Static Modeling by Petrel

Petro Eenrgy Kimia(petek-co.ir) [email protected]

[email protected] @petekco Petro Eenrgy Kimia(petek-co.ir) [email protected]

telegram.me/petekcompany

petek-co.ir :


Model Overview 2.a,,Geology: 2.a Sedimentology,Sequence stratigraphy, Fracturation

1,Well description, stratigraphic layering/ coordinates 2.b,,Facies 2.b (seismic) 3,Geostatistical simulation of lithotypes

4,Well test interpretation

5 ,Petrophysical simulation high resolution grid 7 ,Model uses: 1- Vo Volume lumetric/ tric/ Mapping Mapping 2-Assecc Connectivity

6 ,Petrophysical simulation low resolution grid

3- Scale Up Up for Flow Simulati Simulation on 4- Place Wells/ Wells/ Process Process Design Design

@petekcompany


What is the model

Reservoir Modeling Process

@petekcompany


Petrel Workflow Tools Intro

Zonation & Layering

Well Correlation

Facies Modeling

Fault Modeling

Pillar Gridding

Petrophysical Modeling

Volumetric calculation & well Design

@petekcompany


User Interface

Menu Bar

Tool Bar

Function Bar First Petrel Explorer Window

Second Petrel Explorer Window @petekcompany

Status Bar

Object Information

GH. KARIMI GEOSTATISTICS RESERVOIR MODELER


Petrel Introduction Open and saving petrel project Open second petrel project Petrel explorer ◦ Input tab ◦ Model tab ◦ Result tab The numerical results of volume calculation and simulations are store in this tab.

◦ Templates tab Contain all predefined templates in petrel

Project setting and Unit Visualization ◦ Windows ◦ General Intersection

@petekcompany


Data Import General import procedure: 1-Create a new folder 2-Rename folder 3-Use the “import” option to import into the folder Exercise: make a new folder & rename it and import some data

Wells and well tops procedure: 1- Create well and well tops folders 2-Rename folders if desired 3-Import well data (headers, deviation surveys, logs) 4-Import well tops Exercise: make a new well folder & rename it and import well data and well log

@petekcompany


Edit Input Data Imported data will often need some editing to make them suitable for modeling 1- line data imported without a “flag value” indicating line segments. (polygon and stick) Double click on object and go to operation tab

Select split by horizontal length

execute

2- Irregular 2D grid that needing smoothing, and 2D grids or seismic interpretations containing spikes.

@petekcompany


Edit Input Data Make/Edit polygons Create a new polygon or start on an active polygon

Create new polygon and deactivate the old

Create new line within existing polygon

Close polygon

@petekcompany


Edit Input Data Make/Edit surface This process is where you can grid data into surface. it is a way of preparing the input data. Several type of data can be used as input :point/line/surface in addition to creating surfaces you can also smooth, blank and fill grid, interactively edit grids and tie a grid to data

Make/Edit surface process

Enter the data to be gridded in main input

Give a name to the output

Define the geometry

Click on suggest setting from input /setting the algorithm tab /the surface is stored in input tab after apply

@petekcompany


Edit Input Data QC the surface

Highlight a surface in the input tab

Open the make/edit surface process and input the highlighted surface in the result tab

Geometry, algorithm and input tab are automatically filled in

By deleting the result surface all the parameter are reset. @petekcompany


Edit Input Data Make/Edit surface -interactive smooth

Make sure the Make/Edit surface process is active

Active the smooth area tool in function bar

Select the area of influence in number of grid cell from the bottom toolbar

@petekcompany


Edit Input Data Make/Edit surface –peak remove

Make sure the Make/Edit surface process is active

Active the peak remove tool in function bar

Click directly on the base of peak

@petekcompany


Edit Input Data Surface calculator (is the one of method for create calculated surface – another method use the interpreted surface and isochore)

Right click on a surface and select calculator

Type in a new output name. use the calculator option to create an algebraic expression

Press enter to execute Stored in history field At top of calculator

The new surface will be stored at the bottom of input tab

@petekcompany


Edit Input Data Make isochore Vertical Isochore

Deviated

Isopach

Z top – Z base

Z interpolated – Z base

@petekcompany


Edit Input Data Vertical Well – Isochore point

Highlight one of the well tops and right click on the other well top that will define the isochore interval

Select convert to Isochore point

A new point set data is created in the I input tab .

Right click on the thickness attribute and select use a vertical position The point are now shown as thickness point.

@petekcompany


Edit Input Data Vertical Well – Isochore surface

Drop isochore point in the main input of make surface process

Select the thickness Attribute from the Drop down menu For the attribute field

Set geometry

Visual the new isochore map in the 3D window

@petekcompany



Zonation & Layering

Well Correlation

Facies Modeling

Fault Modeling

Pillar Gridding



@petekcompany


Well Correlation Open well section and view data Creating well section Creating well section in 3D window

Make sure all other well section in the window tab are deactivated

Select the add well to well section icon from function bar

Click on the wells in the 3D window a line appear connecting the wells

Curve filling @petekcompany


Well Correlation Setting up group panels

Right click on a well in well section and select insert group panel

A folder called group is created. Drag and drop curves into group folder

In the well section window there are multiple logs in one panel

@petekcompany


WELL TOP model Use in geo Md .

@petekcompany


Well Correlation Well tops – create and edit

Well tops – flattening : all well top be flat

@petekcompany


Well Correlation Interactive facies interpretation

Flood fill discrete log class

Paint discrete log class

Pick up discrete log class

Create new continuous log

@petekcompany


Well Correlation Well seismic

Right click on the seismic cube and select create well seismic stored in global well log

@petekcompany


Well Correlation Bitmap logs

Import bitmap logs like FMI

Edit bitmap logs

@petekcompany


Well Correlation Fit between two curves- group panel

Insert group panel and select the logs

Select one of the log and double click on it until open setting, go to curve filling

Select new from top to bottom under select interval

Set fill parameter and select other curve in fill edge and fill style

@petekcompany


Well Correlation Curve filling- cut offs and property fill

Insert group panel and select the logs

Select one of the log and double click on it until open setting, go to curve filling

Select new from top to bottom under select interval

Set fill parameter and select curve level in fill edge and fill style (cut off, automatic ) in fill pattern

@petekcompany


Well Correlation Ghost curve For compare between two well logs

Select well & well log as require, click on the well header and drag it to another well

The ghost curve will now appear in window

@petekcompany


well section .

@petekcompany



Zonation & Layering

Well Correlation

Facies Modeling

Fault Modeling

Pillar Gridding



@petekcompany


Fault Modeling Fault modeling is the process of generating a faulted grid and inserting the horizons, zones and layers into it.

Top shape point Mid shape point Key pillar Base shape point

Line between pillars @petekcompany


Fault Modeling Fault type definition 1.

Vertical fault

2.

Linear fault

3.

Listric fault

4.

Curved fault

Almost any type of fault can be modeled in PETREL: Single, branching and crossing fault Reverse and Normal fault Truncated fault Growth fault @petekcompany


Fault Modeling Different input data 1- Fault stick

3- Fault surface

2- Fault polygon

4- 3D line

5- Seismic @petekcompany


Fault Modeling General workflow

1.

Define a fault model

2.

Model the initial set of fault (generate key pillars)

3.

Edit any problem areas

4.

Trim/Execute to the top and base of the model

5.

Connect fault (required for pillar gridding)

6.

Add more faults by repeating steps 2-5

7.

Do the final edits @petekcompany


Fault Modeling Define model

Open define model in the process diagram and rename it

The new model is stored in model tab

Active the fault modeling process

@petekcompany


Fault Modeling Fault modeling from fault polygons + Shift to select all po lygon

Create fault from fault polygon

Edit key pillar vertically

@petekcompany


Fault Modeling Fault modeling from fault sticks + Shift to select all polygon

Create fault from fault sticks, surface or interpretation

Create fault from fault stick

Notice that by choosing this method, a fault will be generated that connect the sticks that you selected. It is better to use create fault from fault sticks, surface or interpretation @petekcompany


Fault Modeling Editing key pillar

By clicking on the plane you can move the shape point in 3D. By clicking on the cylinder you can move the shape point along the tangent of the cylinder

By clicking on the line between shape p oints, you will select all the shape points along the line . You can also combine the icons called select one shape point and select entire pillar, will selecting one key pillar and clicking on the line between shape points.

@petekcompany


Fault Modeling Editing key pillar

Add new pillar to end Add new pillar between

Free movement Move in vertical Move along line tangent only

@petekcompany


Fault Modeling Automated trimming of fault

Before

After

@petekcompany


Fault Modeling Connection of faults

Connect two faults Disconnect two faults

Select one pillar of each fault to be connected

Click connect two faults

A connected key pillar be gray color

@petekcompany


Fault Modeling Automatic Connection of faults

Extend distance: will search for pillars within the given distance and connect them

Remove distance: will remove pillars that are less than the given distance from the connection

@petekcompany


Fault Modeling Automatic generation of fault Make sure fault modeling in process diagram be active

Right click on the folder of fault sticks or polygon and select convert to fault in fault model

Make folder in fault model and move all the similar fault in it

@petekcompany



Zonation & Layering

Well Correlation

Facies Modeling

Fault Modeling

Pillar Gridding



@petekcompany


Pillar Gridding Make a grid based on the define faults.

The pillars are inserted in between the faults, and there will be one pillar in every corner of each grid cell.

The increment of the grid in the I and j direction are set in the process.

The result of pillar gri dding is a “skeleton grid”, defined by all the faults and all the pillars.

The skeleton is a mesh grid consisting of a top, a mid and a base skeleton grid.

Create grid adjusted to the mid point of the key pillars.

Extrapolate the pillars to the top and base shape points.

@petekcompany


Pillar Gridding Pillars

Pillars + Edges

A 3D grid is a 2D grid mesh extended into the third dimension. A 2D grid mesh is defined by rows and columns oriented in the X and Y direction.

Pillars + Edges + Base & Top = Cells @petekcompany


Pillar Gridding The pillar gridding process Purpose: Create the skeleton grid for the 3D model. Limit the model in the I and j direction

Input: Fault Model Tasks: 1.

Build boundary

2.

Insert trends and directions

3.

Decide on the increment in I and j direction

4.

Plan the segments

5.

Build the mesh grid

@petekcompany


Pillar Gridding Boundary: Polygon, boundary segment or part of boundary. The faults themselves can be set as part of the boundary.

Trend: Guidance for the grid cell shape and orientation. They can be Defined connecting one fault to another, along faults or in between faults. Trend can not cross faults.

Directions: Guidance for orientation of cells along faults.

Segments: Faulted compartment closed by faults and/or grid boundary.

@petekcompany


Pillar Gridding Workflow 1.

Start pillar gridding process

2.

Select the faults

3.

Define the boundary

4.

Define grid name and increment

5.

Build mid skeleton grid (click apply)

6.

Tune mid skeleton grid

7.

Extend pillars to top and base skeletons (click Ok)

@petekcompany


Pillar Gridding Defining a boundary

1- Create boundary (closed polygon) Create boundary segment A boundary must be define around the area of interest. This is done by using a boundary polygon or by creating a more complex boundary involving faults as part of the boundary together with boundary segment

2- Used in combination

3- Convert from polygon ◦

Active the pillar gridding process

◦

Active the correct fault model

◦

Right click on the project boundary in the input tab

◦

Select convert to boundary on the active fault model

Set part of grid boundary

@petekcompany


in pillar gridding when we define the trend, the trend I and j do cross together. To create the 3d grid we must to define two trend in each border and it is better to define i , j in our boundary.

@petekcompany


Pillar Gridding Directions and Trends When the fault geometries are complex, you might need to help the gridding process by giving directions to fault and trends in the grids.

I-Trend J-Trend

(Flow-Fracture-Fault Direction are define trend) First select the I and J trends, then simply digitize the trend where you want to place it.

I-Direction J-Direction Arbitrary- Deselecting a fault direction Automatic direction assignment

@petekcompany


Pillar Gridding Number of cells

Highlighting a trend (endpoints will be become Yellow)

Click on the set number of cells on connection icon

Specify number of cells. Click apply to re-run pillar gridding with the updated changes

@petekcompany


Pillar Gridding Defining Segments

Set part of grid boundary Set part of segment boundary

Set no boundary Set no fault

Faulted compartment closed by faults and/or grid boundary

@petekcompany


Pillar Gridding Edit fault model in the pillar gridding process Activate pillar gridding process and select two neighboring pillars (yellow)

Click on the add pillar between icons

See the result in fault model in 3D window (add pillar between two pillars)

@petekcompany


Pillar Gridding QC – Intersection Open 3D window and display one of the skeletons in addition to an I and/or J intersection

Change the style settings for intersections double click intersections and to

Highlight one of the intersections and use the grid player to move in the I and j directions to look for poorly truncated pillars or spike in the skeleton grid

@petekcompany


Pillar Gridding QC – Segments

Double click on the skeleton folder and toggle show as segments

To check segments use the segment filter to toggle on/off or make a visual display of segments as a legend by right clicking on the segment filter selecting create/ update template

@petekcompany



Zonation & Layering

Well Correlation

Facies Modeling

Fault Modeling

Pillar Gridding



@petekcompany


Zonation & Layering Make Horizon

Depth Conversion

Zonation & Layering Make Zone Layering

@petekcompany


Horizon tab

@petekcompany


Setting tab

A select an algorithm filling in gaps (towards fault/undefined areas). Toggle Convergent gridder (better for faulted horizons)

Collapse zones to zero thickness if less than a specified thickness.

@petekcompany


Fault tab

Set distance to fault: this option allows the user to specify a distance from the fault in the project unit. You have the option to set different distance on the front and backside of the fault.

Displacement: allows the user set the maximum and minimum displacement along the fault. Fault

Distance Surface

@petekcompany


Fault tab Fault distance can be different in front or behind the fault (Yellow or Blue)

Press the icon to view the blue/yellow faults in a 3D

@petekcompany


Fault tab

Adjust distance and displacement setting for individual faults: 1.

Make sure that the Use default option is toggled off,

2.

Expand the default for each fault folder in the make horizons work window and select the fault you want to adjust.

3.

After it adjust the distance and displacement setting on selected fault: click apply

@petekcompany


Horizon extrapolation- Distance to fault (problems)

@petekcompany


Fault- horizon intersection

Geometrical relationship between the fault and the horizon. Can be used as input in the make horizon and scale up zone processes. Creating from: 1.

By re-sampling from a previously defined 3D grid

2.

By digitizing them directly on to fault key pillars in a 3D window

3.

From input polygons

@petekcompany


Fault- horizon intersection Open the setting for the horizon to be edited and go to operations tab Select resample from 3D grid to fault model and press resample. A new horizon line folder will appear

@petekcompany


Objects that can be depth converted are: 1.

Surface

2.

Horizon interpretation

3.

Faults

4.

Points

5.

Well logs

6.

Well tops

7.

Seismic data

8.

3D grids

Depth conversion of a 3D grid is a 2 steps process: 1.

make a velocity model

2.

Depth convert the grid using the velocity model.

@petekcompany


The first step of depth conversion process is to create velocity model. The second step is; where the initial time grid is chosen together with the velocity model and the pillar re-creation setting.

Velocity model workflow 1.

Define zone: insert surface/horizon or constant

2.

Define correction :insert well top

3.

Define velocity equation

4.

Enter input for the velocity model

@petekcompany


Define zone

Create new velocity model. (time (TWT) to depth (Z))

Choose a datum Select seabed from input tab as a input

Select the top horizon from the 3D time grid

@petekcompany


Correction

Input well tops for corrections: 1.

Select none for seabed and well tops for another horizon

2.

Get well tops from input tab, press blue arrow to insert into correction

Correction Option

@petekcompany


Velocity equation

At each location the velocity is constant

(Z0=0) At each location the velocity is changes in the vertical direction by a factor of k. V0 the velocity at datum. Z the distance of the point from datum. typical value for k between 0 and -0.2

For Z0 is not zero.

These should have an attribute representing average velocity between the point in the cube and the datum.

Include the property representing average velocity .such grids can be created by sampling data into the grid or using data analysis and petrophysical modeling to extrapolate from well data.

@petekcompany


Velocity equation- Option for input of V0 and k Constant A constant value Surface Defining value at each X-Y location. cover the whole area of the velocity zone Well TDR (constant/surface) The value will be estimated using the time-depth relationship (TDR) through the zone for each well (constant average or interpolated respectively)

Correction (constant/surface) Data in the correction column will be used to define the value. petrel will find the value at the location of each of th e correction points such that the resultant conversion will match the correction point . (constant average or interpolated respectively)

@petekcompany


Depth convert the 3D grid

@petekcompany


Select the stratigraphic interval

Append the number of zones select isochore as input type

Input isochore and well tops

@petekcompany


Parameter setting

Select build from

Select build along

Top horizon isochors added from top of interval

Vertical thickness (TVT)

Base horizon Isochore added from base of interval

Stratigraphic thickness (TST)

Both base and top horizon Only active if rest zone is defined

Along pillar Use when pillars are close to vertical

@petekcompany


Minimum cell thickness

Use min.cell thickness to collapse all thin cells less than the given length (project unit)

No min.cell thickness Min.cell thickness start from base Min.cell thickness start from top

@petekcompany


Select the zone division (four different type)

Specify the number of layer and cell thickness

Optional input of a reference surface

@petekcompany


@petekcompany


Geometrical Modeling Geometrical modeling: The process where you can use some pre-defined functions to generate generate properties.( e.g. bulk volume, depth, height above contact, and more).

Geometrical modeling can be done when no input data is available values are assigned ass igned based on grid geometry. Note that if logs are available, you can upscale the well logs and do facies/ petrophysical modeling directly.

@petekcompany


Geometrical Modeling Property filter options There are three type of property filters:

1) The I-J-K fifilter It is useful for QC

2) The index filter Skips a users defined amount of cells in all directions

3) The value filter Filters property values If bulk volume has negative value toggle on value filter and select Max=0 to view only negative cell value

@petekcompany


Geometrical Modeling Creating simple facies: 1- in geometrical modeling select create new property 2- selec selectt assign facies between between surface as method method 3- for converted horizons to surface, double click on horizons go to operation and toggle fill in faulted areas and press make surface. 4- selec selectt fluvial facies as property property template template 5- Sele Select ct which facies facies code code to use 6- it is good to open fluvial facies template from the templates templates tab to check that code are correct.

@petekcompany


Scale Up Well Logs process

The up-scaling is theprocess where toassign values tothe cells inthe 3D gridthat is penetratedby the welllogs.

1)

Create new property and select input from well logs

2) Select log from drop-down menu (capture from global well log folder)

3- select the best scale up setting

@petekcompany Log as bias

This optionis usedto ensure that the upscaledvalues for apetrophysical property are appropriate for thefacies property forthe same cell.Porosity canbe biasedto afacies log.


Scale Up Well Logs Averaging methods -The following methods for up-scaling the well logs are available.

Arithmetic mean - Typically used for properties such as porosity, saturation and net/gross because these are additive variables.

Harmonic mean - Gives the effective vertical permeability if the reservoir is layered with constant permeability in each layer. The harmonic mean works well with log nor mal distributions. Used for permeability because it is sensitive to lower values. The method is not defined for negative values.

Geometric mean - Normally a good estimate for permeability if it has no spatial cor relation and is log normally distributed. The geometric mean is sensitive to lower values, which will have a greater influence of results. The method is not defined for negative values.

RMS (Root Mean Squared) - Will provide a strong bias towards high values.

Most of - Will select the discrete value which is most r epresented in the log for each particular cell.

Mid Point Pick – Will pick the log value where the well is halfway through the cell. This is essentially a r andom choice and is therefore more likely to give a pr operty with the same distribution of values as the original well log data.

Random Pick – Picks a log point at random from anywhere within the cell. This random option avoids the smoothing tendency of other methods and is therefore more likely to give a property with the same distribution of values as the original well log data.

@petekcompany


@petekcompany


Scale Up Well Logs Method (Scale up well logs) Simple - All cells the well trajectory goes through (touches).

Through cell - The well trajectory must go through two opposite cell walls (top and base - opposite sidewalls) of a cell for the cell to be included.

Neighbor cell - This option will average log values from all cells immediately adjacent to the upscaled cell and belonging to the same layer as the upscaled cell. Therefore if there are three adjacent cells along the well path which belong to the same layer, the first will get an average value of the logs from cells 1 and 2, the second from cells 1, 2 and 3 and the third from logs within cells 2 and 3.

@petekcompany


Scale Up Well Logs Setting- treat log

As point: Only the points inside the cell will be used as point for the average value given to the cell.

As line: All points along the well trace that have part of their inside the cell will be used.

@petekcompany


Scale Up Well Logs Qc of upscaled logs

1- Press the show results in well section or set up the well section display

1-Go to make log tab of setting for wells

2- toggle raw logs from global well logs and upscaled property from properties folder

2- select the property to make log from

3- View the logs and upscaled property

3- press make log button, the log is stored in the global well logs folder and can be toggled on from there

@petekcompany


Scale Up Well Logs Statistic check of the scaled up well logs

@petekcompany


Data Analysis Histogram, CDF (cumulative distribution function)



and Crossplots

Open the window menu bar and select new histogram window

Select perm log from global well logs

Show CDF curve icon

Click on the show viewport setting And change the data range.

In the setting press log value

@petekcompany


Data Analysis Histogram, Cdf and Crossplots

Open function window

Select phi on the X axis and perm on Y axis

Display the perm axis to logaritmic

Calculate the correlation coefficient between the two properties by clicking the make linear function from cross plot

@petekcompany


Data Analysis- Variogram Identify any regional trends in data analysis before you begin the variogram analysis

Two aspects to a variogram: 1- how similar are two values right next to each other 2- how far away do points have to be before they bear no relation to each other. Use spherical model- unless you have good reasons for a model change

@petekcompany


Data Analysis- Variogram Measure variability with distance Large distance=Large variability Calculated in 3 directions: -Horizontal major -Horizontal minor -Vertical

HOLE EFFECT

@petekcompany


Data Analysis- Variogram Direction in variogram The goal of horizontal variogram analysis is to: ◦ ◦

Determine if anisotropy is present Quantify the degree of anisotropy in term of major and minor variogram model ranges.

The vertical ranges in stratified geology are usually a fraction of those identified for the horizontal direction. This vertical vs. horizontal anisotropy is to be expected.

Vertical variogram ◦

Plenty of data

◦

Easily estimate

Horizontal variogram ◦

Usually very little data

◦

Usually implied from geology knowledge

◦

Can be derived from correlated data source

@petekcompany


Data Analysis- Variogram Direction in variogram Often well data is far too sparse to facilitate variogram modeling in the horizontal direction. for example: what if you have a single exploration well. This situation support only vertical variogram analysis, no pairs are available in the horizontal panel. It is common to turn to a correlated secondary source of data. Given a reasonable correlation, one can justify the horizontal analysis on the secondary data. this data is used as a proxy or substitute for the purpose of interpreting the direction and major and minor range values.

COMMENT The vertical modeling should be done on raw log data and make sure the simbox is turned off.for lateral modeling the upscaled logs should be used and the simbox should then then be on. @petekcompany


Data Analysis- Variogram Variogram map and sample variogram generation in Petrel Variogram map: Good for visualizing anisotropy and its direction

Sample variogram: Good for finding major and minor range horizontally

Other variogram type for point data and surfaces: -Classical variogram: used as default in petrel -Pairwise relative: each pair is normalized by the square average -Logarithmic: logarithmic values used instead of original values. -Semimadogram: uses the absolute difference instead of the squared difference.

@petekcompany


Data Analysis- Variogram Variogram map calculation of point data set Open setting for the point data Go to variogram tab and select variogram map and change the setting under the XY range tab Click the execute button to calculate variogram map Open map window and display the variogram map. eventually push the view all in viewport icon

@petekcompany


Data Analysis- Variogram Sample variogram calculation and Define a variogram model 1.

Generate sample variogram on selected parameter and define the settings then press the execute

2.

Display the sample variogram in a function window.

3.

Click on the make variogram for sample variogram icon. define the model, range and nugget for display sample

4.

Right click on the variogram model and select setting, the major and minor range are display.

@petekcompany


Data Analysis- Kriging Simple Kriging Ordinary kriging

Default kriging algorithm in petrel is simple kriging

@petekcompany


Data Analysis- Simulation Principle Normal score Stationary( mean of the data does not change) No trend

Note: The make/edit surface process automatically transforms input data to normal score. The simulation results are backtransformed automatically In the property modeling a normal score transformation is used automatically as well, unless you see the transformations in the data analysis process.

@petekcompany


Data Analysis- Simulation The interaction between a CDF and a variogram Before running a SGS, the variogram and distribution setting needs to be specified. The variogram gives the range, Azimuth and etc. A CDF will be set up from the normal scored data. However it is possible to change the CDF, by changing the output data range.

Influence of variogram model type Exponential and spherical model give similar results.

Gaussian model gives smooth result

@petekcompany


Data Analysis-

Setting for Property Modeling

Condition with facies

@petekcompany


Data Analysis-

Proportion : vertical facies variation

Thickness: individual facies interval thickness

Setting for Facies Modeling

Probability: calibration with seismic attribute (AI). Relation between continuous and discrete logs

Variogram: spatial facies continuity @petekcompany


Incorporate the maximum amount of data Well Data

Seismic Data

Production

Outcrop

Other Geological Studies

Integrated Study Structure (Horizon, Fault) •Stratigraphic Modeling •Facies Images •

Deterministic Information

Sedimentological Model •Facies Description •Connectivity •

Conceptual Information

Statistical Information

Histogram Variogram •Correlation •Trend •Variation • •

@petekcompany


The main division in the modeling algorithms available in Petrel is between deterministic algorithms and stochastic ones.

Deterministic algorithms will always give the same result with the same input data. Stochastic algorithms on the other hand use a random seed in addition to the input data, so whilst consecutive runs will give similar results with the same input data, the details of the result will be different. There are of course advantages to both of these methods. Deterministic algorithms will generally run much quicker and are very transparent – it is easy to see why a particular cell has been given a particular value. The disadvantage is that models with little input data will automatically be smooth even though eviden ce and experience may suggest that this is not likely. Getting a good idea of the uncertainty of a model away from the input data points is often difficult in such models.

Stochastic algorithms such as Sequential Gaussian Simulation are more complex and therefore take much longer to run, but they do honor more aspects of the input data, specifically the input data’s variability. This means that local highs and lows will appear in the result which are not steered by the input data and whose location is purely an artifact of the random seed used. The result will therefore have a distribution more typical of the real case although the specific variation are unlikely to match. This can be useful, particularly when taking the model further to simulation as the variability of a property is likely to be just as important as it’s average value. The disadvantage is that some important aspects of the model may be random and it is therefore important to perform a proper uncertainty analysis with several realizations of the same prop erty model with different random seeds.

@petekcompany


Reliable permeability data is rarely available to the same extent as porosi ty data and as there is often a relationship between these two properties, it is common practice to base permeability model directly upon the porosity model. There are a number of ways to do this which vary in complexity as well as in requirements in data and time. When evaluating which option to use, it is important to consider how much data is required to make the result meaningful and how much benefit the additional complexity will bring.

Simple linear relationship The simplest method of generating a permeability model from a porosity model is to use a straightforward transformation from a porosity / permeability cross plot. This can be useful if the relationship is relatively good and there is little input data or if a rapid result is required for early stage assessment. a random element can be added via a normal distribution to generate a simple cloud type transform, moving away from the unrealistic straight linear type relationship. Incorporating porosity values in the equation i.e. in the definition of the standard deviation, allow the user to create more complex transformations. One major drawback of this method is that whilst a statistically reasonable fit may be achieved, the distribution does not take spatial continuity into account. i.e. cells with very high values of permeability may be located next to cells with low values. This is unlikely to have an important effect in reservoirs with very good or very poor permeability distributions but may have a marked effect in reservoirs with important spatial trends, particularly if the model is used for simulation purposes. The cross plot to the left shows the relationship between porosity and permeability in the upscaled cells and a best fit line through those, whilst that to the right shows the modeled porosity and permeability using the best fit multiplied by a normal distribution.

@petekcompany


How to generate a Permeability model from a linear poro-perm relationship 1. Upscale porosity logs and model the porosity model as normal. 2. Upscale permeability logs. 3. Open a function window and plot Porosity (X) against Permeability (Y), displaying upscaled cells only. 4. Use and to make the axes logarithmic if required. 5. Press Make Linear Function to generate a best fit function through the cross plot points. The function will appear in a function folder in the input tab. If one of the 6. Rename the function as PoroPerm. 7. Open the property calculator and type Permeability = PoroPerm (Porosity).

To generate a simple cloud type transform use the equation Permeability = PoroPerm (Porosity)*Normal(1,0.15) where 0.15 is the standard distribution to apply at each cell.

A teardrop shape with greater variation at higher porosity values can be added by making the standard deviation a function of the normalized porosity i.e. Permeability = PoroPerm(Poros ity)*Normal(1,0.15*(porosity/0.2)) where 0.2 is the average por osity.

@petekcompany



Zonation & Layering

Well Correlation

Facies Modeling

Fault Modeling

Pillar Gridding



@petekcompany


.

.

.

@petekcompany


Facies Modeling •

Overview Deterministic

Interactive drawing, seismic volume extraction, indicator kriging

FACIES MODELING

Pixel base

Indicator simulation

Object base

Facies with defined shapes, object, fluvial

Stochastic

@petekcompany


Facies Modeling Overview If well logs are upscaled they can be used in deterministic and stochastic modeling. if no logs are available, only used unconditional stochastic method and interactive drawing

Two modeling setting buttons are available: 1.

Common

2.

Zone setting

@petekcompany


Facies Modeling Common and Zone setting Common setting for all zones are entered and the zone setting for specific zone .

Common tab 1.

Use filter

2.

Ensure that all cells ge t value

3.

Number of realization

Zone setting ◦

◦

Define zone setting for all zone Leave the zone button off ◦ ◦ Press the leave zone Use one algorithm for all zones ◦ Define zone setting for each zone ◦ Press the zone button ◦ Select the interest zone Select the method and algorithm ◦

@petekcompany


Facies Modeling Sequential Indicator Simulation

Make sure to pick the upscaled property and select SIS method

Select the facies from the template

If do the data analysis click the data analysis icons, if do not it do it at this setting

@petekcompany


Facies Modeling When we select this option only use the property that was upscaled and be in the property folder. When we do not select this option we will use all of the property in the template window

When we select the use existing Property we can define the facies Template and we do data analysis Just select this button.

When we did data analysis and variogram modeling we select these buttons and do not need do another data analysis

@petekcompany


Facies Modeling Object modeling

-Select the correct property - select object modeling method

- Select the fluvial channel icon to insert channel body -Select facies property to mach channel - and levee

NOTE: -The object modeling combine to another facies modeling at the same zone in the background -Ellipse is the part of body lithology. When we had part of body we select the ellipse and rename it on modify name of body

@petekcompany


Facies Modeling Object modeling- channel setting Layout Specify channel direction, wavelength and amplitude

Channel Specify channel width and thickness

Levee Specify levee width and thickness. Levee smaller than channel

@petekcompany


Facies Modeling •Object modeling- Ellipse setting Geometry -Select the body shapes from drop d awn menu - set the orientation, width and thickness

Rules -Specify whether the facies will replace other facies or not.

Trend Vertical (function curve) Area (probability map)

@petekcompany


Facies Modeling Object modeling- background

When we do not know what is the background When the back ground is the one lithology When we had another lithology in the background and the channel is the one of the lithology

@petekcompany


Facies Modeling Interactive facies modeling

Brush type Facies type Radius

Profile @petekcompany


Facies Modeling (Truncate Gaussian Simulation Algorithm) • Facies transition simulation (Truncate Facies transition simulation allows a stochastic distribution of the facies based on given transition between facies

@petekcompany


Petrel Wor orkfl kflow ow Too ools ls Intro

Zonation & Layering

Well Correlation

Facies Modeling

Fault Modeling

Pillar Gridding



@petekcompany


Petrophysical Modeling Common and Zone setting

@petekcompany


Petrophysical Modeling Sequential Gaussian Simulation Method

@petekcompany


Petrophysical Modeling QC Result

QC result in histogram -Go to setting and select histogram tab check that the histogram fallow the distribution 1- Raw Data 2- Up scaled cells 3- 3D grid

Filter 4- Use zone filter 5- Filter on other property value by pressing the filter button and go to property f ilter in setting for the properties folder

@petekcompany


Petrophysical Modeling Property Calculator This calculator can be used to create new 3D properties and do operations between properties

Creating a new property model:

Right click on property folder in the 3D grid and select calculator Change the property template to NET/GROSS and type into the while formula field: NG=0.8 Type in NetVol = Bulk_Volume*NG. select the net volume template. Press ENTER.

When we do not any template, generate one Some of property we generate them on Geometrical Modeling

@petekcompany



Zonation & Layering

Well Correlation

Facies Modeling

Fault Modeling

Pillar Gridding



@petekcompany


/ petek-co.ir

:


/ petek-co.ir

:

Course Title Of petro Energy Kimia

)

( 0000

0

00000

0

0000

7

9

EX-09

92

RE-27 & 30

6

11 011

E-06

Que$tor

49-4 60000

EX-25

Petrel

200000

EX-42

00000 )

49(

EX-48

)Visage(

4 -94 0000

EX-12

97 -92

) DFN(

700000

EX-27

600000

-

CMR Geolog

7298002346 :

729648340 -8 :

233

Info@Petek -co.ir, Training@Petek -co.ir :

:

www.Petek -co.ir :

EX-16

Course Title Of petro Energy Kimia

00000

DR-07

6 -

400000

4 -46

Top

DR-37

Drive

700000

99-9

DR-08

4900000

RE-07 )ESP(

)IADC WellSharp(

No

Training Course

1 2 3

Case 1: IADC WellSharp Certification Introduction Course Driller Course Supervisor Course

Training Duration Location

Min. Batch Size

Price per head in USD $

Tehran Tehran Tehran

4 days 5 days 5 days

10 10 10

1,450 1,800 1,800

Tehran Tehran Tehran

4 days 5 days 5 days

10 10 10

1,450 1,600 1,600

Case 2: In-house Certification 1 2 3

Introduction Course Driller Course Supervisor Course

7298002346 :

729648340 -8 :

233

Info@Petek -co.ir, Training@Petek -co.ir :

:

www.Petek -co.ir :

Static Modeling by Petrel

Recommend Documents