Initial data analysis Modeling fracture parameters
Fracture Modeling Course Introduction Overview What is Fracture Modeling? Naturally Fractured Reservoirs Fluid Flow Simulation Models Fracture Modeling approaches Fracture Modeling Workflow Data Set - Location Data Set - Geological description –
–
Stratigraphy/Mechanical zones Fractures
Data Set - Comparative Outcrop studies
What is Fracture Modeling? Purpose and Process Purpose
Create simulation properties for matrix and fractures to be able to predict reservoir behavior
Why?
Many reservoirs are dual porosity/dual permeability (Naturally fractured); leading to high flow zones not representative of the matrix flow capacity Flow simulators have problems simulating these kind of reservoirs.
Process
Multi-disciplinary approach; Use analyzed fracture data from wells Building a Fracture model (DFN+IFM) Upscale fracture permeability, porosity and connection factor between matrix and fractures from the Fracture model These data can subsequently be simulated
Naturally Fractured Reservoirs Simple Classification of Reservoir types I. Fractures provide essential Porosity and Permeability –
Requires large reservoir tank or thick pay zones to be economical (no matrix porosity)
II. Fractures provide essential reservoir Permeability –
Most reservoirs with storage in matrix but low matrix permeability
III. Fractures assist Permeability in already producible reservoir –
Higher porosity lithologies
IV. Fractures provide no additional Porosity/Permeability –
Fractures act as Flow Barriers
100% KF . m r e P l a t o T f o %
II III
IV
I
Naturally Fractured Reservoirs Example of Reservoir types MATRIX DISCHARGE
I. Fractures provide essential Porosity and Permeability
II. Fractures provide essential reservoir Permeability Fluid communication from Matrix to Fractures is important Fracture Morphology essential !
–
–
III. Fracture assist Permeability in already producible reservoir IV. Fractures provide no additional Porosity/Permeablity
Crossflow
No Crossflow
M F
M Morphology M to F communication Good Recovery Factor Good waterflood sweep efficiency
Morphology Restricted communication Poor Recovery Factor in tight Matrix Poor waterflood sweep efficiency
•
•
•
•
•
•
Fluid Flow Simulation Models How to approximate nature? Reality captured in 3D Models
Ideally hydrocarbon flow takes place in a Single Porosity / Permeability system However in Dual Porosity reservoirs, fluids exist in two interconnected systems (matrix and fractures). This must be accounted for in Simulation models.
A simplification of the real reservoir is done when creating a dual porosity model Fluid flow and transport exist in both the connected fractures and matrix blocks Two overlapping continua, where both are treated as porous media
Dual Porosity model types
Simple layer model (sheet of parallel fracture sets) Matchsticks model (2 orthogonal fracture sets) Sugarcube model (3 orthogonal fracture sets)
Layered Model
Match Stick Model
Sugar Cube Model
Real Reservoir
Fluid Flow Simulation Models DFN vs. Dual Porosity models DFN Model Non Uniform Geometry Variable Fracture Orientation Variable Fracture Length Variable Aperture - -> Variable Intensity and Interconnectivity
Fluid Flow Simulation Models Standard approaches to fracture modeling Equivalent Non-Fractured Medium
Equivalent Continuum –
Bulk response for equivalent porous media
Layered Model
Dual Porosity (DP) –
Separate Matrix and Fracture blocks
Discrete Fracture Network (DFN) – –
Physical fracture representation Upscaled to Dual porosity properties
DFN Model
Real Fractured Medium
Fluid Flow Simulation Models Petrel 2010 approach to fracture modeling
Property Model
Implicit Fracture Model (IFM) –
–
Yields directly fracture porosity and permeability as properties Upscaled to Dual porosity properties
Real Fractured Medium
Combined Model
Discrete Fracture Network (DFN) – –
Physical fracture representation Upscaled to Dual porosity properties
DFN Model
Fracture Modeling Workflow Petrel – Overall Fracture modeling workflow Well data
Data Analysis
Model Parameters
Upscale & Simulate Create Fracture model
Fracture Modeling Workflow Petrel – Specific Fracture modeling processes DFN
Fracture intensity
IFM
Hybrid IFM / DFN model
Data Set Teapot Dome – Wyoming (USA) Teapot Dome is located in central Wyoming. A comprehensive Data Management project has been conducted to digitize USAand compile all available data. Data is available e.g. for research and software testing/training.
Achnowledgements: Thanks to Rocky Mountain Oilfield Testing Center and U.S. Department of Energy for using Teapot data
Data Set – Stratigraphy (Outcrops @ Alcova anticline) N
Cretaceous Measverde Fm
Teapot Sandstone Parkman Sandstone
East
Unit 5: Unit 4:
Unit 3:
West
Unit 2:
Unit 1:
Fluvial Ss Non-Marine Carb. Sh with localized coal White Beach Ss
1
Shoreface/Beach Ss
Shallow Marine Interbedded Ss and Sh
10m
2 1 Section Location, Number Quaternary Alluvium Mesaverde Fm
NPR3 Boundary
0
Reworked from: S.Raeuchle et al, 2006 and Cooper, S. 2000
3
Undifferentiated 0
Carboniferous Tensleep Fm
5
1 km
4
Data Set – Mechanical Zones (Mesaverde Fm. Outcrops) Mechanical zones Separating units according to mechanical properties is important due to mechanical influences on fracture characteristics.
Generalized Stratigraphic column – Parkman Sandstone Mb. (Mesaverde Fm.) Fluvial Ss Non-Marine Carb. Sh with localized coal 3: White Beach Ss
N
Unit 5: Unit 4:
Unit
Unit 2:
Unit 1:
1
Shoreface/Beach Ss
Shallow Marine Interbedded Ss and Sh
10m
Compiled from Mallory et al., 1972; Spearing, 1976, and Rocky Mountain Oilfield Testing Center field data.
5
3
Undifferentiated 0
From: Cooper, 2000; Cooper et al., 2001, 2003.
2 1 Section Location, Number Quaternary Alluvium Mesaverde Fm
NPR3 Boundary
0
1 km
4
Data Set – Mechanical Zones (Tensleep Sst. Outcrops) Stratigraphic systems Separating units according to stratigraphic architecture is also important for prediction of complex fracture development in low-complex reservoir facies.
Compiled from Zahm & Hennings, 2009 (AAPG Bulletin)
Data Set – Fracture Intensity (Tensleep Sst. Outcrops)
Fracture intensity at multiple scales High variability in fracture intensity was demonstrated, caused by original depositional architecture, overall structural deformation and diagenetic alteration of the host rock. Fracture intensity depends on stratigraphic scale.
1. Throughgoing fractures
3. Facies Bound fractures
2. Sequence Bound fractures
4. Lamina Bound fractures
Compiled from Zahm & Hennings, 2009 (AAPG Bulletin)
Data Set – Faults at Teapot Dome (Outcrops) Map of faults and representative hinge-perpendicular fractures
Map of faults and representative hinge-parallel fractures
Data Set – Fractures at Teapot Dome (Outcrops) Fracture map of a pavement surface Illustrating the nature of throughgoing fractures and cross fractures at the top of a single sandstone bed at Teapot Dome N
Throughgoing fractures
Cross fractures
covered
0
Illustrations from: Cooper, 2000
1m
Conceptual 3D model of fracture outcrop patterns developed at Teapot dome.
Data Set – Fractures related to Lithology (Outcrops) Throughgoing fractures N Fluvial Ss Non-Marine Carb. Shwith localized coal Unit 3: White Beach Ss Unit 5: Unit 4:
Unit 2:
Shoreface/Beach Ss
Unit 1:
Shallow Marine Interbedded Ss and Sh
10m
0
Quaternary Alluvium Mesaverde Fm
A
Undifferentiated Charted Locality NPR3 Boundary
N 0
n = 24
1 km
Rotation to Fold Hinge
n = 23
Illustrations from: Cooper, 2000
B
Data Set – Infer Outcrop observations to subsurface 3D models? Surface outline (boundary) of subsurface 3D grid