- Home
- Conferences
- Conference Proceedings
- Conferences
EAGE Workshop on Innovative Reservoir Modeling into Digital Proliferation
- Conference date: September 27-28, 2022
- Location: Kuala Lumpur, Malaysia
- Published: 27 September 2022
-
-
Prediction of Facies & Reservoir Properties in Carbonate Reservoir through Geo-body Modelling: Mumbai Offshore Case Study
More LessSummaryThe study area is located on a westerly dipping gentle homoclinal part of Mumbai High-Deep Continental Shelf and has a full coverage of 3D seismic data. Eight exploratory wells have been drilled out of which five wells proved to be oil-bearing with three pays. These pays lie within Oligocene limestone. The crestal well is enigmatic and is characterized by the absence of hydrocarbons.
Geo-body modelling approach has been adopted to characterize the reservoir variability and facies architecture. Two reflectors within the reservoir zones were mapped and window-based 3D-RMS attributes were generated. Based on the integration of seismic attributes with petrophysical studies, three geo-bodies had been extracted within the pay sequences. The extracted geo-bodies were modelled geo-statistically with a proper calibration with seismic attributes and petrophysical properties.
This integrated approach of geo-body extraction and geostatistical modelling is very effective in delineating facies architecture and reservoir heterogeneity. Both seismic and well inputs are efficiently collated to characterize the reservoir. Seismic attribute analysis helped in geo-body extraction and bring the facies architecture and geostatistical modelling helps in exhibiting the heterogeneity. The present study helped in understanding the facies architecture and prediction of reservoir properties within the study area and its potential.
-
-
-
Scenario Modelling using the EMBER Algorithm
By C. DalySummaryThe Ember algorithm was recently introduced with the objective of simplifying and speeding up the modelling process by using ideas from machine learning and from Geostatistics. In this paper, the method is further developed to construct scenario models. These are useful in helping understand the sort of uncertainty that remains in the reservoir by constructing widely differing models which nonetheless honour available data. Since the scenario models can be blended in parametric way, this approach is promising for optimization problems like history matching.
-
-
-
Uncertainty Analysis for Optimal FDP with Limited Data: A Case Study from the J Field
Authors H. Mohd Ali, K.H. Ling and G. SenSummaryMultidisciplinary integration and 3D grid approach provided a meaningful way to study the probabilistic volumetric ranges as compared to a simplistic map-based approach. The static case selection done based on the integrated static and flowline-based connectivity gave the right choice of key subsurface parameters that controlled both the in-place volume and connected pore volume
-
-
-
Probabilistic Machine Learning for Parameters Estimation of Capacitance Resistance Modeling and Uncertainty Assessment.
Authors M. Thiam and A. NakhaeeSummaryIn this study, we summarize mainly the estimation of interwell connectivity using the capacitance resistance model, derived from the work Sayarpour et al. al (2009) , using a probabilistic data-driven approach. The uncertainty associated with the model can be quantified in this way in order to make better decisions. As well as inverse reservoir characterization, the study analyzes and forecasts field performance.
-
-
-
Confining Reservoir and Net-to-gross Uncertainty by Geological Analogues, Turbidite Channel-Levee Reservoir, Deepwater Greenfields, Offshore NW Borneo
Authors M.F. Abdul Kadir, M.A. Omar, Z.Z. Tuan Harith, D. Uli and D.J. ShieldsSummaryConstructing a static reservoir model of deep-water turbidite channel-levee reservoir for a greenfield is often challenging with very limited data available including well, cores and coupled with gas-affected seismic to delineate reservoir architecture and connectivity. There is high uncertainty in lateral extend and net-to-gross (NTG) of an overbank element as upside potential that contributes to uncertainty in hydrocarbon in-place volumetric. Therefore, understanding overbank geometrical relation with NTG distribution from geological analogues is an important aspect of quantifying uncertainty. A robust facies modelling approach was adopted by ensuring both field data and geological analogues are closely linked to capture uncertainty related to turbidite sand lateral extent and quality for the channel, levee, and overbank elements. Application of geological analogues proved to be very crucial with limited data available, which emphasizes the lateral reservoir quality resultant from a better understanding of lateral interpolation limit. The robust modelling approach adopted for the turbidite channel-levee reservoir model for the static model provided a better insight into turbidite reservoir architecture and NTG uncertainty.
-
-
-
Revisited High-Resolution Sequence Stratigraphy and Seismic Geomorphology Perspectives for Turbidite Architectures in Reservoir Model, Offshore NW-Borneo
Authors A. Bera, M.F. Abdul Kadir, M. Mubin, Z.Z. Tuan Harith, D. Uli and T.M.S. Tengku HassanSummaryReservoirs in two deep-water gas fields, identified from 3D seismic DHI, consisted of sheet-sand facies with high net-to-gross and good lateral with deep-water fan depositional system. However, contradictory results were observed in subsequent exploration wells targeting similar bright amplitude features. This triggered the need to re-visit the existing reservoir model based on the turbidite fan lobe complex for the proposed development plan of these gas fields.
Revisit of existing data was conducted using integrated high-resolution sequence stratigraphy and seismic geomorphology analysis to de-risk subsurface uncertainty. High-order chrono-stratigraphic boundaries were defined after the elimination of structural elements and fluid effects on seismic. High-density horizon interpretation was done to generate detailed seismic attribute maps for each turbidite event. Seismic facies integrated with sedimentology analysis from log motif analysis and borehole image interpretation to define turbidite depositional facies elements.
This novel approach has revealed the architecture of deep-water turbidite reservoirs leading to the establishment of a completely new reservoir correlation and conceptual depositional facies model. With this new perspective on the reservoir model, new insights into reservoir complexities have been incorporated for the optimal field development plan and well location optimization.
-
-
-
A Malay Basin Field Reservoir Model Improvement Constrained by Forward Stratigraphic Modelling and Geostatistical Seismic Inversion
SummaryAn integrated Forward Stratigraphic Modelling (FSM) and Geostatistical Stochastic Inversion (GSI) was utilized to improve static model in a field in the Malay Basin where early production wells showed the high uncertainty in oil-originally-in-place, facies distribution and reservoir connectivity.
FSM prediction combined with regional seismic, cores and well log data have provided a robust scenario of reservoir characteristics for static model. The FSM result were compared with the Geostatistical Stochastic Inversion (GSI) for property distribution away from the well control. The dynamic modelling was calibrated to field and wells performance (production history, MDT, DST, etc.) taking into account main remaining uncertainties and risks and evaluation of multiple field development options.
With thorough integrated analysis of A field and its surroundings, integrated FSM and GSI derived static model reflects accurate facies distribution of the area compared with conventional workflows. It was used as an aid for Field A development optimization and increased the probability to find good reservoir facies as proven from findings of recently drilled development wells.
-
-
-
Conceptual Depositional Environment for Static Model Utilizing Seismic Sequence Stratigraphic Method; Field A, Inboard Sabah
Authors M.Z. Ishak and M.F. MostapaSummaryThis geological study was commenced to re-evaluate the previous geological work up to reservoir simulation for potential field development. All previous studies were conducted based on lithostratigraphy methods with datum at SRU for field wide well correlation. As new seismic dataset was made available, seismic sequence stratigraphic interpretation have been utilized to guide well correlation, conceptual depositional environment, structural grid, facies model and subsequently fluid contacts scenarios.
By utilizing sequence stratigraphic methods, new stratigraphic structural grid and facies model has been developed as part of the 3D static model. Area of non deposition and onlap stratal termination during Sand 6–9 deposition as well as sediment direction from seaward dipping clinoforms in the north has been captured for a more robust reservoir lateral connectivity scenario and fluid contacts for reservoir simulation.
-
-
-
M-70 Static Model: Capturing Heterogeneity through Integration of Re-processed Seismic, Dynamic Data, and Conceptual Geology.
SummaryDiscovery of a sizeable gas reserves in M-70 sand from Well-A (2014) & Well-B (2017), the exploration & appraisal well respectively, had led to Field X being economically attractive to be developed and tied-in to a nearby gas hub facility. Peninsular Malaysia’s increasing gas demand serves as a value driver for the development and production of Field X, which include installation of a light-weight structure and drilling of three wells that can produce up to 70 MMscf/d.
Re-processing of the 3D seismic (2010 acquisition) commenced in August 2019 and was completed in April 2020. In general, the re-processing objectives include; to improve fault imaging to further understand compartmentalization of Field X and to produce optimum amplitude response suitable for seismic inversion and AVO work. Interpretation and analysis from the re-processed seismic were used to enhance reservoir characterization & distribution and optimize target locations for the development wells. This paper focuses on the reservoir model update, the techniques applied in utilizing the Vp/Vs attributes as one of the main static model input and the forecasted rates from the model.
-
-
-
A Systematic Methodology and Powerful Data Visualization in Integrating Multidiscipline Analysis in Fluid Contact Determination.
Authors W.M.N. Wan Mohamed Zin and M.S. HendrawatiSummaryThis methodology highlights the improvement in the common multidisciplinary analysis integration gap by utilizing the power visualization of a well fluid ID map (e.g., water-up-to, oil-down-to, gas-down-to) in conjunction with stick plot fluid analysis to comprehend the key uncertainty, opportunity, and risk. Both subsurface data inconsistencies and human mistake are immediately detectable. It also facilitates fluid ID iteration between Geomodeler (GG) and Petrophysicist (PP). This study expands on the issue of the impact of the vertical dimension on fluid analysis and the incorporation of production allocation in minimizing the fluid contact uncertainty with the involvement of the Reservoir engineer (RE).
-
-
-
An Ensemble Approach for Managing Reservoir Uncertainty: A Real Field Application
More LessSummaryIn the current energy climate, it is crucial to minimise the risks associated with drilling and prevent inadequate decision making. Reservoir modelling provides critical insight to highlight these risks and evaluate reservoir uncertainties. However, the challenge to account for multiple uncertainties efficiently remains. Ensemble-based approach has been proposed as an alternative to ‘base-case centered’ methodology, to propagate uncertainty during models’ data conditioning whilst ensuring geological consistency. In this technical paper, we apply the ensemble-based approach to a mature real field. We also use the ensemble-based algorithm - Ensemble Smoother – as the optimisation tool for history matching of ensemble models.
-
-
-
A Novel Horizontal Well Modeling Method for Thin-bedded Reservoir and its Application
More LessSummaryFor thin-bedded reservoirs, complex spatial position relationship between formation surfaces and horizontal wells introduces difficulty for accurate structural modeling, whilst existing algorithms are not good at addressing this issue during structural modeling. Besides, aggregation of horizontal well sampling points in the dominant reservoir results in the data statistics that do not conform to the geological body, sampled data tend to stay within a narrow zone and forms the clustering effect. This study proposes a horizontal well modeling method for thin-bedded reservoirs to improve the accuracy of the structural model and ensure the representativeness of the data. Aiming at the difficulty, the method of layer-by-layer iteration, virtual point control, and local grid refinement is adopted to ensure the thickness variation trend of each thin layer and thus ensure the accuracy of the structural model. For vertical wells with dense well patterns, the data of vertical and horizontal wells whose trajectory inclination angle is less than a certain value are used for data analysis. For horizontal well pattern, the method of data rarefying is adopted to incorporate the horizontal well information to the most degree and reduce uncertainty.
-
-
-
Improving Carbonate Build-Up Reservoir Model by Integrating Novel Process-based FSM Technology, Case study from Central Luconia.
SummaryCarbonate reservoir modelling can be challenging and often result in unsatisfactory outcomes in predicting complex reservoir heterogeneity. An innovative approach to constructing a robust carbonate reservoir model by incorporating a novel process-based forward stratigraphic modelling (FSM) was adopted into the modelling workflow. The FSM result is integrated with geophysical analysis intended to address subsurface challenges related to field performance and optimise well deliverability. The model-building workflow incorporates multi-discipline data into a geologically realistic reservoir model for fluid flow simulations, intending to tackle reservoir heterogeneity and subsurface challenges. The application of novel FSM technology allows for an improved understanding of complex carbonate internal architecture and in turn, develops a more comprehensive knowledge of reservoir heterogeneity. The integrated static model is then carried on to dynamic simulation and the best fit facies model realisation is selected with the criteria that closely match material balance, pressure trend, and flow behaviour. Also, lateral and vertical connectivity uncertainty is addressed by modelling the baffle facies as part of the final static facies model. The innovative and integrated workflow has produced an enhanced capability of the dynamic model to provide a more accurate and realistic prediction of well and overall contributing to much-improved field performance.
-
-
-
Derisking Well Placement through Innovative Geological Driven Modeling for Greenfield Deepwater Turbidite in NW of Borneo
Authors D. Uli, D.J. Shield, M.F. Abdul Kadir and Z.Z. Tuan HarithSummaryBuilding a static reservoir model for a deepwater greenfield that most accurately represents geological data is the starting point in de-risking well placement and optimisation. A sound depositional model and chronostratigraphic control are prerequisites for a robust static reservoir model. The recent study has led to a significant revision of the chrono-stratigraphic correlation and sedimentological model that shifted from a turbidites-lobes system with high reservoir continuity to a more complex turbidites-channels system. This revision caused a substantial impact on reservoir distribution, connectivity, and well placement. Depofacies model was constructed for channel-levee reservoirs based on depositional facies maps inferred from high-resolution seismic sequence stratigraphic analysis, and seismic sedimentology analysis of spectral decomposition seismic attribute. Given the poor relationship between facies and seismic quantitative interpretation results, two static model scenarios were constructed. In the first scenario, due to the limitation of well data especially in levee-overbank, integration of 2D trends from geological analogues was used to govern the net and non-net electro-facies proportion for a realistic base-case net-to-gross. In the second scenario, additional steps were performed to generate sand distribution trends from 3D process-based turbidite modelling as a 2D secondary constrain and incorporated into the electro-facies modelling.
-
-
-
Improved Reservoir Characterization Using Advanced Petrophysical Evaluation and High-Resolution 3D Modeling in Heterolithics Thin Bedded Reservoirs
By S. JohariSummaryAccurate reservoir characterization is a key step in developing, monitoring, and managing a reservoir and optimizing production. Reservoir characterization requires such very high degree of prediction accuracy that any deviation from expectation may result in huge losses and wasted efforts through enormous man-hours and huge investments. The naturally laminated of heterolithic thin-bedded sand make it more challenging in building the representative 3D geological model and has an impact on volumetric calculation and reserve estimation. In this study, advanced petrophysical evaluation and high-resolution 3D modeling were adopted to empower geomodeler to generate a robust reservoir model for simulation and production forecast purposes. Using S field as a case study, the results indicated that the model successfully preserved the characterization of thin-bedded sand and shale leading towards an improved quantification of thin-bedded sand contribution and potential representatative volumetric estimation. The thin-bedded reservoir characterization and high-resolution 3D modelling technique successfully address the poor history match issues due to early water breakthrough in the prediction model but not in the actual production data. From the volume assessment, the model has successfully yielded additional 20 percent of OIIP from previous study.
-
-
-
Improving Reservoir Characterization through Application of Sequence Stratigraphy in a Carbonate Reservoir
More LessSummaryThe sequence stratigraphy study was conducted in 2020 to address reservoir thickness prediction away from the core area in a Cretaceous carbonate field in Iraq. The study updated depositional understanding and addressed the reservoir thickness and reservoir characterization to allow for more precise characterization
-
-
-
Integrated Reservoir Mapping while Drilling through High Definition Deterministic Inversion Enhances Horizontal Well Optimization.
SummaryA horizontal well trajectory targeting multi-layer reservoirs in a brownfield with uneven fluid contact movements due to a long production history was generally developed from 3D Reservoir Model. This paper aims to showcase the recent case study of an application of High-Definition Deterministic Inversion that has successfully mitigated these fluid contact and structure dip uncertainties, resulting in the successful drilling a challenging horizontal well. Reservoir properties and logs from nearby offset wells were used as input for inversion modelling, prior to drilling. Several cases of depth structure and fluid contact scenarios were incorporated into a pre-modeling workflow to capture all possible well-landing elevation outcomes. High-Definition Deterministic Inversion methods were used to optimize the trajectory during the pre-drill planning as well as in real-time, during the drilling of horizontal well. These models were used from the first penetration into the target reservoir (near to landing point), while drilling the horizontal section in the first target sand and thin shale layers, then while chasing and landing in the next target sand and finally while drilling up to the well’s total depth. The result and information from this inversion was incorporated into the 3D reservoir model to identify future potential re-development opportunities.
-