B02 - Rtaex01
B02 - Rtaex01
B02 - Rtaex01
1. Introduction
This tutorial provides a description of the options and workflow in KAPPA-Workstation. This includes creation of
new documents and analyses, loading of pressure and rate data, extraction of the loaded production data, Decline
Curve Analysis, use of loglog analysis tools (new in G5), example of analytical and numerical modeling,
specialized plots, sensitivity and forecast, and the creation of interpretation file templates (new in G5). The tutorial
finishes by pointing to a few items that users may decide to explore on their own.
Before starting this session, the user is expected to have installed KAPPA-Workstation and started the RTA (Topaze
NL) module. The tutorial will use the three files (below, left) located in the Examples folder in the Installation
directory. Topaze NL starts (below, center) and brings the user to the File page. The active option is New and
recent and a Blank icon can be seen towards the top left of the screen (below, right).
- Step 1: initialization of the main document options: reference time and location, general information, units
and general comments. Keep everything as default and click .
- Step 2: main options of the first analysis in this document. Input the main test parameters. Those
highlighted with red fields have a significant impact on the results and should not stay at default value. If
the default happens to be the answer one may enter the same value or select Accept default using a right
mouse click. If any field remains, a red warning message will be carried out throughout the interpretation.
For this session, set the pay zone (h) to 100 ft and the porosity () to 0.08. Click .
- Step 4: definition of the constant parameters and/or pseudo-functions that will be used in analytical or
linear numerical models (linearity is required for superposition). The computed pseudo-pressure and
pseudo-time integral can be viewed in tabulated and graphical form using (below, right). The choice
of pseudo pressures (none, not normalized or normalized) affects the plot axes only. The model is always
generated in pseudo pressures. Click on .
- Step 6 (new in G5): Topaze NL G4 would initialize the default model as a fully penetrating vertical well in a
homogeneous reservoir bounded by a circular boundary. This can now be overridden at any stage of the
analysis, even at initialization stage. If the user knows or suspects that a given model should apply there is
no need to start with an irrelevant default. For this session, do not change the default at this stage. Click
on .
Click on the Load Q, , icon in the control panel on the left to load the Rate History. The rate information
is stored in RTAEX01 Rates.xlsx. Click on and select the file to load. A file preview is shown (below, left).
with the possibility to change the tab/worksheet. brings a dialog where the file information is
interpreted line by line (below, right). The collapsible panels on the left offer detailed load options. The top
right section has a set of editable information while the bottom left window gives the result of the format
processing. As the input file is very simple, with time stored as durations, the default format will work. Click on
to proceed.
The main Topaze NL screen is displayed again with a history plot showing the loaded rates.
The Extract p, , icon in the control panel accesses the manual extraction dialog. We will revisit the dialog
later. The Automatic Extraction, , icon appears in lieu of the Extract p icon when the shift key is pressed.
In this state, click on the icon and the entire loaded production history, using the default extraction settings
will be extracted on a DCA plot, generated at the end of the extraction.
Change the top right plot to q vs Q by right clicking on it and selecting the plot type in Decline curve options
(below left). Similarly, change the bottom one to log(q) vs log(t) to end up with something as shown below,
right:
Click on the Parameters icon in the Plot options panel at the top and change the model to Duong. The slider
bar at the bottom allows the user to set the relative weight of rates and cumulative production while regressing
on the model parameters. Move the slider to mid-way between q and Q and run the regression on initial rate,
M and A Exponents (each decline curve will have its own set of parameters):
Several DCA plots may be created with different decline models on each.
Close the dialog and restore the DCA plot by double clicking on the plot title bar.
Before proceeding with loading pressure data, access the RTA Settings through located at the top right of
the Topaze NL window and select the automatic plots as indicated below:
Based on the above selection, when an extraction is made only the normalized rate cumulative plot will be
created (in addition to the loglog plot and Blasingame plot which are always created on extraction). This allows
the user to save overcrowding the workspace with plots the user is not intending to use. The unchecked plots
can always be created at any time during the analysis using the New plot option in the ribbon (subject to plot
prerequisites being met). This will be visited later in the tutorial.
Back in the Topaze NL main workspace, the history plot is displayed with both rate and pressure data.
Since an extraction already existed, even though we deleted the DCA plot, loading pressure data will also
launch the extraction dialog. This time, the extraction will be based on both pressures and rates, unlike before,
when we had rates only.
If no extraction had existed, the dialog can be manually called using the Extract p, , icon in the control
panel.
Reset the interval to the complete history by clicking on , set the starting value to 1hr, then click on
to proceed with extraction.
The main Topaze screen (below, right) has four plots (loglog, Blasingame, normalized rate - cumulative &
history) and a result windows where a red warning indicates that some key parameters remain at their default
value.
7. Analysis tools
Instead of a simple circular boundary, centered around the well, we will model the boundary with a centered
rectangle i.e. North-South boundary distances are the same and East-West boundary distances are the same.
Click on the 'Tools icon in the Analysis control panel. This recalls the Analysis tool dialog, step 6 of the
new document wizard (below, left). Change the Boundary model to Rectangle (centered) and click on
to validate.
Depending on the position of the limit markers on the loglog plot, the North-South and East-West distance
values observed in the result window can be a little different.
Hide the tool parameters and restore the loglog plot by double clicking on the plot title bar.
When plots are not maximized and several plots exist in the workspace, the plot area may be reduced
by the presence of plot scales. In such instances, the scales may be hidden using this option.
The extracted interval can be edited by clicking on Selection and interactively changing the time range
in the consequent popup plot. Its impact on the plots is immediate.
The loglog plot uses the equivalent time Te = Q/q. Small rate values can be filtered in order to avoid
excessively large values of Te by clicking on Filter and specifying the criteria in the filter dialog.
Certain data points can be selected in a lasso by clicking Time interval selection, then holding the left
click down in a plot to draw a lasso around them. The corresponding points are automatically
highlighted in all other plots (below, right).
The Show values option, , in the plot tool bar in any plot, can be used to display the time and data values
inside the plot, e.g. in the loglog plot (below, left)
The Analytical, , icon in the control panel accesses the manual analytical dialog. Model and parameters
have been initialized from the settings and results of the loglog tool. Clicking on the button would
generate the model with these parameters, but we may as well call the automatic model directly. So click on
to exit the manual analytical model dialog
The Automatic Analytical, , icon appears in lieu of the Analytical icon when the shift key is pressed. In
this state, click on the icon and the model is executed in a single command, with the resulting curves displayed
on the visible plots.
In the ribbon at the top, the single step response can be turned on/off using . Turn the option off to see
the true pressure response on the loglog plot (inset image above).
Turn the single step response back on. It will be left on for the remainder of this tutorial.
In this session, the normalized rate cumulative plot (which is already created) will be used to estimate the
drainage area size. Maximize the normalized rate cumulative plot and display results in the plot using the
option in the ribbon at the top (below, left). Activate the Line plot option from the ribbon at the top and select
a range of data in boundary dominated flow by holding the left mouse click and highlighting the relevant data
range. A straight line will be drawn based on regression on the data within the selected interval. The volumetric
results will be updated when the line is redrawn (below, right). You may need to reset the zoom to see the
complete line.
Restore the normalized rate cumulative plot by double clicking on the plot title bar.
The destination model may be the analytical or numerical model at the bottom of the icons on the left. Clicking
on one of these two buttons will transfer the relevant active transferrable results to the model and execute
the model at once. If for example you select the Analytical model in the top column and send to the numerical
model you will do exactly what you just did when calling the numerical model with the analytical values.
However, the dashboard establishes a more flexible bridge between the different sources of results.
In this session, select the normalized rate - cumulative plot from the list of specialized analyses and click on
the To analytical icon:
Since the PV has no information on the relative North-South and East-West boundary distances, they are all
reset to the same value, which gives the same PV as calculated from the normalized rate cumulative plot.
The Improve, , icon in the control panel accesses the manual improve dialog with two tabs defining the
targets (below, left) and the parameter controls (below, right). One can select the regression parameters, set
their ranges and apply different weighting on various sections. One may also choose between a match on the
cumulative production or the production rates or both. In this session, select the cumulative production as the
target (below, left) and choose the permeability and boundary distances (below, right) as regression
parameters.
Clicking on the button would run the regression with the defined settings.
The existing analytical model is deleted, but the model parameters are still retained in memory. Click on the
Numerical, , icon to access the manual numerical model dialog (above right). The numerical model can
be defined automatically based on the diagnostics (analysis tools) or from the analytical model. To initialize
the numerical model from the analytical one click on and click on . In addition to
the model response at the well, a 3D plot with the reservoir geometry, the static and dynamic reservoir
properties is also generated.
With a numerical model initialized it is possible to consider many more complex options either geometrical (in
the map ribbon), related to the fluid behavior (PVT), etc. In the following sections, we will look at the PVT a
little. Switch back to the Analysis Tab.
Now that the model is generated, it can be used to forecast future production. Click on the Forecast, ,
icon to access the forecast setup. Select the Constant pressure forecast option. Set a producing pressure of
1750 psia and forecast for a duration of 3 months and generate the forecast:
Validate the setup with . The model will be removed. Access the numerical model dialog and
regenerate the model (a zoom reset on the history plot may be required).
Adding desorption to the model leads, quite logically, to an overall higher gas production.
Click on the Improve, , icon in the control panel accesses the manual improve dialog. Select the rates and
the cumulative production as the regression targets (below, left) and the reservoir surface and the regression
parameter (below, right). Click on .
Click on the Compare, , icon in the Analysis ribbon at the top. In the compare dialog, select Analysis 3 and
Analysis 2. The models, as well as the forecasts, are displayed with a specific color in the various plots:
The color assigned to each analysis in compare can be changed by right-clicking on the analysis in the compare
dialog (below, left). Alternatively, this may also be done in the Analysis Information dialog.
15. Sensitivity
Once a model is generated, sensitivity on various model parameters can be run to see their influence on the
model response or forecast.
Click on the Sensitivity, , icon to access the sensitivity dialog. The default is to generate 5 sensitivity runs
between a minimum and maximum value. Select the Forecast pressure as the sensitivity parameter and define
to generate five sensitivity runs for forecast pressure between 1400 and 1800 psia (below, left). Click on
to generate the sensitivities.
The computation of the various responses is executed on parallel threads on a multicore PC and the results are
displayed on a dedicated history plot (below, right).
Alternatively the results of the Sensitivity runs can be displayed on the loglog plot by selecting in
the Sensitivity dialog.
The template stores the setting of the document and the active analysis at the time it was saved. The next time
the user wants to create a new document, they will just have to select the template. Notice that the wizard,
which takes the user through the six initialization steps, has the settings of the current analysis. This can be
checked with the parameter values in Step 2 (below left) or the PVT and diffusion setup (below right).
However, you may want to explore the capabilities of the numerical models a little further. In the session above
we had just initialized the NL numerical model from the analytical model and gone one step further to add
desorption in the NL model. You may go to the 2D-Map tab, load the field bitmap file RTAEX01 Field.jpg,
scale the field, define the contour, faults, position the producing well properly, add an injection well and display
the resulting grid in 2D or 3D, enter the interference well schedule, create layers, select a more complex PVT,
etc.
KW also has a comprehensive contextual online help, including How to topics, Examples and FAQs, to assist
the users whilst using the software. Users are encouraged to consult these.