We demonstrate how key geological uncertainties in a giant onshore carbonate reservoir in the Middle East, most notably fracture permeability and saturation distribution, impact the quality of the history match and change the performance forecasts of a planned Miscible Water Alternating Gas (MWAG) injection process. To achieve this, we used a history matching and multi-objective optimisation (MOO) workflow that was tightly integrated with an innovative reservoir modelling workflow that paid particular attention to the fracture and saturation modelling. Different geological models for the reservoir were designed by integrating static and dynamic data. These data indicated the need to consider fault-related fractures and to update the saturation distribution in the reservoir model. The effective medium theory was therefore used to estimate effective permeability in order to capture the presence of low-intensity fault-controlled fractures in the reservoir. The integration of Special Core Analysis (SCAL) and log-derived J-functions allowed us to build alternative saturation models that honoured well data with great accuracy. The resulting history matched models therefore accounted for the key geological uncertainties present in the reservoir. Afterwards, MOO was applied for each history matched model to identify well controls that optimally balanced the need to maximise the time on the plateau rate while adhering to the field's gas production constraints. Our results clearly show that including low-intensity fault-controlled fractures in the reservoir model improved the quality of the history match for the gas oil ratio (GOR), bottom hole pressure (BHP) and water cut. This is especially true for wells located near faults, which were difficult to match in the past. Moreover, our results further show that the updated saturation model improved the quality of the history match for the water cut, particularly for wells located in the transition zone. These different history matched models yielded different production forecasts, where the time at which the reservoir can be produced at the plateau rate varied by up to ten years. Applying MOO for each history matched model then allowed us to identify well controls for the MWAG injection that could extend the time at which the reservoir would be produced at the plateau rate for up to nine years and the risk of losing production plateau down to two years, while always adhering to the current field operational constraints. We demonstrate how the integration of MOO with an innovative workflow for fracture and saturation modelling impacts the prediction of a planned MWAG injection in a giant onshore carbonate reservoir. Our work clearly illustrates the potential of integrating MOO with new reservoir characterisation methods to improve the quantification of uncertainties in reservoir performance predictions in carbonate reservoirs.
{"title":"History Matching Under Geological Constraints Coupled with Multi-Objective Optimisation to Optimise MWAG Performance - A Case Study in a Giant Onshore Carbonate Reservoir in the Middle East","authors":"S. AlAmeri, Mohamed AlBreiki, S. Geiger","doi":"10.2118/196715-ms","DOIUrl":"https://doi.org/10.2118/196715-ms","url":null,"abstract":"\u0000 We demonstrate how key geological uncertainties in a giant onshore carbonate reservoir in the Middle East, most notably fracture permeability and saturation distribution, impact the quality of the history match and change the performance forecasts of a planned Miscible Water Alternating Gas (MWAG) injection process. To achieve this, we used a history matching and multi-objective optimisation (MOO) workflow that was tightly integrated with an innovative reservoir modelling workflow that paid particular attention to the fracture and saturation modelling.\u0000 Different geological models for the reservoir were designed by integrating static and dynamic data. These data indicated the need to consider fault-related fractures and to update the saturation distribution in the reservoir model. The effective medium theory was therefore used to estimate effective permeability in order to capture the presence of low-intensity fault-controlled fractures in the reservoir. The integration of Special Core Analysis (SCAL) and log-derived J-functions allowed us to build alternative saturation models that honoured well data with great accuracy. The resulting history matched models therefore accounted for the key geological uncertainties present in the reservoir. Afterwards, MOO was applied for each history matched model to identify well controls that optimally balanced the need to maximise the time on the plateau rate while adhering to the field's gas production constraints.\u0000 Our results clearly show that including low-intensity fault-controlled fractures in the reservoir model improved the quality of the history match for the gas oil ratio (GOR), bottom hole pressure (BHP) and water cut. This is especially true for wells located near faults, which were difficult to match in the past. Moreover, our results further show that the updated saturation model improved the quality of the history match for the water cut, particularly for wells located in the transition zone. These different history matched models yielded different production forecasts, where the time at which the reservoir can be produced at the plateau rate varied by up to ten years.\u0000 Applying MOO for each history matched model then allowed us to identify well controls for the MWAG injection that could extend the time at which the reservoir would be produced at the plateau rate for up to nine years and the risk of losing production plateau down to two years, while always adhering to the current field operational constraints.\u0000 We demonstrate how the integration of MOO with an innovative workflow for fracture and saturation modelling impacts the prediction of a planned MWAG injection in a giant onshore carbonate reservoir. Our work clearly illustrates the potential of integrating MOO with new reservoir characterisation methods to improve the quantification of uncertainties in reservoir performance predictions in carbonate reservoirs.","PeriodicalId":354509,"journal":{"name":"Day 3 Thu, September 19, 2019","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122746835","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
David Rafael Contreras Perez, R. A. Zaabi, Bernato Viratno, C. Sellar, M. I. Susanto
This paper summarizes an efficient workflow for building a reliable static model reference case by improving the accuracy of well placement in a hydrocarbon bearing structure. This is beneficial in optimising upcoming well target position and trajectory planning as well as during the dynamic history matching process. In a non-operated venture, the ability to generate an up-to-date static model that maintains pace with operations, provides valuable insight to advise the operator on the upcoming drilling plan and continuously supports the dynamic model for reserves booking, is highly sought after. The systematic approach described in this paper is applied to a geo-model from a Middle East carbonate reservoir consisting of over 50 wells with good quality PSDM seismic data. The workflow presented begins with seismic mapping, utilizing volume-based modelling techniques, followed by structural element correction using borehole images (e.g. structural formation dip and true stratigraphic thickness estimate) and finally introduces alternative control points, which enable drilled wellbore trajectories to be structurally anchored, based on layer thicknesses and structural trends within the target reservoir. Using this approach it is possible to generate a consistent structural model that honours geological markers, measured dip ranges and structural trends seen from seismic data and image logs. During the process one learns more about data quality (e.g. scale of data resolution and depth of investigation), associated with specific fields and carbonate reservoirs through the interaction between geological, geophysical and petrophysical disciplines and ensures their correct use. Data are used to improve the raw interpreted seismic horizons by calibrating mapped thickness distribution against the well tops. 2D visualizations are generated on a well-by-well basis, including map views, curtain sections (along each horizontal well), composite cross-sections and 3D visualizations to show inter-well relationships within different geological layers. As a result the well is placed in the correct structural position. Correct well placement, especially of highly deviated/horizontal wells, provides more accurate identification of reservoir sweet spots, leading to improved well target position and trajectory planning for upcoming wells, and a robust baseline to achieve production/well test history match during the dynamic modelling process.
{"title":"Maximising the Use of Horizontal Well Data in the Structural Framework of the Reservoir Modelling Workflow: A Case Study of a Middle East Carbonate Reservoir","authors":"David Rafael Contreras Perez, R. A. Zaabi, Bernato Viratno, C. Sellar, M. I. Susanto","doi":"10.2118/196638-ms","DOIUrl":"https://doi.org/10.2118/196638-ms","url":null,"abstract":"\u0000 This paper summarizes an efficient workflow for building a reliable static model reference case by improving the accuracy of well placement in a hydrocarbon bearing structure. This is beneficial in optimising upcoming well target position and trajectory planning as well as during the dynamic history matching process. In a non-operated venture, the ability to generate an up-to-date static model that maintains pace with operations, provides valuable insight to advise the operator on the upcoming drilling plan and continuously supports the dynamic model for reserves booking, is highly sought after.\u0000 The systematic approach described in this paper is applied to a geo-model from a Middle East carbonate reservoir consisting of over 50 wells with good quality PSDM seismic data. The workflow presented begins with seismic mapping, utilizing volume-based modelling techniques, followed by structural element correction using borehole images (e.g. structural formation dip and true stratigraphic thickness estimate) and finally introduces alternative control points, which enable drilled wellbore trajectories to be structurally anchored, based on layer thicknesses and structural trends within the target reservoir.\u0000 Using this approach it is possible to generate a consistent structural model that honours geological markers, measured dip ranges and structural trends seen from seismic data and image logs. During the process one learns more about data quality (e.g. scale of data resolution and depth of investigation), associated with specific fields and carbonate reservoirs through the interaction between geological, geophysical and petrophysical disciplines and ensures their correct use. Data are used to improve the raw interpreted seismic horizons by calibrating mapped thickness distribution against the well tops. 2D visualizations are generated on a well-by-well basis, including map views, curtain sections (along each horizontal well), composite cross-sections and 3D visualizations to show inter-well relationships within different geological layers. As a result the well is placed in the correct structural position. Correct well placement, especially of highly deviated/horizontal wells, provides more accurate identification of reservoir sweet spots, leading to improved well target position and trajectory planning for upcoming wells, and a robust baseline to achieve production/well test history match during the dynamic modelling process.","PeriodicalId":354509,"journal":{"name":"Day 3 Thu, September 19, 2019","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131190040","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
M. I. Susanto, C. Sellar, David Rafael Contreras Perez
This paper presents a diagnostic workflow to understand and implement rock and fluid modeling in a diagenetically heterogeneous and hydrodynamically pressured Middle East carbonate field. The workflow allows interactive field data integration, provides guidance for reservoir property distribution and fluid contact generation in order to improve reserves and forecasting estimation. The workflow is useful to a reservoir modeler in QA/QC role and in this case it proves particularly applicable in an organization with constrained resources during the farm-in process. The workflow runs on numerical methods within the static model to avoid database discrepancy during the diagnostic process. Using the core (CCAL, SCAL), log and pressure database, the geoscientist can assess subsurface modeling outputs from the simplest to more complex deterministic scenarios. The process aims to minimize the discrepancy between data input and model output while continuously honoring the data, maintaining realistic correlations (e.g. between static permeability and water saturation) and respecting inherent uncertainty. Using a data-rich Middle East carbonate reservoir, the pre- and post-diagnostic comparison of 3D modeled reservoir properties to the input data are demonstrated. Diagnostic steps have helped to understand potential subsurface scenarios and thus minimize the discrepancy post exercise. The value of the workflow is its ability to pinpoint the key uncertainties in rock and fluid modeling from the field’s vast dataset in a shorter diagnostic time. The application of the workflow in this carbonate reservoir case study increases the importance of geological and property driven rock type classification and its 3D distribution in matching the water saturation profile. This proved particularly challenging in this case study due to the field’s compartmentalization - fluid contact scenario.
{"title":"An Efficient Approach to Diagnose and Improve 3D Reservoir Model Quality in a Highly Diagenetic Deterogeneous and Dynamic Pressure Carbonate Field Case Study","authors":"M. I. Susanto, C. Sellar, David Rafael Contreras Perez","doi":"10.2118/196656-ms","DOIUrl":"https://doi.org/10.2118/196656-ms","url":null,"abstract":"\u0000 This paper presents a diagnostic workflow to understand and implement rock and fluid modeling in a diagenetically heterogeneous and hydrodynamically pressured Middle East carbonate field. The workflow allows interactive field data integration, provides guidance for reservoir property distribution and fluid contact generation in order to improve reserves and forecasting estimation. The workflow is useful to a reservoir modeler in QA/QC role and in this case it proves particularly applicable in an organization with constrained resources during the farm-in process. The workflow runs on numerical methods within the static model to avoid database discrepancy during the diagnostic process. Using the core (CCAL, SCAL), log and pressure database, the geoscientist can assess subsurface modeling outputs from the simplest to more complex deterministic scenarios. The process aims to minimize the discrepancy between data input and model output while continuously honoring the data, maintaining realistic correlations (e.g. between static permeability and water saturation) and respecting inherent uncertainty.\u0000 Using a data-rich Middle East carbonate reservoir, the pre- and post-diagnostic comparison of 3D modeled reservoir properties to the input data are demonstrated. Diagnostic steps have helped to understand potential subsurface scenarios and thus minimize the discrepancy post exercise. The value of the workflow is its ability to pinpoint the key uncertainties in rock and fluid modeling from the field’s vast dataset in a shorter diagnostic time. The application of the workflow in this carbonate reservoir case study increases the importance of geological and property driven rock type classification and its 3D distribution in matching the water saturation profile. This proved particularly challenging in this case study due to the field’s compartmentalization - fluid contact scenario.","PeriodicalId":354509,"journal":{"name":"Day 3 Thu, September 19, 2019","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116855446","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
P. Diatto, Anita Cerioli Regondi, S. Doering, D. Italiano, I. Maffeis, M. Marchesini, Marco Martin
With the aim of improving the understanding of production behaviour in a multi-discovery asset and the evaluation of near-field exploration opportunities, an integrated study has been carried out involving three different disciplines: Fluid Thermodynamics (PVT), Organic Geochemistry and Petroleum Systems Modelling (PSM). The synergistic workflow has been undertaken starting from an accurate quality check of the initial dataset related to fluid samples and lab tests. By merging PVT and geochemical data, it was possible to carry out a robust statistical survey and explore correlations across different parameters and features; in this way, strict connection among many physical parameters and some oil maturity and biodegradation indices were identified. In the following step, after geo-referencing the fluid samples in the framework of the Petroleum Systems Model and tracking the locations of the source rocks, a reliable interpretation of the oil expulsion and migration history became possible over the whole reservoir fluid system. Finally, taking into account the simulated fluid phase envelopes, further insights were drawn in terms of the fluid phase behavior in different areas, contributing to reduce uncertainty and exploration risk for future activity in nearby prospects.
{"title":"Regional Reservoir Fluid Analysis and Interpretation based on the Integration of Petroleum Systems, Organic Geochemistry and PVT","authors":"P. Diatto, Anita Cerioli Regondi, S. Doering, D. Italiano, I. Maffeis, M. Marchesini, Marco Martin","doi":"10.2118/196733-ms","DOIUrl":"https://doi.org/10.2118/196733-ms","url":null,"abstract":"\u0000 With the aim of improving the understanding of production behaviour in a multi-discovery asset and the evaluation of near-field exploration opportunities, an integrated study has been carried out involving three different disciplines: Fluid Thermodynamics (PVT), Organic Geochemistry and Petroleum Systems Modelling (PSM). The synergistic workflow has been undertaken starting from an accurate quality check of the initial dataset related to fluid samples and lab tests. By merging PVT and geochemical data, it was possible to carry out a robust statistical survey and explore correlations across different parameters and features; in this way, strict connection among many physical parameters and some oil maturity and biodegradation indices were identified. In the following step, after geo-referencing the fluid samples in the framework of the Petroleum Systems Model and tracking the locations of the source rocks, a reliable interpretation of the oil expulsion and migration history became possible over the whole reservoir fluid system. Finally, taking into account the simulated fluid phase envelopes, further insights were drawn in terms of the fluid phase behavior in different areas, contributing to reduce uncertainty and exploration risk for future activity in nearby prospects.","PeriodicalId":354509,"journal":{"name":"Day 3 Thu, September 19, 2019","volume":"50 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130305048","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Sergio Roberto Mata García, Javier Carrasco Hernández, J. López
This study examines which is the margin of usability for Artificial Intelligence (AI) algorithms related to the rock properties distribution in static modeling. This novel method shows a forward modeling approach using neural networks and genetic algorithms to optimize correlation patterns among seismic traces of stack volumes and well rock properties. Once a set of nonlinear functions is optimized in the well locations, to correlate seismic traces and rock properties, spatial response is estimated using the seismic volume. This seismic characterization process is directly dependent on the error minimization during the structural seismic interpretation process, as well as, honoring the structural complexity while modeling. Previous points are key elements to obtain an adequate correlation between well data and seismic traces. The joint mechanism of neural networks and genetic algorithms globally optimize the nonlinear functions and its parameters to minimize the cost function. Estimated objective function correlates well rock properties with seismic stack data. This mechanism is applied to real data, within a high structural complexity and several wells. As an output, calibrated petrophysical time volumes in the interval of interest are obtained. Properties are used initially to generate a geological facies model. Subsequently, facies and seismic properties are used for the three-dimensional distribution of petrophysical properties such as: rock type, porosity, clay volume and permeability. Therefore, artificial intelligence algorithms can be widely exploited for uncertainty reduction within the rock property spatial estimation.
{"title":"Seismic Characterisation using Artificial Intelligence Algorithms for Rock Property Distribution in Static Modeling","authors":"Sergio Roberto Mata García, Javier Carrasco Hernández, J. López","doi":"10.2118/196647-ms","DOIUrl":"https://doi.org/10.2118/196647-ms","url":null,"abstract":"\u0000 This study examines which is the margin of usability for Artificial Intelligence (AI) algorithms related to the rock properties distribution in static modeling. This novel method shows a forward modeling approach using neural networks and genetic algorithms to optimize correlation patterns among seismic traces of stack volumes and well rock properties. Once a set of nonlinear functions is optimized in the well locations, to correlate seismic traces and rock properties, spatial response is estimated using the seismic volume. This seismic characterization process is directly dependent on the error minimization during the structural seismic interpretation process, as well as, honoring the structural complexity while modeling. Previous points are key elements to obtain an adequate correlation between well data and seismic traces. The joint mechanism of neural networks and genetic algorithms globally optimize the nonlinear functions and its parameters to minimize the cost function. Estimated objective function correlates well rock properties with seismic stack data. This mechanism is applied to real data, within a high structural complexity and several wells. As an output, calibrated petrophysical time volumes in the interval of interest are obtained. Properties are used initially to generate a geological facies model. Subsequently, facies and seismic properties are used for the three-dimensional distribution of petrophysical properties such as: rock type, porosity, clay volume and permeability. Therefore, artificial intelligence algorithms can be widely exploited for uncertainty reduction within the rock property spatial estimation.","PeriodicalId":354509,"journal":{"name":"Day 3 Thu, September 19, 2019","volume":"71 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127003973","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
T. Taha, P. Ward, G. Peacock, R. Bordas, U. Aslam, S. Walsh, R. Hammersley, E. Gringarten
This paper presents a case study in 4D seismic history matching using an automated, ensemble-based workflow that tightly integrates the static and dynamic domains. Subsurface uncertainties, captured at every stage of the interpretative and modelling process, are used as inputs within a repeatable workflow. By adjusting these inputs, an ensemble of models is created, and their likelihoods constrained by observations within an iterative loop. The result is multiple realizations of calibrated models that are consistent with the underlying geology, the observed production data, the seismic signature of the reservoir and its fluids. It is effectively a digital twin of the reservoir with an improved predictive ability that provides a realistic assessment of uncertainty associated with production forecasts. The example used in this study is a synthetic 3D model mimicking a real North Sea field. Data assimilation is conducted using an Ensemble Smoother with multiple data assimilations (ES-MDA). This paper has a significant focus on seismic data, with the corresponding result vector generated via a petro-elastic model. 4D seismic data proves to be a key additional source of measurement data with a unique volumetric distribution creating a coherent predictive model. This allows recovery of the underlying geological features and more accurately models the uncertainty in predicted production than was possible by matching production data alone. A significant advantage of this approach is the ability to utilize simultaneously multiple types of measurement data including production, RFT, PLT and 4D seismic. Newly acquired observations can be rapidly accommodated which is often critical as the value of most interventions is reduced by delay.
{"title":"History Matching Using 4D Seismic in an Integrated Multi-Disciplinary Automated Workflow","authors":"T. Taha, P. Ward, G. Peacock, R. Bordas, U. Aslam, S. Walsh, R. Hammersley, E. Gringarten","doi":"10.2118/196680-ms","DOIUrl":"https://doi.org/10.2118/196680-ms","url":null,"abstract":"\u0000 This paper presents a case study in 4D seismic history matching using an automated, ensemble-based workflow that tightly integrates the static and dynamic domains. Subsurface uncertainties, captured at every stage of the interpretative and modelling process, are used as inputs within a repeatable workflow. By adjusting these inputs, an ensemble of models is created, and their likelihoods constrained by observations within an iterative loop. The result is multiple realizations of calibrated models that are consistent with the underlying geology, the observed production data, the seismic signature of the reservoir and its fluids. It is effectively a digital twin of the reservoir with an improved predictive ability that provides a realistic assessment of uncertainty associated with production forecasts.\u0000 The example used in this study is a synthetic 3D model mimicking a real North Sea field. Data assimilation is conducted using an Ensemble Smoother with multiple data assimilations (ES-MDA). This paper has a significant focus on seismic data, with the corresponding result vector generated via a petro-elastic model. 4D seismic data proves to be a key additional source of measurement data with a unique volumetric distribution creating a coherent predictive model. This allows recovery of the underlying geological features and more accurately models the uncertainty in predicted production than was possible by matching production data alone.\u0000 A significant advantage of this approach is the ability to utilize simultaneously multiple types of measurement data including production, RFT, PLT and 4D seismic. Newly acquired observations can be rapidly accommodated which is often critical as the value of most interventions is reduced by delay.","PeriodicalId":354509,"journal":{"name":"Day 3 Thu, September 19, 2019","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130289768","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
M. Simonov, A. Shubin, A. Penigin, D. Perets, E. Belonogov, A. Margarit
The topic of the paper is an approach to find optimal regimes of miscible gas injection into the reservoir to maximize cumulative oil production using a surrogate model. The sector simulation model of the real reservoir with a gas cap, which is in the first stage of development, was used as a basic model for surrogate model training. As the variable (control) parameters of the surrogate model parameters of gas injection into injection wells and the limitation of the gas factor of production wells were chosen. The target variable is the dynamics of oil production from the reservoir. A set of data has been created to train the surrogate model with various input parameters generated by the Latin hypercube. Several machine learning models were tested on the data set: ARMA, SARIMAX and Random Forest. The Random Forest model showed the best match with simulation results. Based on this model, the task of gas injection optimization was solved in order to achieve maximum oil production for a given period. The optimization issue was solved by Monte Carlo method. The time to find the optimum based on the Random Forest model was 100 times shorter than it took to solve this problem using a simulator. The optimal solution was tested on a commercial simulator and it was found that the results between the surrogate model and the simulator differed by less than 9%.
{"title":"Optimization of Oil Field Development using a Surrogate Model: Case of Miscible Gas Injection","authors":"M. Simonov, A. Shubin, A. Penigin, D. Perets, E. Belonogov, A. Margarit","doi":"10.2118/196639-ms","DOIUrl":"https://doi.org/10.2118/196639-ms","url":null,"abstract":"\u0000 The topic of the paper is an approach to find optimal regimes of miscible gas injection into the reservoir to maximize cumulative oil production using a surrogate model. The sector simulation model of the real reservoir with a gas cap, which is in the first stage of development, was used as a basic model for surrogate model training. As the variable (control) parameters of the surrogate model parameters of gas injection into injection wells and the limitation of the gas factor of production wells were chosen. The target variable is the dynamics of oil production from the reservoir. A set of data has been created to train the surrogate model with various input parameters generated by the Latin hypercube.\u0000 Several machine learning models were tested on the data set: ARMA, SARIMAX and Random Forest. The Random Forest model showed the best match with simulation results. Based on this model, the task of gas injection optimization was solved in order to achieve maximum oil production for a given period. The optimization issue was solved by Monte Carlo method. The time to find the optimum based on the Random Forest model was 100 times shorter than it took to solve this problem using a simulator. The optimal solution was tested on a commercial simulator and it was found that the results between the surrogate model and the simulator differed by less than 9%.","PeriodicalId":354509,"journal":{"name":"Day 3 Thu, September 19, 2019","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132244832","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Gas injection is a proven EOR method in the oil industry with many well-documented successful field applications spanning a period of more than five decades. The injected gas composition varies between projects, but is typically hydrocarbon gas, sometimes enriched with intermediate components to ensure miscibility, or carbon dioxide in regions such as the Permian Basin, where supply is available at an attractive price. Miscible nitrogen injection into oil reservoirs, on the other hand, is a relatively uncommon EOR technique because nitrogen often requires a prohibitively high pressure to reach miscibility. Unlike other injection gases, the minimum miscibility pressure for nitrogen decreases with increasing temperature. In fact, in deep, hot reservoirs containing volatile oil, nitrogen may develop miscibility at a pressure similar to the MMP for hydrocarbon gas or carbon dioxide. The phase behavior is more complicated than what can be captured by correlations and hence requires equation-of-state calculations. Results from a recent EOR screening study in ADNOC indicate that a couple of high-temperature oil reservoirs in Abu Dhabi may be potential targets for miscible nitrogen injection. This paper discusses key aspects of the EOS modeling. Advanced gas injection PVT data are available to enable a fair comparison between nitrogen, carbon dioxide and lean hydrocarbon gas. In this work, we have modelled and analyzed the phase behavior of two volatile oil systems with respect to nitrogen, hydrocarbon gas, and carbon dioxide injection, as part of a reservoir simulation study, which will be covered in a subsequent publication; see Mogensen and Xu (2019). Nitrogen behaves differently from hydrogen carbon gas, despite the fact that the two gases lead to similar minimum miscibility pressures. At the prevailing reservoir pressure, the swelling factor with hydrocarbon gas is four times higher than for nitrogen. Furthermore, the reservoir fluid density increases during swelling with nitrogen, whereas it decreases as a result of hydrocarbon gas swelling. The same trend is observed for viscosity. Injection gas blends with various proportions of nitrogen and carbon injection shows that the MMP is constant when more than 35-40% nitrogen is present in the blend.
在石油行业，注气是一种经过验证的提高采收率方法，在过去的50多年里，有许多成功的现场应用记录。不同项目注入的天然气成分不同，但通常是碳氢化合物气，有时富含中间成分以确保混相性，或者在二叠纪盆地等地区以具有吸引力的价格供应二氧化碳。另一方面，向油藏注入混相氮气是一种相对不常见的提高采收率技术，因为氮气通常需要过高的压力才能达到混相。与其他注入气体不同，氮气的最小混相压力随着温度的升高而降低。事实上，在含有挥发油的深层热储层中，氮气可能在与烃类气体或二氧化碳的MMP相似的压力下形成混相。相位行为比关联所能捕捉到的更为复杂，因此需要状态方程计算。ADNOC最近的一项EOR筛选研究结果表明，阿布扎比的几个高温油藏可能是注混相氮气的潜在目标。本文讨论了EOS建模的关键方面。先进的注气PVT数据可以对氮气、二氧化碳和贫烃气体进行公平的比较。在这项工作中，我们模拟并分析了两种挥发油系统在氮气、碳氢气体和二氧化碳注入方面的相行为，作为油藏模拟研究的一部分，这将在随后的出版物中介绍;参见Mogensen and Xu(2019)。氮气的行为与碳氢气体不同，尽管这两种气体的最小混相压力相似。在现行储层压力下，含烃气的膨胀系数是含氮气的4倍。此外，在氮气膨胀过程中，储层流体密度增加，而在烃气膨胀过程中，储层流体密度降低。粘度也有同样的趋势。不同比例的氮气和碳的注气混合物表明，当混合物中氮气含量超过35-40%时，MMP是恒定的。
{"title":"Potential Applicability of Miscible N2 Flooding in High-Temperature Abu Dhabi Reservoir","authors":"K. Mogensen, Siqing Xu","doi":"10.2118/196716-ms","DOIUrl":"https://doi.org/10.2118/196716-ms","url":null,"abstract":"\u0000 Gas injection is a proven EOR method in the oil industry with many well-documented successful field applications spanning a period of more than five decades. The injected gas composition varies between projects, but is typically hydrocarbon gas, sometimes enriched with intermediate components to ensure miscibility, or carbon dioxide in regions such as the Permian Basin, where supply is available at an attractive price.\u0000 Miscible nitrogen injection into oil reservoirs, on the other hand, is a relatively uncommon EOR technique because nitrogen often requires a prohibitively high pressure to reach miscibility. Unlike other injection gases, the minimum miscibility pressure for nitrogen decreases with increasing temperature. In fact, in deep, hot reservoirs containing volatile oil, nitrogen may develop miscibility at a pressure similar to the MMP for hydrocarbon gas or carbon dioxide. The phase behavior is more complicated than what can be captured by correlations and hence requires equation-of-state calculations.\u0000 Results from a recent EOR screening study in ADNOC indicate that a couple of high-temperature oil reservoirs in Abu Dhabi may be potential targets for miscible nitrogen injection. This paper discusses key aspects of the EOS modeling. Advanced gas injection PVT data are available to enable a fair comparison between nitrogen, carbon dioxide and lean hydrocarbon gas. In this work, we have modelled and analyzed the phase behavior of two volatile oil systems with respect to nitrogen, hydrocarbon gas, and carbon dioxide injection, as part of a reservoir simulation study, which will be covered in a subsequent publication; see Mogensen and Xu (2019). Nitrogen behaves differently from hydrogen carbon gas, despite the fact that the two gases lead to similar minimum miscibility pressures. At the prevailing reservoir pressure, the swelling factor with hydrocarbon gas is four times higher than for nitrogen. Furthermore, the reservoir fluid density increases during swelling with nitrogen, whereas it decreases as a result of hydrocarbon gas swelling. The same trend is observed for viscosity. Injection gas blends with various proportions of nitrogen and carbon injection shows that the MMP is constant when more than 35-40% nitrogen is present in the blend.","PeriodicalId":354509,"journal":{"name":"Day 3 Thu, September 19, 2019","volume":" 4","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"113951775","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Improved reservoir knowledge is key to extracting additional value from existing oil and gas assets. However, given the uncertainty in the subsurface, it is always a question if our current development strategy is the most robust choice, or if there are alternatives that can further increase the value of our field. This paper presents a novel solution that enables the asset team to answer these questions in a new way. Furthermore, the solution helps teams quickly identify and screen new opportunities that ultimately increase both subsurface understanding and the value of the field. The solution combines a quasi- Newton gradient based numerical optimization scheme with a stochastic simplex approximate gradient (StoSAG) algorithm. Because the algorithm is non-intrusive with respect to the fluid flow simulator, we can directly apply the solution on any flow optimization problem without the need to access the simulator source code. The solution is implemented using a microservice architecture that allows for efficient scaling and deployment either on cloud-based or internal systems. We demonstrate the proposed solution on a field containing 11 oil producers and 7 water injectors by optimizing the water injection and oil production rates. The machine learning algorithm allows us to quickly explore different drainage strategies, given the current understanding and associated uncertainties of the reservoir. Specifically, the software solution suggests that 6 of the 18 pre-defined well targets are high risk and/or of little value. Running a second development scenario where we do not drill these six wells reduces the investment cost of this field by 163 MUSD and increases the expected net present value per well of the field by 48 percent. Compared with the reactive control drainage strategy approach, we increase the expected net present value of the field by 9.0 %, while simultaneously lowering the associated risk.
{"title":"Drainage Strategy Optimization - Making Better Decisions Under Uncertainty","authors":"J. Sætrom, K. Wojnar, M. Stunell","doi":"10.2118/196683-ms","DOIUrl":"https://doi.org/10.2118/196683-ms","url":null,"abstract":"\u0000 Improved reservoir knowledge is key to extracting additional value from existing oil and gas assets. However, given the uncertainty in the subsurface, it is always a question if our current development strategy is the most robust choice, or if there are alternatives that can further increase the value of our field. This paper presents a novel solution that enables the asset team to answer these questions in a new way. Furthermore, the solution helps teams quickly identify and screen new opportunities that ultimately increase both subsurface understanding and the value of the field. The solution combines a quasi- Newton gradient based numerical optimization scheme with a stochastic simplex approximate gradient (StoSAG) algorithm. Because the algorithm is non-intrusive with respect to the fluid flow simulator, we can directly apply the solution on any flow optimization problem without the need to access the simulator source code. The solution is implemented using a microservice architecture that allows for efficient scaling and deployment either on cloud-based or internal systems. We demonstrate the proposed solution on a field containing 11 oil producers and 7 water injectors by optimizing the water injection and oil production rates. The machine learning algorithm allows us to quickly explore different drainage strategies, given the current understanding and associated uncertainties of the reservoir.\u0000 Specifically, the software solution suggests that 6 of the 18 pre-defined well targets are high risk and/or of little value. Running a second development scenario where we do not drill these six wells reduces the investment cost of this field by 163 MUSD and increases the expected net present value per well of the field by 48 percent. Compared with the reactive control drainage strategy approach, we increase the expected net present value of the field by 9.0 %, while simultaneously lowering the associated risk.","PeriodicalId":354509,"journal":{"name":"Day 3 Thu, September 19, 2019","volume":"62 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128296250","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Data from seismic to production is integrated to build models to provide estimations of parameters such as petroleum volumetrics, pressure behavior, and production performance (Fig. 1). The level of confidence of these models depends on the representativeness of the data. The quality of the generated models is based on the data interpreted and integrated aimed to build computational realizations of petroleum reservoirs. Reservoir dynamic simulation is the most applied process that integrates all reservoir data, where an Equation of State (EOS) is coupled with the objective to estimate the fluid thermodynamic state at each computational step. The simulation consists of iterative mathematical computations in which the reservoir-defined conditions at the previous time step is an input to determine the properties at the next and subsequent time steps. The calculated pressure is a fundamental variable in each time step, which means that a representative and high level of confidence Pressure Volume Temperature (PVT) model is required to avoid scale-up of errors resulting from fluid pressure estimation. A PVT modeling includes three main stages: Fluid sample and data acquisitionLaboratory analysis and fluid characterizationThe EOS model. The emphasis in this work is on the EOS model, which is the fluid model used for the simulation process. The objective of this work is to analyze the main uncertainties associated with typical EOS modeling and defining the level of confidence of these EOS approaches. In this work, some of the most-used approaches for EOS modeling are reviewed. An assessment of these methods is also provided based on their application to actual petroleum fluids with the objective of defining their statistical level of confidence. First, the study analyzes the sources of critical uncertainties in a PVT EOS model. Second, a statistical number of PVT laboratory studies of petroleum fluids is used to determine the level of confidence of four approaches that are based on the two well-known Peng-Robinson and Soave-Redlich-Kwong EOS. Third, statistical analysis is performed to determine the level of confidence of the different methods. Fourth, a correlation to determine the optimal number of pseudo-components is defined. These steps include: Characterization of fluid and heavy componentsTuningLumping. As a result of this study, one can conclude: The level of confidence of the four analyzed approachesThe significance of the difference between the analyzed methodsA correlation to determine the optimal number of pseudo-components. In this work, a statistical analysis over some of the most-used EOS modeling approaches and on a set of petroleum fluid PVTs was performed to determine the level of confidence of four EOS modeling methods. In addition, a correlation was introduced for a priori determination of the optimal number of pseudo-components in a PVT fluid.
{"title":"EOS Workflows Uncertainties and Implications in Reservoir Modeling","authors":"Angulo Yznaga, Reinaldo Jose","doi":"10.2118/196629-ms","DOIUrl":"https://doi.org/10.2118/196629-ms","url":null,"abstract":"\u0000 Data from seismic to production is integrated to build models to provide estimations of parameters such as petroleum volumetrics, pressure behavior, and production performance (Fig. 1). The level of confidence of these models depends on the representativeness of the data. The quality of the generated models is based on the data interpreted and integrated aimed to build computational realizations of petroleum reservoirs.\u0000 Reservoir dynamic simulation is the most applied process that integrates all reservoir data, where an Equation of State (EOS) is coupled with the objective to estimate the fluid thermodynamic state at each computational step. The simulation consists of iterative mathematical computations in which the reservoir-defined conditions at the previous time step is an input to determine the properties at the next and subsequent time steps. The calculated pressure is a fundamental variable in each time step, which means that a representative and high level of confidence Pressure Volume Temperature (PVT) model is required to avoid scale-up of errors resulting from fluid pressure estimation.\u0000 A PVT modeling includes three main stages: Fluid sample and data acquisitionLaboratory analysis and fluid characterizationThe EOS model.\u0000 The emphasis in this work is on the EOS model, which is the fluid model used for the simulation process. The objective of this work is to analyze the main uncertainties associated with typical EOS modeling and defining the level of confidence of these EOS approaches. In this work, some of the most-used approaches for EOS modeling are reviewed. An assessment of these methods is also provided based on their application to actual petroleum fluids with the objective of defining their statistical level of confidence.\u0000 First, the study analyzes the sources of critical uncertainties in a PVT EOS model. Second, a statistical number of PVT laboratory studies of petroleum fluids is used to determine the level of confidence of four approaches that are based on the two well-known Peng-Robinson and Soave-Redlich-Kwong EOS. Third, statistical analysis is performed to determine the level of confidence of the different methods. Fourth, a correlation to determine the optimal number of pseudo-components is defined. These steps include: Characterization of fluid and heavy componentsTuningLumping.\u0000 As a result of this study, one can conclude: The level of confidence of the four analyzed approachesThe significance of the difference between the analyzed methodsA correlation to determine the optimal number of pseudo-components.\u0000 In this work, a statistical analysis over some of the most-used EOS modeling approaches and on a set of petroleum fluid PVTs was performed to determine the level of confidence of four EOS modeling methods. In addition, a correlation was introduced for a priori determination of the optimal number of pseudo-components in a PVT fluid.","PeriodicalId":354509,"journal":{"name":"Day 3 Thu, September 19, 2019","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133662652","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：