pressure equalization after the end of the planned build-up was achieved by mechanical wireline.

5.

The well produced sand during the test and sand had settled out above the shut-in valve during the build-up, thus it was not possible to immediately pull the tool string out of the hole. A period of cleaning and bailing was necessary, and we have this long static period after the equalization of the shut -in tool.

The flowing pressure was unstable and periodic, “slugging” of the pressure was caused by long flow lines between the wellhead and the gathering station (onshore environment).

Fig B06.5

B06.2 • Synchronization and anomaly detection The observed pressure difference between the gauges appears stable during the build-up, however after the equalization of the shut-in tool there is a definite change in the difference value. This may indicate that the fluid between the gauge sensing points changed and justifies further investigation to determine if any phase segregation could have taken place during other parts of the build-up, which would have caused those parts to be invalid for transient analysis. It is evident that the stable difference during most of the build-up reflects no changes in the fluid (no phase segregation), however it is necessary to investigate what could have happened during early time. Due to the




Doc V3.20.00

coarse time scale it is necessary to enlarge the early time build-up event. However, before we do so we will synchronize the two data channels. By enlarging an event, such as the start of the build-up, we can readily observe that one gauge is lagging behind the other, (see below for illustration of this).

Fig B06.6 The synchronization can be done automatically in the current zoom window. However in this case we will use “mouse synchronization” first. Make sure the Active gauge is the “Upper” in the “Active” droplist. Click on

or use the pop-up menu

available with a right click in the plot window: “Shift Active”>>”Mouse”. Click on the Upper gauge and holding the mouse button down drag it on top of the Lower gauge. See below.

Fig B06.7




Doc V3.20.00

When you shift a data set with the mouse you cannot avoid to also shift in the Y direction, thus it is critical to reset the Y (pressure shift) in order to retain the correct difference value. Right click and chose the menu “Shift Active”>>”Reset Y”. We can now check if the synchronization is satisfactory by zooming in further (see below), and then running an automatic shift, which executes non linear regression in the current zoom window in order to minimize the difference.

Fig B06.8 Make sure the Upper gauge is the Active gauge and the Lower gauge is the Reference gauge in the drop lists. Click on

or right click in the plot window and choose the menu “Shift Active”>>”Automatic”. The automatic

shift will take place (Hint: If this does not seem to work, the current window may contain too much data and it may be necessary to zoom in on the event further in order for Automatic shift to work). The synchronization is finished and it is now time to look for any phase segregation and anomalies. If we go back to the default scale we can see that now the difference channel is much smoother.

Fig B06.9




Doc V3.20.00

We will now enlarge the early time data of the build-up with its difference channel. We can see from the plot that phase segregation takes place after a short time of stabilization, just after the well was shut-in by the downhole valve. Double click in the difference plot to maximize it and zoom in on the early build-up.

Fig B06.10 During the segregation the gauge pressures are affected by this wellbore phenomenon and cannot be used for transient analysis. The fact that this takes place at early time aggravates the implication as the diagnostics are done in log-log coordinates and a short time may represent several early time log cycles, therefore leaving the interpreter with very little valid data at intermediate and late time due to the compression of the data.

B06.3 • “Differential Pressure Analysis” B06.3A • Principle and Background The analysis is based upon the difference in pressure measured between tandem pressure gauges (the simplest case), or a combination of pressure differences if more gauges are used during the survey. The study of these differences can reveal the following problems and has a direct impact on the choice of the data measurements to be used for transient analysis: •

Phase segregation in the wellbore

•

Fluid interface movements (oil, gas and water)

•

Temperature anomalies affecting the pressure gauge and / or identification of gauges with technical problems, such as:

•

Pressure gauges outside of claimed accuracy and resolution specifications

•

Gauge drift

•

Gauge battery running out

•

Other technical or electronic malfunctions




Doc V3.20.00

By convention the pressure difference between gauges is calculated so that an increase in the “difference channel” represents an increase in the fluid density between the gauge sensing points, and a decrease represents a reduction of the fluid density. The “difference channel” behavior will be identical whatever the gauge offset may be (the upper gauge may well read a higher pressure than the lower gauge, possibly due to a gauge problem, but the “difference channel” would have the same identifiable shape). The simple analysis is based upon the study of the pressure and temperature differences between two gauges placed in the test string at different depths. The figure below illustrates schematically what happens at the pressure sensors of two sensing points, if a ‘gas-oil’, ‘oil-water’, ‘gas-water’ or a mixed interface is moving downwards.

) p o t P t o b P ( e r u s s e r p l a i t n e r e f f i D

Upper gauge

Ptop

Lower gauge

Pbottom TIME

The example assumes that any “background” behavior is following a constant transient or is in “pseudo-steady state”. Once the interface hits the “upper sensor” the pressure at this sensing point remains constant as the interface moves towards the lower pressure point. The pressure at the “lower sensor” declines linearly if the fluid interface movement is constant, and becomes constant again after the interface has moved below the lower pressure point. The difference in pressure between the two sensing points follows the difference in fluid gradient between oil and gas.

B06.3B • The guided interpretation The “Differential Pressure Analysis” does not stop with a qualitative evaluation of the phenomenon. It is now necessary to establish what fluid is present at what time. This will give an estimate of the quality of the measurements. To do this we will need to determine the delta-p for several events, and use a special spreadsheet constructed for this purpose. As shown before a stabilized difference channel is always indicative of no change in the fluid density between the gauge sensing points.




Doc V3.20.00

In order to illustrate the changes in fluid density that took place in this particular test we will select events and use the QA/QC plot to read the values in time and delta-p. The figures below illustrate the choice of events.

Fig B06.11

Fig B06.12 The events have been numbered from 1 to 6, and the values for each event correspond to the entries in the spreadsheet shown in Table 1. The values are simply read off the difference channel using the option “Pick Point”

or the mouse pointer. The values are not exact so “eye balling” the value is justified which again

justifies slight changes in the values in order for the “Differential Pressure Analysis” to converge to an explainable and consistent result.




Doc V3.20.00

Each delta pressure value corresponds to a particular fluid density, or mixture density, of the fluid present between the gauge sensing points. The first event, taken when the well is on a high flowrate, caused by the added component of friction in the annular space created in the tubulars due to the placement of the gauge. When the well is shut in the difference drops and stabilizes at a certain level, after equalization this level increases and stabilizes again. At early time we have identified phase segregation. Thus in sequence we can determine the following: 1.

Event 2 and 3: Constant fluid density

2.

Event 4 and 5: Constant fluid density, but lighter than Event 2 and 3

3.

Event 6: Constant fluid density, heavier than Event 4 and 5

Therefore we have three possible scenarios as no gradient survey was run at the end of the test and we can only guess at what fluid was present before pulling out of hole. Scenario 1:

Event 1: Flowing Event 2 and 3: Oil Event 4 and 5: Gas Event 6: Oil

Scenario 2:

Event 1: Flowing Event 2 and 3: Water Event 4 and 5: Oil Event 6: Water

Scenario 3:

Event 1: Flowing Event 2 and 3: Water Event 4 and 5: Gas Event 6: Water

In the spreadsheet, the delta-p values are entered in column 3 and the assumed liquid (column 5) and corresponding densities (column 6). An implied offset is then calculated (column 8) which is simply the difference between the delta-p value from the difference channel and the theoretical value (column 7) calculated using the assumed density and the known distance between the sensing points. This implied offset must be the same (or close) for all fluid assumption as long as no gauge drift takes place. The resulting residual difference (difference between Dp from QA/QC and the corrected delta-p (column 9) using one single value of the offset, should therefore go to zero (column 10). D I F FE R E N T I A L P R E S S U R E A N A L Y S I S EVENT

From Difference Channel Time dp

dGradient

(lower-upper)

[1]

Event Event Event Event Event Event

1 2 3 4 5 6

[2] (hr)

[3] (psi)

17.10 17.18 17.22 17.33 22.50 32.00

Assumed Gradient

Calculated dp

[6] (psi/m)

[7] (psi)

between Gauges

[4] (psi/m) 6.30 2.90 2.90 0.20 0.20 2.90

Assumed Fluid

-1.22 0.00 -0.97 0.00

[5]

Friction Oil OIl Gas Gas Oil

Implied Offset

Distance Between Sensors

2.340 1.120 1.120 0.147 0.147 1.120

Assumed Accuracy Offset:

[8] (psi)

6.51 3.11 3.11 0.41 0.41 3.11

-0.21 psi

Accuracy Offset:

2.78 m TVD

Lower Gauge = Upper Gauge - X psi Pressure Gradients Gas gradient (psi/m) Oil Gradient (psi/m) Water gradient (psi/m)

0.147 1.120 1.680

Table 1

Residual Difference

(lower-upper)

Input Fields Upper Gauge Lower Gauge

Observed dp corrected

-0.21 -0.21 -0.21 -0.21 -0.21 -0.21

[9] (psi)

[10] (psi) 6.51 3.11 3.11 0.41 0.41 3.11

0.00 0.00 0.00 0.00 0.00 0.00




Doc V3.20.00

The analysis will only work with the right assumptions and Table 1 clearly shows this.

DIFFERENTIAL PRES SURE ANALYSIS EVENT

From Difference Channel Time dp

dGradient

(lower-upper)

[1]

Event Event Event Event Event Event

1 2 3 4 5 6

[2] (hr)

[3] (psi)

17.10 17.18 17.22 17.33 22.50 32.00

Assumed Gradient

Calculated dp

[6] (psi/m)

[7] (psi)

between Gauges

[4] (psi/m) 6.30 2.90 2.90 0.20 0.20 2.90

Assumed Fluid

-1.22 0.00 -0.97 0.00

[5]

Friction Water Water Oil Oil Water

Implied Offset

Observed dp corrected

Residual Difference

[9] (psi)

[10] (psi)

(lower-upper)

2.340 1.680 1.680 1.120 1.120 1.680

[8] (psi)

6.51 4.67 4.67 3.11 3.11 4.67

-0.21 -1.77 -1.77 -2.91 -2.91 -1.77

8.07 4.67 4.67 1.97 1.97 4.67

1.56 0.00 0.00 -1.14 -1.14 0.00

Input Fields Upper Gauge Lower Gauge Distance Between Sensors

Assumed Accuracy Offset:

-1.77 psi

Acc uracy Offs et:

2.78 m TVD

Lower Gauge = Upper Gauge - X psi Pressure Gradients Gas gradient (psi/m) Oil Gradient (psi/m) Water gradient (psi/m)

0.147 1.120 1.680

Table 2 Table 2 illustrates that with a different assumption we see a spread of the implied offsets, thus the "Differential Pressure Analysis” will not work.

B06.4 • Conclusions We determined the validity of the data for transient analysis, and identified the fluid phases present in the wellbore at different test times. We were able to determine a true gauge offset which is a direct measure of the quality of the tools, and obviously this offset will have to be within the accuracy band claimed by the manufacturer. The identification of the fluid present at the end of the build-up is critical for proper correction of the average drainage area pressure to datum. When an offset is obviously outside of expected specification, caution must be adapted with respect to calculated reservoir pressures from the gauge responses and interpreted pressures values (Pi and Pbar) from the subsequent transient analysis.




Doc V3.20.00

B07 – Saphir guided interpretation : 2-D Map and Numerical model (1)

B07.1 • Introduction The following guided session uses the file FieldMap.bmp installed during the set up of Saphir. It is assumed that the user has at least been through the Guided Saphir Interpretations # 1 through # 5. Launch Saphir and go through the process of starting a new project using the default path. Standard test, oil and default PVT, well and reservoir parameters. This example is for Saphir advanced users only.

B07.2 • 2 D-Map and Voronoi Click on the tab “2-D Map”. The “Tested well” has been defined in the middle of the default rectangle and the coordinates of the well is at (0,0).

Fig B07.1 Double click on the tested well and the well radius can be set and the coordinates can be confirmed at (0,0)




Doc V3.20.00

Fig B07.2 The second tab “Production”, is where the rate history of the “Tested well” can be defined. This rate history can be loaded from a file or entered from the keyboard as normal.

Fig B07.3 The rate history can also be loaded through “Load Q” in the Interpretation Page

.

At this point, we will not define a rate history, as we will proceed to illustrate some of the features of the 2-D Map. In the the tool bar click click on the 2-D Map area :

to load a bitmap. This menu is also available through a right click of the mouse in




Doc V3.20.00

Choose “Load another bitmap” and load the bitmap “FieldMap.bmp” from the Examples directory.

Fig B07.4 The tested well can be moved to any of the indicated wells on the map. Just hover the mouse pointer above the well, and click and drag it to the well that is the tested well on the map. In our demonstration we will move the tested well to the bottom middle well called P01.

Fig B07.5 We can now define the other wells in the field by using a right click to access the appropriate menu or the add well button

. We add all the wells which are numbered from Well #1 to Well #n.




Doc V3.20.00

Create the well on the well P03 as a fractured well. To do it, click on the button

.

Click once to set the position of one end of the fracture and a click a second time at the position of the second fracture end. You can adjust the position of the well, just by a drag and drop. You can adjust the length and the orientation the fracture, with a drag and drop on the end of the fracture. The fracture settings can also be modified numerically in a table (see “Well Info” below).

Fig B07.6 A double click on any of the added well will bring up the dialogue specific specific for each well well where coordinates coordinates with with reference to (0,0) can be refined and the production history of each well can be defined. Any well can be set as fractured on not and the fracture parameters (length, angle) angle) can be input in this well info info dialog. Any well can be se t as “Tested “Tested well” or declared declared to be taken into account or not in the simulation.

Fig B07.7




The contour of the field can be drawn using

Doc V3.20.00

the draw contour button. The trick is to click and follow the

contour around and clicking each time to change the direction of the boundary. The option acts as a “rubber band” and the contour is finished by a double click. Anytime a mistake is made, the ESC key will set you back to the previous point.

Fig B07.8

Fig B07.9 This map has a scale indicator, and we will use this to set the scale of the 2-D Map. Using the zoom-in feature, we zoom in on the scale indicator and use

, set scale. If the distances between wells are known, this can

also be used, and it is possible to snap the scale line drawn by the user to a beginning and ending well. Any other known distances between points on the map can also be used.




Doc V3.20.00

Fig B07.10 Enter 1000 m and click OK, this will rescale the whole 2-D Map.

Fig B07.11 This map has fault indications and it is necessary to include this in order to build the total picture. With the button

or the right mouse button, we will add the faults in a similar way to that which the contour

was defined. Hint: it is best if the fault starts at a node defined on the contour. This will ensure closure between the contour and the fault. A double click ends the fault. The faults in the 2-D Map can be sealing or have some transmissibility. Double clicking on a defined fault brings up the dialog where the Leakage factor can be changed.




Doc V3.20.00

Fig B07.12

The 2-D Map has now been totally defined and we can turn off the bitmap by using the button

.

Fig B07.13

Clicking on

will calculate the automatic Voronoi gridding, the local grid refinements for each well and the

gridding necessary to take into account the defined faults. The grid can be modified by clicking on the “Settings” button

, then the “Grid Control” tab, displaying a dialog where the user can change settings such

as the grid shape, or decide to use global or individual well settings for the gridding. The Voronoi gridding is the basis for the numerical solution of the pressure at the Tested Well and solves the influence of the tested well on itself and the influence of any interfering well on the Tested Well.




Doc V3.20.00

Fig B07.14 By checking out the “Automatic” option, the Grid control allows also to change the type of grid set around the fracture. It can be either pseudo radial (default) or elliptic, a zoom on the fracture well shows the difference:

Fig B07.15: Pseudo Radial fracture grid

Fig B07.15: Elliptic fracture grid

B07.3 • Using the Voronoi model The previous sections described how to build a problem using the 2-D Map facility. Some features, such as the multiple-well options are usable by both the analytical and numerical solutions. The others can only be modeled using the Saphir numerical model, called “Voronoi” from the gridding it uses. The following section describes the generation of a Voronoi model. Let us start a new interpretation using the default values and going to Saphir’s Test Design option. Start a new Saphir project and choose the first button in the Interpretation (2) page, model with the well in the middle of a rectangle of 10,000 ft x 6,000 ft.

. We will define an analytical




Doc V3.20.00

Fig B07.17 To define the production history click on

and enter a production of duration 4,000 hrs at 1,000 stb/d and a

build-up of 2,000 hrs.

Fig B07.18 Generate the analytical solution, and extract the production period to create the diagnostic plot with the appropriate derivative. Rename “Analysis 1” to “Rectangle” by clicking on the tab.

Fig B07.19




Doc V3.20.00

Pseudo-steady state is reached at around 100 hours drawdown. We will now proceed to analyze the analytically calculated response numerically, using the Voronoi model. Click on 2-D Map. We have to click on the square contour, choose “set as a rectangle”, input the rectangle size and a click on

will display the Voronoi grid.

Fig B07.20

Click on the “Rectangle” tab and choose “Model” button in the interpretation page

. This time click on the

“Numerical” tab. Select ‘pressure fields’ by checking the check box.

Fig B07.21 Click on

to set the pressure field time steps. Check “Display during generation” and choose ‘Control output

by time’ = 120 hrs to generate 50 pressure fields. Generate the numerical model. A perfect match between the numerical and analytical solution is found.




Doc V3.20.00

An animation of the pressure fields is played during generation of the model in the geometry plot.

Fig B07.22

Maximize the geometry plot, and click on the button

to advance to the first pressure field.

Fig B07.23




Doc V3.20.00

It is possible to play back the pressure field animation manually or automatically. The user can also decide which field to display by using a pick option.

Show first field Show previous field Show next field Show last field Stop animation Run animation Run animation loop Click on

to edit the settings.

Fig. B07.24

Choose or confirm Shading, Pressure and Interpolated in the Display Tab. Choose 7 colors and click on choose the method of min-max of the color scale, choose “All fields min-max”. Then click on

to run the animation. The figure B07.23 illustrates the animation at field= 3360 hrs.

to




Doc V3.20.00

Fig. B07.25

The animation can also be seen in 3-D. Click on

.

The figure below illustrates the same time field as Fig. B07.23, just before the shut-in.

Fig. B07.26

Click on

to switch back to 2-D.




Doc V3.20.00

Click back on the 2-D Map tab, and click on the contour of the rectangle. We see that nodes appear and it is possible to drag one of the nodes close to the well, thus making the rectangle smaller.

Fig B07.27 Make a new analysis based upon the first and call this analysis Small Rect.

Fig B07.28 Generate the Voronoi model again (uncheck pressure fields), and we see clearly the influence of the proximity of the boundaries and that pseudo-steady state is reached earlier due to the smaller reservoir volume.




Doc V3.20.00

Fig B07.29 In order to illustrate the functionality of the Voronoi model we will add a fault. Click back on the 2-D Map tab and draw a fault as illustrated in the graphic below after having dragged the node back to approximately where the original rectangle was.

Fig B07.30 Define a new analysis called “One Fault”, and generate the response.



Saphir V.3.20 - © KAPPA Engineering 1990-2003 WARNING: We

Doc V3.20.00

are not suggesting that this is a realistic fault ! We are merely demonstrating a facility with an

extreme behavior. However, it should be noted that such a configuration can be extended to multiple connecting tanks. This can be achieved by implementing composite zones, which will be demonstrated in the next Guided Saphir Interpretation.

Fig B07.31 The influence of this added fault and the fact that the diffusion has to go through a small opening where we are starting to “see” the bigger reservoir before pseudo-steady state is reach is very evident in the calculated numerical model. We can add another fault to illustrate the functionality further. Go to the 2-D Map and draw a second fault as illustrated below.

Fig B07.32




Doc V3.20.00

Define a new analysis and call it “Two Faults”. Generate the Voronoi model. We can see that the channel caused by the two concentric faults also shows up in the model as a half slope.

Fig B07.33 Finally we will add an other well to illustrate this feature. In the 2-D Map we will add a well with the following injection history 2,000 hrs 0 (production), 3,000 hrs at -1,000 BOPD (injection), 1,000 hrs build-up. Right click in the 2-D Map and choose the option “Create well”, or click on the button

in the toolbar.

Position the well inside the first channel and click to place it. Double clicking on the well after it has been positioned will bring up the dialog where the coordinates of the well can be changed and the production can be entered.




Doc V3.20.00

Fig B07.34 Call the new analysis: “Add Well” and now generate the Voronoi model and add the new well to the numerical solution.

Fig B07.35

Fig B07.36 The influence from the additional well can be clearly observed. Maximizing any of the plots and clicking on the

button enables you to visualize all the generated solutions.




Doc V3.20.00

Fig B07.37






Doc V3.20.00

B08 – Saphir guided interpretation: 2-D map and Numerical model (2)

B08-1 • Starting the session

Create a new session accepting the default values, select 2D map, load the Bitmap FieldMap.bmp, draw the contour and the faults, set the tested well in P01 and the other wells (see session B07.2) Note: for the purpose of this session, do not set the scale and keep the default one. Save the session under Sapb08. Input the rate history through the keyboard: 1000 hours of 1000 bpd production rate followed by a 1000 hours build up.

B08-2 • Basic pressure behaviour

Go to the control panel interpretation (2) and select “Test design”. Select numerical tab. Press generate in the “Test design” dialog and extract the first production period when the simulation is completed:

Fig. B08.1 Go back to the main screen and save the session.




Doc V3.20.00

B08-3 • Composite Numerical

The composite Numerical Saphir facility allows the user to introduce a certain level of reservoir heterogeneity by defining regions where different values of the reservoir characteristics can be entered. B08-3.1 • Radial composite well option Go to 2D map and double click on the “tested well” symbol.

Fig. B08.2 Tested well information box Check the “radial composite gridding” and input 500 ft as interface distance Ri. Press OK. Clicking on the

button, drag and drop the anchor sign just outside the circle surrounding the well:

Fig. B08.3 Composite anchor setting




Doc V3.20.00

Double-click on the Composite anchor and change the name to “Outer zone”. A specific set of reservoir characteristics will be assigned to the zone outside the circle.

Select the tab Analysis and the Control Panel “Interpretation” and click on the Select the “Numerical” tab: In the Numerical dialog box input 0.1 as D and M values:

Fig. B08.4 Numerical model dialog box Note that M and D follows the zone color in the 2-D Map. Then, press “generate”. Maximize the log-log plot.

Fig. B08.5 Log-log Composite Numerical model

Model button.




Doc V3.20.00

The Composite Numerical model shows on the log log plot the divergence from the homogeneous behaviour due to the change of mobility at 500 ft from the well. Note: in the 2-D Map, the “white” zone corresponds to the pressure and time matches and is the reference. The M and D ratio are then: (white zone condition)/(colored zone condition). Setting the Composite anchor outside of the circle surrounding the well defines the composite system with the same convention as in the analytical model, the M and D ratio are: (inner condition)/(outer condition). B08-3.2 • Composite reservoir zonation Select the “2-D Map” tab.

Click on the

icon to create a fault.

Draw a fault isolating a part of the map (i.e. the West part as Shown in the fig.B08.6), end with a double-click. The isolated zone is now in white (no specific parameters values).

Fig. B08.6 Zoned 2-D Map Click on

to create a “composite anchor” drag and drop it into the isolated zone. The zone changes color.

Double-click on the “Outer zone “ anchor and change the name to “East zone”, double-click other anchor and change the name to “West zone”.




Doc V3.20.00

Fig. B08.7 Zoned 2-D Map Double click on the fault, and click on the composite limit icon

(yellow and green) in the dialog:

Fig. B08.8 Fault information dialog box The fault is now a composite limit and its transmissibility (leakage) can be specified, press OK:

Fig. B08.9 Zoned 2-D Map Repeat the same procedure to create another zone in the North corner:




Doc V3.20.00

Fig. B08.10 Zoned 2-D Map

We have now a radial composite well in a composite reservoir. Create a New analysis from analysis #1.

Click on the

model icon. The dialog is still the Numerical model dialog:

Fig. B08.11 Numerical model parameters input dialog box

The yellow zone corresponds to the radial composite well status. The green zone corresponds to the “North” region. The pink zone corresponds to the “West” region. In this case, we will have the following permeability contrasts if we assume that the fluid and porosity to be the same in all zones:




Doc V3.20.00

Note that M and D follows the zone names and colors in the model dialog and in Fig. B08.10. The 2D map tool bar allows zooming in, zooming out, hiding or displaying the text, the grid and the composite zones. Well zone (white):

kh = 1000 md.ft

Channel zone (yellow):

kh = 10000 md.ft

West zone (pink):

kh = 100 md.ft

North zone (green):

kh = 50 md.ft

Press Generate.

Fig. B08.12 Composite Numerical Model We can observe that the radial composite effect occurs before the initial radial flow, then the multiple faults effect and the closed sysem behaviour (Pseudo Steady State) are superimposed . B08-3.3 • Fissured composite reservoir zonation Create a New Analysis from analysis #2. In order to keep this session at reasonable level of complexity, go to 2D map, double click on the tested well and uncheck the radial composite option.

Remove the Composite Anchor “East Zone” by clicking on the accept to delete it. The composite zone color are back to the “3 zones” color pattern:

icon in the tool bar, then on the anchor,




Doc V3.20.00

Fig. B08.13

Go to analysis #3 and press on the

model icon. The dialog is still the Numerical model dialog:

Fig. B08.14 The North (pink) zone and the West (yellow) zone kept the same D and M ratio values. The default reservoir model is “homogeneous”, it can be changed to 2 porosity PSS model. Select the 2 porosity PSS for the three zones and change the Omega and Lambda value to, respectively: Keep the default for the reference (white) zone. Note : it can be changed with the “pick” option. ω =

0.01 and λ = 1e-7 for the West (yellow) zone

ω =

0.001 and λ = 1e-8 for the North (pink) zone




Doc V3.20.00

Fig. B08.15 Press generate:

Fig. B08.16 We can observe the 2 Porosity transition of the first zone and the North and West zonations act like a reduction of the reservoir area on the Pseudo Steady State.

B08-4 • Constant pressure boundary

Create a new analysis from analysis #3. Go to 2D map, delete the composite limits and the Composite anchors by clicking on the icon then on the item to delete.

and picking




Doc V3.20.00

The reservoir is is now homogeneous and the map is is totally white. Double click on the contour. The field contour properties dialog is displayed, selmect the “Sealing vs Ct Pressure” tab:

Fig. B08.17

Check “segment in sequence” , press on the icon

, the Field contour dialog disappears and the cursor

changes. Click on one segment of the contour, this segment turns to blue and is then to constant pressure, repeat the same procedure for each segment you want to switch to constant pressure. Change the segments as indicated in the circle on Fig. B08.18. Press to stop the sequence.

Use the same procedure to switch back a segment to No-flow by pressing the icon Exit by pressing OK.

Go to analysis #4 and press on the

model icon. The dialog is still the Numerical model dialog.

Select “Reservoir Model” as “Homogeneous” and:

Fig. B08.18 Press on Generate:




Doc V3.20.00

The pressure curve exhibits clearly the constant pressure effect at late time.

Fig. B08.19

Note: an additional drawdown derivative curve is automatically calculated and plotted as soon as a constant pressure limit is used. This is done in order to facilitate the diagnosis by allowing the comparison between the multirate multirate derivative to the drawdown derivative. derivative.

B08-5 • Pressure fields generation and display

The pressure value can be b e calculated at a t each cell of the gridding in order to visualize the pressure distribution change during the well test or the simulation: simulation:

Press on the

model icon. Numerical model dialog is displayed. Check the “pressure field” option:




Doc V3.20.00

Fig. B08.20

Press on the setting icon

to set the pressure field calculation parameters:

Fig. B08.21 Ask for the pressure fields to be generated during the simulation simulation in order to follow the evolution and set the “Control Output by time” period to 40 hours. Press OK and then generate: generate:

Fig. B08.22 The Geometry Plot, generated during the simulation, shows very clearly the effect of the constant pressure limits. Other well pressure history: history : When the “pressure field” field” option was selected, double clicking clicking on an “other well” displays the pressure and rate history of this one. Double click on the Geometry Plot header to maximize the view.




Doc V3.20.00

The pressure maps can be visualized or modified, using the top tool bar or the drop menu displayed when right clicking on the geometry plot:

The commands work like in a standard movie recorder.

Go to the first field and run the animation by pressing

and then

Fig. B08.23

Press on

to set the view parameters:




Doc V3.20.00

Fig. B08.24 The displayed item can be selected and the result can be previewed without getting out of the setting panel by pressing “Apply”. Let’s visualise the various pressure representation. Select “Solid” and press “Apply”. The pressure is represented as a plain color in each cell:

Fig. B08.25

Select “Isovalue” and press on the icon setting “Apply”:

, set the isovalues spacing to 10 psi, press OK and then




Doc V3.20.00

Fig. B08.26 Go back to “Interpolated”, press “Apply” and select the “Color Scale” tab:

Fig. B08.27

The default color scale is Select “7 colors”, Press on

to adjust the coloring: the default scale is “all field min-max” based on the min-max of all the

calculated fields during the simulation.

Fig. B08.28 Select “track compressible zone” which is a scaling based on the calculation of the radius of investigation, its objective is to concentrate the coloring around the zone of maximum pressure gradient it focuses on the radius of investigation.




Doc V3.20.00

Press “Apply” :

Fig. B08.29

The option “Center on initial pressure” set the color scale between a percentage below and above the initial pressure, which can be useful only in case on injection/production. Keep the default, and select the tab “Time settings»:

Fig. B08.30 The selected is presently the final (2000 hours), change it to a field at the end of the production period by pressing the “pick option”

:

Pick at the end of the production period:




Doc V3.20.00

Fig. B08.31 Press “Apply”:

Fig. B08.32

Go to Color Scale tab, press on dialog:

, select “All fields min/max», accept, Press OK to exit from the setting




Doc V3.20.00

Fig. B08.33

Press on

to switch to the 3D view:

Fig. B08.34 The orientation and the aspect can be changed either with the mouse (drag and drop) or with the thumb wheels at the bottom.

Press on

display or hide the Y-axis scale (pressure in this case).

Go to the first field and run the animation by pressing

B08-6 Field data input and display

and then




Press on

Doc V3.20.00

to switch back to the 2D view:

Fig. B08.35 Go to the 2D map screen .

Press on the

“edit data field” icon :

Fig. B08.36 Select “thickness” and press on “load” , answer “Yes” when the dialog ask if the tested well is in coordinates (0,0), and load the file Sapb08-thickness.asc:




Doc V3.20.00

Fig. B08.37 Press on “OK”. Click on

to set view parameters. Choose display tab and select ‘thickness’.

Press on “Apply” to visualize the thickness color map:

Fig. B08.38

Exit by pressing OK and press again on

to set the view parameters. Select the tab “Data Interpolation”:




Doc V3.20.00

Uncheck the “Automatic” option:

Fig. B08.39 The default interpolation technique is the Krigging. Select “Linear Interpolation” and press “Apply»:

Fig. B08.40 The interpolation is not better, therefore, go back to the automatic option.




Doc V3.20.00

Follow the same procedure to load the porosity map data contained in the file Sapb08-porosity.asc. Press “Apply” in the 2D data edition to visualize the porosity map, and then press OK to exit:

Fig. B08.41 Go to Analysis#4. Double click on the maximized Geometry Plot to return to the main screen, if necessary.

Press on the

model icon. Numerical model dialog is displayed. Check the “pressure field” option and

check the “Include thickness field” and “Include porosity field” to take them into account in the simulation. We can press on

to display the thickness field or on

for the porosity field:




Doc V3.20.00

Fig. B08.42 Press on “Generate”:

Fig. B08.43 Maximize the geometry plot. Press

. In the setting dialog, Color Scale, press

compressible zone». Play the animation. Display the field at 160 hours:

to access to Color Scale min-max and select “Track




Doc V3.20.00

Fig. B08.44 Press on

to switch to the 3D view:

Fig. B08.45

Press

for the setting dialog access.

The data governing the shading and the vertical scale can be specified in this dialog. Select “Thickness” for the shading and “pressure” for the vertical scale: the color then depends on the thickness and the shape on the pressure:




Doc V3.20.00

Fig. B08.46 Play the animation in 3D.




Doc V3.20.00

B09 – Saphir guided interpretation #9: Material Balance

B09-1 • Material Balance The material balance option is available to model gas tests in a closed system. It takes into account the changing fluid compressibility with respect to the depleting average pressure. B09-1.1 • Demonstration and guided visit Open the session with the file Sapb09.ks3. The pressure data has been generated with a numerical simulator and takes into account the exact PVT at the actual pressure in any point of the reservoir.

Fig B09-1 Analytical model closed square without material balance

Fig B09-1a Zoomed area of Log-Log plot




Doc V3.20.00

The data clearly exhibits pseudo-steady state behaviour of a closed reservoir. The match is made with an analytical model of a well in a 4000ft x 4000ft square closed reservoir without the gas material balance option. It is observed that the analytical model simulation exhibits a constant linear pressure decline higher than the decline given by the numerical simulation. This is due to the fact that the fluid compressibility increases when the pressure depletes. In other words, the same amount of extracted fluid makes the pressure decline less. The match is poor at late time in the Log-Log plot (production #1) and does not match the build-up at all in the History plot. Create a new analysis from the previous one. Click on the “Model” button

to call the Model dialog:

Fig B09-2 Model dialog Check the “material balance” option box and click “Generate”:




Doc V3.20.00

Fig. B09-3 Closed system model with material balance option

Note: the “2D Geometry plot” shown on this screen copy can be enabled in the “Settings” control panel, “Interpretation” icon, “Misc” tab, by checking the “Reservoir Geometry” box on. The analytical model utilizing the “material balance” calculation option now shows complete agreement with the simulator results. The History plot also displays a yellow curve that is the result of the “calculated average pressure versus time” during the extended production period. The following “P/Z vs Cumulative Production” plot shows the comparison between the results of two analytical model simulations, one using the “material balance” option, the other one without.

p/z vs Cumulative Production 6000 5000 Z / 4000 g v 3000 a P 2000 1000 0 0

10000

20000

30000

40000

Cumulative production With material balance Without material balance

Fig. B09-4 OGIP calculation plot It can be seen that without the “material balance” option, the average reservoir pressures simulated by the uncorrected analytical closed system model will erroneously underestimate the original gas -in-place.






Doc V3.10.00

B10 – Saphir guided interpretation #10: Real Time Acquisition

B10-1 • Real time acquisition B10-1.1 • Starting the real time simulation In the Saphir folder start the KDataSup version 2.30T software. The icons in the toolbar call the options which can also be accessed from the Options menu. They are, from left to right: - Import:

Import an ASCII file to simulate an acquisition.

- Spy:

Control dialog of the different data visible to RTKAPPA.

- Client:

Connect to a server PC via a serial link.

- Server:

Use the PC as a data server for client PC’s connected via a serial link.

- Client:

Export to a network socket.

- Server:

Import from a network socket.

- DLL load:

Load using an external DLL.

- Help:

Online Help

Fig B10.1 Data Supervisor command window

Click on the “Import” icon

to open an untitled Import window.

Use the browse option to navigate to the data file Sapb10.dat, supplied in the Examples folder. When the file has been selected the Import button is enabled. Leave the acquisition rate (Read every) at 1 second. The program will read the data lines in the file at a speed of 1 line every second and send the pressure reading to RTKAPPA. Click on the Import button

to start acquiring data.




Doc V3.10.00

The boxes at the right side of the window now show the time & pressure at the start of the acquisition, the current time & pressure reading, and the current point number.

Fig B10.2 Import command window

B10-1.2 Loading data in real time in Saphir Start Saphir and create a new interpretation Sapb10 with the default settings (overwrite if the example has been done already). Create the interpretation. In the QA/QC control panel, click on “Load”

, then in the “Load data” dialog, select “Real Time”:

Fig B10.3 Load data origin box The real time options are displayed in the window suggesting that you load from the channel Sapb10.dat, which is the only available channel. If there were more channels sending data, you would select them from the droplist.




Doc V3.10.00

Fig B10.4 Real time load option The window shows the reference date/time (t=0) for the Saphir session, but since we started Saphir some time after KdataSup, the first points in the loaded pressures would then be carrying negative time values. We can set the reference time of the Saphir session to the first time in the data to be loaded by simply clicking on the button. The refresh rate can be left at 1 second, as you want to see the data as it is loaded. Click on the OK button. The pressure data is then acquired and displayed in the pressure history plot, which a utomatically adjusts the scale to the range of the data. At this time there is no flow rate history.

Fig B10.5 Saphir real time pressure acquisition plot in QA/QC panel Switch to the “Interpretation” control panel.




Doc V3.10.00

Fig B10.5b Saphir real time pressure acquisition plot in the main screen The real time toolbar buttons allow you to pause, resume loading after a pause or to stop the load. Stop the load.

Pause in the acquisition without losing any data.

Resume loading after a pause.

Update the data.

Get information on the Real time acquisition status (loaded channel, data point number).

Fig B10.6 Real time acquisition status box To “Extract dP” and create a Log-Log plot, the user must first load a flow rate history. Click on “Load Q” “Extract dP”

and select “Keyboard”, and input 0.25 hours @ 100 STB/D. and accept the default settings. The Log-Log and Semi-log plots are now displayed and

refreshed in real time.




Doc V3.10.00

Fig B10.7 Log-Log and Semi-log plot in real time

Once you start to see a representative derivative response, generate the model (Improve)

and the regression

as usual. At any time, the program will work as if the load was completed, but whenever the

program returns to the main screen, the data will be updated on all existing plots.

Fig B10.8 Real time drawdown analysis If you wait for a while, you will see the simulated build-up being imported in real time from the file. At this time, go to the “Edit Rates” tab, split the rate history (add a build-up), synchronize the rate and pressure history as in guided session #2.




Doc V3.10.00

Fig B10.9 Split rate to create the buildup in the rate history

Switch to the “Analysis 1” tab, “Extract dP”

of build-up #1 and proceed with the build-up interpretation.

Fig B10.10 Build-up analysis in real time. Several channels can be loaded at the same time, which encompasses that not only real time interpretation can take place but also real time quality control in the QA/QC page.




Doc V3.20.00

B11 – Guided interpretation #11: Multilayer Analysis with static rates This guided interpretation covers the analysis of Multilayer Test Pressure Data, using bottom hole layer rate data acquired during Production Logging surveys. This guided session will use the data sets stored in the files Sapb11.asc, Sapb11.rat , Sapb11-plt.asc, stored in the examples directory which was installed during the setup of Saphir. In order to make the demonstration clear and straightforward, the data are synthetic (simulated) and the gauge acquisition problems were created to simulate some common field problems. It is assumed that the user is conversant with the functionality taught in the previous guided sessions.

B11.1 • Starting the New Interpretation This oil well test progam consists in a 26 hours clean up period followed by a 51 hours build-up, then the well was produced on increasing chokes for 26, 32, 39, 38 hours respectively and then closed in for an 85 hour final build-up. Layer 1 and Layer 2 bottom hole rates are recorded with a flowmeter in stabilized conditions at the end of the clean up eriod and at the end of the main production period. The surface rate data are stored in Sapb11.rat, the Production Data summary are given in the file Sapb11plt.asc. Create a new session, setting the reference date to 01/01/02 and accepting the default PVT and formation parameters values. Go to the “File” menu to save as Sapb11 in the example subdirectory.

B11.2 • Loading the data sets Click on

in the Interpretation control panel page. Choose to load from an Ascii file (default) and in the

file browser select the file Sapb11.rat.

Fig B11.1




Click on the

Doc V3.20.00

icon to load the pressure data from Ascii and accept the default format :

Fig B11.2 We can oberve that pressure data are missing at the end of the the Clean up and main flow period due to the PLT runs.

Click on

while pressing “Shift” to extract automatically the last Build up:

Fig B11.3




Doc V3.20.00

B11.3 • Initial standard analysis

Click on

improve Icon, while pressing “Shift”,

to generate the automatic model and perform the

improve on all the parameters :

Fig B11.4 The results are a skin value of 0.14 and a kh value of 674 md.ft for a single layer with a 30 ft thickness.

B11. 4• Initial multilayer analysis Create a new analysis and select Multilayer in the dialog :

Fig B11.5




Doc V3.20.00

The user is asked for the layer information :

Fig B11.6 Set the number of layers to two and keep the thicknesses equal to 15 ft. Two new icon appear in the Interpretation Control Panel :

which recall the previous dialog for the layer definition is necessary

and

Click on

to load the layer rate data.

while pressing “Shift” to extract automatically the last Build up , press on

access to the model dialog :

Fig B11.7 Accept the proposed values and generate:

to




Doc V3.20.00

Fig B11.8

Click on the

improve icon, the improve dialog is displayed:

Fig B11.9

Check in all the parameter for the regression and press run. The pressure match is perfect as shown in the flowing figure.




Doc V3.20.00

Fig B11.10 Two additionnal plots appear : The Mutilayer Diagram Diagram

which summarizes summarizes the layer geometry and the the Layer rate plot which shows shows the

calculated layer contribution. Double click on this plot header and press the right mouse button :

Fig B11.11 Select “show legend”. The default display is “simulated layer contributions”, pressing “simulated layer rates” displays the simulated rate at the top of each layer (layer 2 and Layer 1+2). We can observe here that the contribution are equal, consequence of the analysis results : equal kh and skin.




Doc V3.20.00

B11. 4• Final multilayer analysis Create a new Multilayer analysis from the previous one. This analysis will use the layer rate information. information. Loading layer rate

Press on the

Layer rate icon. To get access to the Layer rate dialog :

Fig B11.12 Press “Stabilized “Stabilized Q” to enter the layer rate data contained in the file Sapb11-plt.asc : Date 01/01/02 time 18:00:00 measured rate top of layer 2: 300 bpd

Fig B11.13 Press OK and specify that the data corresponds to layer 2 only:




Doc V3.20.00

Fig B11.14 Repeat the operation for the other measures : Date 01/01/02 time 18:00:00 measured rate top of layers 1+ 2:

2410 bpd

Date 09/01/02 time 00:00:00 measured rate top of layer 2:

270 bpd

Date 09/01/02 time 00:00:00 measured rate top of layers 1+ 2:

2210 bpd

Specify the layer rate description :

Fig B11.15 You can go to the browser to rename re name the layer rate data to “Layer “ Layer 2 plt1”, “Layer 1+2 plt1”, “Layer 2 plt2”, “Layer 1+2 plt2”:




Doc V3.20.00

Fig B11.16

Or by double clicking on the data name in the browser to modify the data channel properties:

Fig B11.17 Double click on the Layer Rates plot header and select in the drop menu (right mouse button) “Simulated Data Rate” to display the calculated rate corresponding to the acquisition conditions:

Fig B11.18 We observe that the simulated rates for layer 1+2 (total rate) are in agreement with the measured one, but that the simulated layer 2 rate is not. The pressure match is good but the rate match is wrong. This is due to the fact that the previous analysis used only the total rate and resulted identical layer in the absence of more information.




Doc V3.20.00

Improve using the layer rates

Click on the

improve icon, the improve dialog is displayed, specify the regresssion on the “simulation”

and “include layer rates”, select all the parameters for the regression:

Fig B11.19

Press “run”, the regression is performed on the complete pressure history but using in addition the match with the layer rates. The pressure match is as good as before . Double click on the Layer Rates plot header and select in the drop menu (right mouse button) “Simulated Data Rate” to display the calculated rate corresponding to the acquisition conditions:




Doc V3.20.00

Fig B11.20

The measured and simulated layer rates are in complete agreement, proving that the analysis is now correct.

Press on

to access to the model dialog and to check the new results :

Fig B11.21 The constrast of permeability (40 md and 4.5 md) explains the difference in layer contribution. The global analysis using the total rate was just giving a set of average formation parameters.






Doc V3.20.00

B12 – Guided interpretation #12: Multilayer Analysis, transient rates

This guided interpretation covers the analysis of Multilayer Test Pressure Data, using bottom hole layer rate data acquired during Production Logging surveys. This guided session will use the data sets stored in the files Sapb12.asc, Sapb12.rat , Sapb12 Bottom.asc and Sapb12 Bottom+middle.asc, stored in the examples directory which was installed during the setup of Saphir. In order to make the demonstration clear and straightforward, the data are synthetic (simulated) and the gauge acquisition problems were created to simulate some common field problems. It is assumed that the user is conversant with the functionality taught in the previous guided sessions.

B12.1 • Starting the New Interpretation This oil well test program consists in a 100 hour production period at 1000 bpd followed by a 100 hour build-up, then the well was produced on a greater choke for a 100 hour production period at 1500 bpd and then closed in for an 200 hour final build-up. The bottom hole rates are recorded with a flowmeter during all the first production period on the top of the bottom layer and during the second production period on the top of the middle layer. The surface rate data are stored in Sapb12.rat, the Production logging data are given in Sapb12 Bottom.asc for the first flow and in and Sapb12 Bottom+middle.asc for the second flow. Create a new session, setting the reference date to 01/01/02, accepting the default PVT and formation parameters values and with a standard analysis. Go to the “File” menu to save as Sapb12 in the example subdirectory.

B12.2 • Loading the data sets

Click on

in the Interpretation control panel page. Choose to load from an Ascii file (default) and in the

file browser select the file Sapb12.rat.

Fig B12.1




Click on the

Doc V3.20.00

icon to load the pressure data from Ascii file Sapb12.asc and accept the default format :

Fig B12.2

Click on

while pressing “Shift” to extract automatically the last Build up:

Fig B12.3




Doc V3.20.00

B12.3 • Initial standard analysis

Click on the

model Icon, while pressing “Shift”, to generate the automatic model.

Fig B12.4 A limit effect can be observed on the derivative curve. We will make an attempt of analysis with the production period #2.

Extract the production period #2, press on

to display the Model Catalog, select a Closed Circle as

Boundaries condition and enter the limit distance by using the “pick” option. The limit distance should be of the order of magnitude of 1300 ft. Impose the Pi to 5000 psi and generate.

Click on

while pressing “shift” to perform an automatic improve, it is made on the production #2 period:




Doc V3.20.00

Fig B12.5 The results are a skin value around 0 and a kh value of 515 md.ft and a distance to the limit of 1170 ft, for a single layer with a 30 ft thickness. When displaying the Build up #2, we see that the match is not satisfactory and no other improve on the build up can give better result.

Fig B12.6




Doc V3.20.00

B12. 4• Initial multilayer analysis Create a new analysis and select Multilayer in the dialog :

Fig B12.7 The user is asked for the layer information :

Fig B12.8 Set the number of layers to three and keep the thicknesses equal to 10 ft.

Click on

while pressing “Shift” to

extract automatically the last Build up , press on

access to the model dialog :

Fig B12.9

to




Doc V3.20.00

Enter the permeability equal to 17 (from the previous analysis) for all the layers, the PI equal to 5000 psi. Set the boundaries condition to Closed circle and set the distance to 1200 ft for each layer. Generate . The resulting match is similar to the analytical one since the three layers are identical :

Fig B12.10

Click on the

improve icon, the improve dialog is displayed:

Fig B12.11

Check in all the parameter except Pi, select the regression on the “simulation” and press run. The resulting pressure match does not show any improvement




Doc V3.20.00

Fig B12.12 Loading layer rate

Press on the

Layer rate icon. To get access to the Layer rate dialog :

Fig B12.13 Press on the “Load” button to load the layer rate data. The standard Data Load dialog is displayed, select the file Sapb12 bottom.asc and specify “Instantaneous” for the time scale :




Doc V3.20.00

Fig B12.14 Presss “Load”. Proceed the same way to load the data file “Sapb12 bottom+middle.asc”. Specify the layers corresponding to the measures in the “Edit layer rate” dialog :

Fig B12.15

Double click on the Layer Rates plot header, ask to display the legend and select in the drop menu (right mouse button) “Simulated Data Rate” to display the calculated rate corresponding to the acquisition conditions:




Doc V3.20.00

Fig B12.16 This plot demonstrates in which measure the analysis is wrong : the individual contributions are not correct. This is due to the fact that the analysis cannot evaluate the permeability contrast by taking into account the layer contribution. Improve using the layer rates

Click on the

improve icon, the improve dialog is then displayed, specify the regression on the

“simulation” and “include layer rates”, select all the parameters for the regression:

Fig B12.17




Doc V3.20.00

Press “run”, the regression is performed on the complete pressure history but using in addition the match with the layer rates. The pressure match is better than before :

Fig B12.18 Double click on the Layer Rates plot header and select in the drop menu (right mouse button) “Simulated Data Rate” to display the calculated rate corresponding to the acquisition conditions:

Fig B12.19 The measured and simulated layer rates are in quite good agreement, but the pressure match should be improved. Using the regression to improve the pressure match always goes back to the same approximative match.




Doc V3.20.00

The regression calculation found a local minimum and cannot get out to converge to the absolute minimum.

Press on

to access to the model dialog and to check the new results :

Fig B12.20 The permeability contrast and the various limit distance are a good approach to the solution but are blocked at this level.

B12. 4• Final multilayer analysis Create a new Multilayer analysis from nothing and start again using the same procedure. The layer rates are already loaded but the layer rate distribution has to be entered again in “edit layer rate”. When the first model is described enter the permeability to 17md for all layers and the impose Pi to 5000 psi, but keep the limit distance to 3000 ft. Generate :




Doc V3.20.00

Fig B12.21 The match is not satisfactory use the improve to perform a regression on the complete simulation and using the layer rates.

Fig B12.22 It can take few minutes.




Doc V3.20.00

Fig B12.23 The match is now perfect, as good for the pressure data as for the layer rate data. The use of “ Wide Search” When such a local minimum is found and that the regression can not improve more the match it remain the possibility to use in the improve dialog the option “Wide Search”. The option will make calculate 300 (depending on the setting we specified) run of simulation in order to detect the global minimum, then starting from this point it will run a standard regression, avoiding like that to “fall” in a local minimum.






Doc V3.20.00

B13 – Saphir guided interpretation #13: Numerical multiphase

B13-1 • Multiphase The objective of this tutorial is to illustrate Saphir 3.20 multiphase abilities. In this guided session the user will be led through the Saphir dialogs and screens to enable the simulation of the pressure response of a multiphase producer (oil and water) coupled with a water injector (into oil). The evolution of the fluid saturations will also be studied.

B13-2 • Tutorial initialization Run Saphir and start a new project. Keep the well and reservoir characteristics at the default values and specify the reference phase as oil, and in the “available rates”, add water. Specify numerical analysis; this will invoke the non linear multiphase simulator with con stant composition. See Fig. B13.1.

Fig B13.1 The next screen is the document or project specific PVT dialog. Automatically “Dead” oil is selected with water. Leave the default PVT characteristics;

Note that a click on the button

in the PVT toolbar will warn the user that default values for all phase

parameters and correlation are in use. Accept the next PVT screen and change Sw to 0.2 in the final PVT dialogue. Click on the button “So and Sg from flash”

to adjust the value of So. The saturations entered here are only used for a constant system

compressibility calculation. The user will later specify the “Analysis” specific PVT and define initial saturations



for model generation and simulation. The next screen shows the PVT characteristics calculated for the reference phase.

will define the new project.




Doc V3.20.00

B13-3 • Numerical “Interpretation” panel After the creation of the new project, Saphir switches automatically to the “Interpretation” panel and creates Numerical 1

new analysis tab.

Two new buttons are added to the “Interpretation” panel.

Edit numerical PVT. This PVT is “Analysis” specific. Thus each analysis can have a different PVT. Edit numerical KrPc. The definition is “Analysis” specific. Each analysis can have different relative permeability definitions.

B13-4 • Tutorial analysis specific PVT and KrPc

Click on

in the “Interpretation” panel. This accesses the analysis specific PVT dialogue and the user

can here change the parameters. Observe that “Dead Oil” and “Water” has been carried over from the document specific PVT defined during start up. In this guided session keep the default values.

button will warn the

user that default parameters and correlations are being used.

Click on

to define the relative permeability definition for this analysis. In the tab “KrData” enter the

following values: Swr = 0.1, Sorw = 0.2, Krwo = 1, Krow = 1. Use the power law model at power 1.65 and 2 respectively for water and oil with no extrapolation. See Fig B13.2.

Fig B13.2




Doc V3.20.00

The user can access the plot of the relative permeabilities versus the water saturation by clicking on the tab “Kr Two-phase”. Fig. B13.3.

Fig. B13.3 It is possible to interactively change the relative permeability curves at the click of the mouse by dragging the handles (yellow squares) on the curves.

B13-5 • Defining the 2-D Map and the wells Click on the tab

to access th e 2-D Map. Double click on the contour and choose set as a

rectangle at 30,000 x 30,000 ft as indicated Fig B13.4

Fig. B13.4 The tested well which is the reference well until the user chooses to change this is at coordinate (0,0). Now add a well at coordinate (-3000 ft, 0), use the button and set the coordinates exactly, see Fig. B13.5.

to place the well approximately. Double click on the well




Doc V3.20.00

Fig. B13.5 Double click on the tested well to define the production history. Click on the tab “Production” and enter 1000 days of Oil rate = 5,000 STB/D and Water rate = 500 STB/D. Fig B13.6. Again double click on the added well, designated Well#1, which is going to be specified as an injection well by entering a negative water rate. Enter the same duration as the “Tested well” and a Water rate = -6,000 STB/D and no Oil production.

Fig. B13.6: Tested well & Well#1 productions




Doc V3.20.00

B13-6 • Running the simulation Click back on the tab “Numerical 1”,

. Choose the option “Test Design” in the “Interpretation (2)”

panel. Check “add other wells” and “store pressure fields”, run the simulation with initial pressure Pi= 5,000 psia, a permeability kh= 1500 mD and initial water saturation Swi= 0.2. See Fig. B13.7.

Fig B13.7 The simulated pressure response is illustrated in Fig. B13.8.

Fig B13.8




Doc V3.20.00

Extract dP to see the loglog plot and click on Model to generate the simulation again with the default parameters that are the same as those of the test design. Click on the button fields” and choose “Display during generation”. Fig. B13.9.

Fig. B13.9 The main screen after generation is illustrated in Fig. B13. 10.

Fig. B13.10

by the “store results




Doc V3.20.00

There are now two match lines on the loglog plot. The white dotted line correspo nds to the stabilisation obtained if only the oil phase is mobile. The green dotted line corresponds to the stabilisation at Swi (initial water saturation specified in the model).The results returned are the absolute permeability kh and the effective permeabilities to oil and water at Swi. The user can now display the simulated pressure history and the simulated phase rates in the history plot. You can right click in the rate area of the history plot and select “Show phase rates” – “All”. The history plot is illustrated in Fig. B13.11. The user can observe that water breakthrough happens at about 4000 hrs production.

Fig. B13.11

B13-7 • Simulation value fields The geometry plot is automatically created.

Maximise this plot and click on the button

to define the

setting of the value fields available after this simulation. It can be seen that the field values “Pressure”, Sw and So is available. See Fig. B13.12




Doc V3.20.00

Fig. B13.12

Fig. B13.13 Leave the shading as selected on Pressure with the other default selections. Visit the tab “Color Scale”, use 7 colors and check “always visible”. The “always visible” option will display a defined color scale of the chosen value on the geometry plot. By setting the maximum and minimum pressure to 5,800 and 3,000 psia respectively as indicated in Fig B13.13 the colors of Fig. B13.14 should be close to yours. Click OK and return to the geometry plot. You can now play back the evolution of the pressure fields by using the “tape recorder buttons” in the toolbar.

Click the

button to return to 0, and then

to play back the pressure evolution. The final pressure field of

the tested well will look something like Fig. B13.14.

Fig. B13.14




Revisit the settings

Doc V3.20.00

option of the geometry plot. This time choose “Shading” for the value Sw (water

saturation). Click on the “Color Scale “ tab and check that you have 3 colors as indicated in Fig. B13.15. Then click the

button to return to 0, and play back the fields using the button

.

Fig. B13.15 The final field in the geometry plot clearly shows that water has broken through, Fig B13.16. is a zoomed version of the plot.

Fig. B13.16

Click on the toolbar button

in the geometry plot. This will display the plot in 3D. Choose the settings button

. Select the setting as indicated in Fig. B13.17.




Doc V3.20.00

Fig. B13.17 The 3D plot will look something like Fig. B13.18 depending on the settings and the zoom applied. The 3D plot exhibits the pressure variation as the vertical scale, showing clearly the injection and the production well. The colour coding is the evolution of Sw (in this case at the field after 7,000 hrs production) and indicates clearly that the injected water is pouring into the producing well.

Fig. B13.18


Additional Saphir Examples • BX01-1


Doc V3.20.00

BX01 – Additional Saphir Example #1: 2 Φ PSS This chapter is for Saphir Advanced users only After going through the guided interpretations B01 to B07 you should now be familiar with the loading/editing of data as well as the default Saphir path. This chapter provides a series of examples on selected models and facilities: All the files are copied to your computer when downloaded from our web site as these files are not copied during the Saphir installation. The style of this Chapter is much more condensed than that of the guided interpretations and features already explained are not detailed here. These brief notes should help you as you work through the examples.

BX01.1 • Pseudo-steady state double-porosity Before starting this example, locate the f ile Sapbx01.ks3. Start Saphir and open the interpretation project Sapbx01. Both rates and pressures are loaded and the flow period has already been extracted.

Fig BX01.1 The pressure derivative shows the typical trends of a reservoir with pseudo-state double-porosity. However, none of the theoretical stabilizations is apparent on the data. In a double-porosity (or 2 layer) model, IARF must correspond to the radial flow of the total system. Even when this section is not developed, it is possible to get a first estimate with the line option and then refine this estimate interactively. In this case, the first estimate will be close enough so that we can let the “Improve” make the necessary refinements. Make the log-log plot active (one click in the plot header). Define a line at the end of the derivative (right click in the log-log plot, Line). Maximize the log-log plot and reset the match IARF straight line.

, the pressure match snaps to the




Doc V3.20.00

Fig BX01.2 Select automatic Model. The model response is of course quite far from the data but: The reset match set the pressure match to the level of the IARF line we defined, and computed the time match accordingly. A skin estimate has been made by the automatic model.

Fig BX01.3




Doc V3.20.00

Select Model, and 2-Porosity P.S.S.

Fig BX01.4 Accept the proposed parameter values, and the value calculated for the Skin. The model requires the input of Lambda and Omega, for which a Pick option exists that can be activated by the

button to the right of the

parameter menu. Click on this button

. Fig BX01.5 The log-log plot is displayed on the screen with the instruction “Pick the minimum of the transition and drag to adjust”. Set the cursor on the derivative at the bottom of the "valley", about the middle of the 4th log cycle. Moving the mouse up and down changes the depth of the valley, and moving the mouse left and right moves it left or right. Adjust for the best fit on the derivative, then click. The parameters Omega and Lambda are displayed and can be edited. Accept the values and generate the response.




Doc V3.20.00

Fig BX01.6 Select automatic improve to obtain the refined match illustrated below.

Fig BX01.7




Doc V3.20.00

BX02 – Additional Saphir Example #2: Horizontal well BX02.1 • Horizontal Well This section illustrates the use of the horizontal well model on data simulated with the Test Design option of Saphir. The interpretation is called Sapbx02.ks3. Open Sapbx02. Extract the build up.

Fig BX02.1 The automatic pressure match has been set at the right level by the program, i.e. on the final stabilization corresponding to radial flow in the horizontal plane. The corresponding value of transmissivity is kr.h, kr being the radial permeability and h the formation thickness. Go to Model, select the horizontal well. In the parameter input screen the value of well lenght should be set to 1000 and zw should be left at the default 25 ft.

Fig BX02.2




Doc V3.20.00

The parameter kz/kr has a pick option and it should be selected. The log-log plot appears and you are asked to Pick the 1st stabilization level on the derivative. Pick as shown below

Fig BX02.3 The type of top and bottom boundaries can be selected from the drop lists. Sealing or gas cap for the top boundary and sealing or water drive for the bottom boundary. There is no evidence of pressure maintenance here so both boundaries should be left as Sealing. Generate the response. A non-linear regression can be run to refine the match. Select Improve and run (Automatic Improve is disabled for this model in view of it’s complexity).

Fig BX02.4




Doc V3.20.00

BX03 – Additional Saphir Example #3: Rate grouping BX03.1 • Automatic Rates Grouping This section illustrates the use of the Test Design option to predict the response to a test sequence and the use of the grouping function to analyze a long production period when the flow-rates decline with time. Start a new Saphir project Sapbx03, using the default parameters and Oil. Load the flowrate file Sapbx03.rat. The sequence consists of 21 periods of 10 hours each with the flowrate declining by a few barrels every 10 hours.

Fig BX03.1 Select the first button in the “Interpretation (2)” control panel “Test Design”

.

The Model menu is opened. Accept the default value for C, 300 mD.ft for kh. Leave the Pi at 5000 psi and the Skin at 0. To introduce gauge limitations click on gauge and, change the gauge type to “Strain” and edit the resolution to 1 psi, gauge type changes to “User defined”.

Fig BX03.2




Doc V3.20.00

Generate the simulated response for a homogeneous infinite reservoir. In order to extract the whole drawdown the first solution would be to merge all the flow periods, if we chose not to take advantage of Saphir’s automatic grouping system which groups consecutive flow periods of the same type into one group. Press on the tab “Edit Rates” and use the button “Merge Rates”,

to merge from period 1 to 21. The

routine calculates the average rate which is applied throughout the drawdown and transforms the declining rate history into one single rate.

Fig BX03.3 Select Automatic Load delta P. The derivative appears to show a constant pressure boundary and a quite good match can be made with a constant pressure circle.

Fig BX03.4 The assumption that the rates could be averaged therefore leads to an erroneous interpretation, as we know the reservoir to be homogeneous infinite. The automatic rate grouping allows a correct interpretation, without any simplification of the true rate history. Reload the flow rate file Sapbx03.rat. Then click the tab “Edit Rates” where it can be seen that Saphir has automatically grouped the declining rates into one production period: Production #1.




Doc V3.20.00

Fig BX03.5 The automatic load extracts the delta P data for the entire sequence, showing infinite behavior which can be easily analyzed.

Fig BX03.6 The general superposition function takes into account the changes in the rate history and normalization with respect to the individual rates is done on the semi-log and log-log. This ensures that consecutive periods in Infinite Acting Radial Flow produce aligned sections on both plots.


Additional Saphir Examples • BX03-4 Doc V3.20.00




Doc V3.20.00

BX04 – Additional Saphir Example #4: Tidal effects

In this example we will demonstrate the facilities developed in Saphir to remove tidal effects. The files Sapbx04raw.asc and Sapbx04seabed.asc are used. It is assumed that the user is already familiar with Guided Sessions 1 through 6.

BX04.1 • Tides Tidal effect on pressure measurements can be strong, especially when data acquisition occurs offshore (for instance in the North Sea). But it may also occur in onshore fields. This type of effect can heavily influence the derivative behavior and errors in the analysis are frequent. Therefore, it is important to remove this effect without removing the true reservoir response before attempting any interpretation.

BX04.2 • Example tide removal Launch Saphir and open a new session using the default oil parameters. Change to the QA/QC page and load the first measurements Sapbx04raw.asc. This file contains the raw data from a build-up which has been affected by the tidal effects in the area. Zoom in until displaying correctly the tide effect.

Fig BX04.1 Next load the file Sapbx04seabed.asc in a new plot in the QA/QC page. This measurement is in fact the pressure response at the seabed by the well head due to tidal effects.




Doc V3.20.00

Fig BX04.2

Fig BX04.3

Calling the tidal facility

, a dialog with a composite plot (see Fig BX04.4). The upper one contains the

active gauge of the active QA/QC plot (this facility cannot be called when a child plot is active). The lower one contains the tidal signal if already loaded. If not, the user needs to load this signal, either coming from a seabed gauge or from a software-generated signal. Note that, entering the well location; a new facility will soon be integrated in Saphir 3 to allow the generation of this tidal table. Anyway, using the droplists the user can choose the correct data channel to be put in each plot. •

The data to be corrected in the upper plot. •

The Tide table in the lower plot.




Doc V3.20.00

Fig BX04.4 In order to phase (synchronize) the tide signal with the measured down hole data the user can do this manually using the mouse by click and drag in the tides plot after clicking the button

or right clicking in the plot for

the contextual menu.

Fig BX04.5

Then the tide amplitude is calculated

, use of the right mouse button will also access this option. The

signal is removed from the data and the resulting curve is shown in the upper plot. The button

“Auto-match time shift and amplitude for tides” that will run a non linear regression to make the

correction can also be used.

Tutorial Saphir 32

Recommend Documents