Geological Modeling Lessons Learned Serdar Kaya, Please contact for any suggestion, comments or contribution [email protected] Outline 29 April 2015 X Systems LLC - Confidential 2 ⢠Important Data and QC Approach ⢠Structural Framework ⢠Depth Match ⢠Structural calibration ⢠Fault interpretation QC ⢠Grid parameters ⢠Property Modeling ⢠Stationarity Analysis ⢠Porosity Reconciliation ⢠Cloud transformation ⢠Multiple realization of property models ⢠Blind Testing ⢠Uncertainty Analysis ⢠Uncertainty cases Data Oriented Approach 29 April 2015 X Systems LLC - Confidential 3 Lesson : Incomplete Data Set Solution: Comprehensive Data Analysis ⢠Complete Data Set with Every Available Data â All penetrating and surrounding wells â Shallower wells â QC data consistency ⢠QC results after every step â Well data in cross sections â Check average maps, histograms â Compare results with raw data Structural Modeling 29 April 2015 X Systems LLC - Confidential 4 3D Grids Depth Shift Cored Wells, Horizontal and Pseudo Wells Update Input Database (Well, core, log, seismic) Sub-zone picking Generate isochore Adjacent wells Inconsistency Generate Sub- zones Generate Layers Generate Reservoir Top Horizon Generate Fault Model Lesson: Outliers in data set Solution: Right workflow Structural Modeling Parameters 29 April 2015 X Systems LLC - Confidential 5 Fault Modeling Gridding Horizons Subzones Layering Structure calibrated with markers Grid orientation, Zig-zag fault, Grid size Fault Data for modeling Subzone well picks and isochores Layer properties Lesson : Inconsistent 3D Framework Model Solution: Right Grid Parameters Core Data Points Shifted Depth Shift of Core Data Lesson: Mismatching flow characteristics Solution: Right depth for core data Depth Correction of Adjacent Wells 29 April 2015 X Systems LLC - Confidential 7 Lesson : Inconsistent Marker Depth Solution: Depth Shift Due to LWD â Wireline Uncertainty Before Depth Shift of Horizontal Well After Depth Shift Structural Correction for Horizontal Well Trajectory 29 April 2015 X Systems LLC - Confidential 8 Lesson : Inconsistent Horizontal Trajectory Solution: Structural Control with Pseudo Wells Before Calibration Calibrated Structure Structural Modeling 29 April 2015 X Systems LLC - Confidential 9 Lesson: Inconsistent SubZone Features Solution: Right Picking Structural Modeling 29 April 2015 X Systems LLC - Confidential 10 ISOCHORE POINTS Top Structure Lesson Learned: Inconsistent thickness variation Solution: QC markers with isochore Structural Modeling 29 April 2015 X Systems LLC - Confidential 11 Lesson : Inconsistent thickness variation Solution: Subzonation Process, Thickness Control Markers Isochore map controls at inter well locations Structural Modeling 29 April 2015 X Systems LLC - Confidential 12 Summary of Troubleshooting for Framework Modeling Top structure calibration Surrounding wells Shallower formations Incorporating the horizontal wells Depth match to pilot Structural control along the trajectory To control thickness use isochore maps Grid Characteristics Orthogonal grids Fixed orientation Zig Zag faults Property Modeling 29 April 2015 X Systems LLC - Confidential 13 Lesson: Inconsistent property variation Solution: Capture Variation with Right Trend Average Porosity at Well Location Porosity Modeling Workflow 29 April 2015 X Systems LLC - Confidential 14 Scale-up porosity to 3D Model SGS with Collocated Co-Krigging to Porosity Map Porosity Model (log porosity reconciled to OB corrected core porosity) KriggingSGS SGS with Data Range Variogram SGS with 5 Km range Krigging with Data Range Variogram 1 2 4 Krigging with 5 km Variogram Krigging Model as 2D Trend Krigging Model without trend Krigging Model with modified trend 5 SGS with 30 Km range Krigging with 30 km Variogram 63 Lesson: Limited Property Models Solution: Scenario Based Modeling Workflow Porosity Modeling 29 April 2015 X Systems LLC - Confidential 15 SGS with Trend Krigging 3D Model Trend Krigging Modified Map Trend 2013 Model Lesson: Defining Accurate Realization Solution: Blind Test on Removed Well Data Porosity Realization Error Mean Std SGS with Trend 0.0159 0.127 Krigging Model Trend 0.0146 0.0124 Krigging Modified Trend 0.0135 0.0109 2013 Porosity Model 0.0293 0.0249 Flank to Crest Well Comparison Permeability Modelling 29 April 2015 X Systems LLC - Confidential 16 Lesson: Limited Property Models Solution: Scenario Based Modeling Workflow Scale-up permeability to 3D Model Permeability (10 wells) (Distribution on Phie-K Cross Plot per sub-zone) GRFS with Porosity Model Trend GRFS without any TrendGRFS with 2D Porosity Map Trend Data range Variogram 5 Km range variogram 1 Cloud Transformation 2 30 Km range variogram Data range Variogram 5 Km range variogram 4 5 30 Km range variogram Data range Variogram 5 Km range variogram 7 8 30 Km range variogram 3 6 9 Layer 4 â Subzone C Edge effect of algorithm Saturation log problem Saturation Modeling â QC of Log Saturation Lesson: Incorrect Log Interpretation Solution: Generate Kriged Sw Model QC the logs Saturation log problem Crest Wells with 100% Water Saturation Saturation Modeling 29 April 2015 X Systems LLC - Confidential 19 Lesson: Accuracy in Saturation Model and Data Solution: Saturation Model Based on J Function and MICP data W7 A A` GDT: 8600 ftss FWL: 8700 ftss W1 W2 W3H W4H W5H W6H Saturation Modeling 29 April 2015 X Systems LLC - Confidential 20 Crest Wells Transition Water Zone Lesson: Accuracy in Saturation Model and Data Solution: Saturation Model Based on J Functions Property Modeling 29 April 2015 X Systems LLC - Confidential 21 Data reconciliation and consistency Trend investigation Multiple realizations Extensive QC steps, and statistical analysis Histogram Cross Plots Maps Cross Sections Blind test by removing well data from data set Summary of Troubleshooting in Property Modelling Model Upscaling 29 April 2015 X Systems LLC - Confidential 22 47 Layers 13 Layers 23 Layers Lesson: Upscaling Course Grid Solution: QC with Cross Sections Upscaled 13 and 23 Layers vs. Geomodel (47 Layers) Porosity Model Uncertainty 29 April 2015 X Systems LLC - Confidential 23 Scale-up porosity to 3D Model SGS with Collocated Co-Krigging to Porosity Map Log Porosity (316 wells) (reconciled to OB corrected core porosity) Porosity Model by KriggingPorosity Model by SGS SGS with Data Range Variogram SGS with 5 Km range Krigging with Data Range Variogram 1 2 4 Krigging with 5 km Variogram Krigging Model as 2D Trend Krigging Model without trend Krigging Model with modified trend 5 SGS with 30 Km range Krigging with 30 km Variogram 63 Lesson: Missing Uncertainty Steps Solution: Scenario Embedded Workflow Scenario Based Uncertainty Krigging versus SGS Trend Variogram Seed number Correlation coefficient Porosity Model Uncertainty 29 April 2015 X Systems LLC - Confidential 24 Final porosity model will be obtained by history match process Uncertainty cases are generated on following parameters Modeling algorithm: SGS versus Krigging Trend Variogram Seed number Bulk shift due to core porosity â log porosity difference Lesson: Missing Uncertainty Steps Solution: Scenario Embedded Workflow Permeability Modeling Uncertainty 29 April 2015 X Systems LLC - Confidential 25 Lesson: Missing Uncertainty Steps Solution: Scenario Embedded Workflow Scale-up permeability to 3D Model Permeability (10 wells) (Distribution on Phie-K Cross Plot per sub-zone) GRFS with Porosity Model Trend GRFS without any Trend GRFS with 2D Porosity Map Trend Data range Variogram 5 and 30 Km range variogram 2 Cloud Transformation 1 3 21 3 Scenario Based Uncertainty Cloud definition Trend Variogram Seed number Permeability Model Uncertainty 29 April 2015 X Systems LLC - Confidential 26 Lesson: History Match with Multipliers Solution: Scenario with Cloud Transformation Density of the points Extended limits Upper Reservoir (R1) Porous 1.2 Sub-zone Extended Cloud Case Permeability Reduced Cloud Case Permeability Extended Cloud Case Permeability Reduced Cloud Case Permeability Data Points Model Points Permeability Modeling Uncertainty 29 April 2015 X Systems LLC - Confidential 28 Final permeability model will be decided after history match Uncertainty cases are generated on following parameters Modeling algorithm: SGS with various trends Cloud definition Variogram Seed number Bulk shift due to uncertainty in measurements Lesson: Missing Uncertainty Steps Solution: Scenario Embedded Workflow CONCLUSION 29 April 2015 X Systems LLC - Confidential 29 Data oriented workflow Correction and calibration in early stage Consistent data set Comprehensive QC steps Scenario based uncertainty analysis Flexible workflow to use for data integration