• Energy Research

Abstract Many processes and techniques have been proposed to improve the heavy oil recovery from fractured reservoirs. Such complex processes require careful design to achieve optimal efficiency with minimal costs and environmental impacts. Steam injection is one of the options for heavy-oil recovery from fractured reservoirs but significant steam requirement for effective matrix heating due to heterogeneous structure poses important challenges in terms of cost, water availability, and environment impacts due to water processing and steam generation. Al-Bahlani and Babadagli, 2008 , Al-Bahlani and Babadagli, 2009a proposed a new process called Steam-Over-Solvent in Fractured Reservoirs (SOS-FR) by adding solvent component to minimize the heat needed. The SOS-FR technique consists of a heating phase using steam injection, subsequent solvent injection, and low temperature steam injection for solvent retrieval and additional oil recovery. Optimization of this process is a critical step to determine optimal injection (and soaking) schedules as the heterogeneous structure of this kind of reservoirs may easily yield an inefficient process due to high cost and excessive use of steam and solvent. In this study, we adopted a global optimization scheme, where genetic algorithm is integrated with orthogonal arrays and response surface proxies for better convergence behavior and higher computational efficiency, to optimize the SOS-FR process for cyclic injection option. The results show that one may be able to double the profit obtained with the benchmark model using the optimal injection scheme suggested by our optimization procedure. The results are valid for models with low fracture density; further work should be performed using models with high fracture density and that the fracture orientation is not in the direction of the injector to the producer.

Abstract Many processes and techniques have been proposed to improve the heavy oil recovery from fractured reservoirs. Such complex processes require careful design to achieve optimal efficiency with minimal costs and environmental impacts. Steam injection is one of the options for heavy-oil recovery from fractured reservoirs but significant steam requirement for effective matrix heating due to heterogeneous structure poses important challenges in terms of cost, water availability, and environment impacts due to water processing and steam generation. Al-Bahlani and Babadagli, 2008 , Al-Bahlani and Babadagli, 2009a proposed a new process called Steam-Over-Solvent in Fractured Reservoirs (SOS-FR) by adding solvent component to minimize the heat needed. The SOS-FR technique consists of a heating phase using steam injection, subsequent solvent injection, and low temperature steam injection for solvent retrieval and additional oil recovery. Optimization of this process is a critical step to determine optimal injection (and soaking) schedules as the heterogeneous structure of this kind of reservoirs may easily yield an inefficient process due to high cost and excessive use of steam and solvent. In this study, we adopted a global optimization scheme, where genetic algorithm is integrated with orthogonal arrays and response surface proxies for better convergence behavior and higher computational efficiency, to optimize the SOS-FR process for cyclic injection option. The results show that one may be able to double the profit obtained with the benchmark model using the optimal injection scheme suggested by our optimization procedure. The results are valid for models with low fracture density; further work should be performed using models with high fracture density and that the fracture orientation is not in the direction of the injector to the producer.

SummaryThe discrete fracture network (DFN) model is widely used to simulate and represent the complex fractures occurring over multiple length scales. However, computational constraints often necessitate that these DFN models be upscaled into a dual-porosity dual-permeability (DPDK) model and discretized over a corner-point grid system, which is still commonly implemented in many commercial simulation packages. Many analytical upscaling techniques are applicable, provided that the fracture density is high, but this condition generally does not hold in most unconventional reservoir settings. A particular undesirable outcome is that connectivity between neighboring fracture cells could be erroneously removed if the fracture plane connecting the two cells is not aligned along the meshing direction.In this work, we propose a novel scheme to detect such misalignments and to adjust the DPDK fracture parameters locally, such that the proper fracture connectivity can be restored. A search subroutine is implemented to identify any diagonally adjacent cells of which the connectivity has been erroneously removed during the upscaling step. A correction scheme is implemented to facilitate a local adjustment to the shape factors in the vicinity of these two cells while ensuring the local fracture intensity remains unaffected. The results are assessed in terms of the stimulated reservoir volume calculations, and the sensitivity to fracture intensity is analyzed.The method is tested on a set of tight oil models constructed based on the Bakken Formation. Simulation results of the corrected, upscaled models are closer to those of DFN simulations. There is a noticeable improvement in the production after restoring the connectivity between those previously disconnected cells. The difference is most significant in cases with medium DFN density, where more fracture cells become disconnected after upscaling (this is also when most analytical upscaling techniques are no longer valid); in some 2D cases, up to a 22% difference in cumulative production is recorded. Ignoring the impacts of mesh discretization could result in an unintended reduction in the simulated fracture connectivity and a considerable underestimation of the cumulative production.

SummaryThe discrete fracture network (DFN) model is widely used to simulate and represent the complex fractures occurring over multiple length scales. However, computational constraints often necessitate that these DFN models be upscaled into a dual-porosity dual-permeability (DPDK) model and discretized over a corner-point grid system, which is still commonly implemented in many commercial simulation packages. Many analytical upscaling techniques are applicable, provided that the fracture density is high, but this condition generally does not hold in most unconventional reservoir settings. A particular undesirable outcome is that connectivity between neighboring fracture cells could be erroneously removed if the fracture plane connecting the two cells is not aligned along the meshing direction.In this work, we propose a novel scheme to detect such misalignments and to adjust the DPDK fracture parameters locally, such that the proper fracture connectivity can be restored. A search subroutine is implemented to identify any diagonally adjacent cells of which the connectivity has been erroneously removed during the upscaling step. A correction scheme is implemented to facilitate a local adjustment to the shape factors in the vicinity of these two cells while ensuring the local fracture intensity remains unaffected. The results are assessed in terms of the stimulated reservoir volume calculations, and the sensitivity to fracture intensity is analyzed.The method is tested on a set of tight oil models constructed based on the Bakken Formation. Simulation results of the corrected, upscaled models are closer to those of DFN simulations. There is a noticeable improvement in the production after restoring the connectivity between those previously disconnected cells. The difference is most significant in cases with medium DFN density, where more fracture cells become disconnected after upscaling (this is also when most analytical upscaling techniques are no longer valid); in some 2D cases, up to a 22% difference in cumulative production is recorded. Ignoring the impacts of mesh discretization could result in an unintended reduction in the simulated fracture connectivity and a considerable underestimation of the cumulative production.

Vapex (vapor extraction) is a nonthermal process that has significant potential in providing a more environmentally friendly and energy-efficient alternative to steam injection. Vaporized solvent injected in-situ dissolves into the oil and reduces oil viscosity, allowing the oil to flow to a horizontal production well via gravitational forces. While compositional simulators are available for assessing the Vapex performance, the simulation process may become difficult when taking into account the uncertainty due to reservoir heterogeneity. A semi-analytical proxy is proposed to model the process, in a way analogous to the steam-assisted gravity drainage (SAGD) model described by Butler, who demonstrated the similarity between two processes with a series of Hele-Shaw experiments and derived an analytical steady-state flow rate relationship that is comparable with the SAGD case. Solvent concentration and intrinsic diffusivity are introduced in this model instead of temperature and thermal diffusivity in SAGD. In this paper, analytical solutions and implementation details for the Vapex proxy are presented. The proposed approach is then applied to various reservoirs discretized with spatially varying rock porosity and permeability values; bitumen drainage rate and solvent penetration are calculated sequentially at grid blocks along the solvent–bitumen interface over incremental time steps. Results from this model are compared against experimental data available in the literature as well as detailed compositional simulation studies. Computational requirement of the proxy in comparison with numerical simulations is also emphasized. An important contribution from this work is that process physics are built directly into this proxy, giving it an advantage over other data-driven modeling approaches (e.g., regression). It can be used as an efficient alternative to expensive detailed flow simulations. It presents an important potential for assessing the uncertainty due to multiscale heterogeneity on effective mass transfer and the resulting recovery performance, as well as assisting decisions-making for future pilot and field development.

Vapex (vapor extraction) is a nonthermal process that has significant potential in providing a more environmentally friendly and energy-efficient alternative to steam injection. Vaporized solvent injected in-situ dissolves into the oil and reduces oil viscosity, allowing the oil to flow to a horizontal production well via gravitational forces. While compositional simulators are available for assessing the Vapex performance, the simulation process may become difficult when taking into account the uncertainty due to reservoir heterogeneity. A semi-analytical proxy is proposed to model the process, in a way analogous to the steam-assisted gravity drainage (SAGD) model described by Butler, who demonstrated the similarity between two processes with a series of Hele-Shaw experiments and derived an analytical steady-state flow rate relationship that is comparable with the SAGD case. Solvent concentration and intrinsic diffusivity are introduced in this model instead of temperature and thermal diffusivity in SAGD. In this paper, analytical solutions and implementation details for the Vapex proxy are presented. The proposed approach is then applied to various reservoirs discretized with spatially varying rock porosity and permeability values; bitumen drainage rate and solvent penetration are calculated sequentially at grid blocks along the solvent–bitumen interface over incremental time steps. Results from this model are compared against experimental data available in the literature as well as detailed compositional simulation studies. Computational requirement of the proxy in comparison with numerical simulations is also emphasized. An important contribution from this work is that process physics are built directly into this proxy, giving it an advantage over other data-driven modeling approaches (e.g., regression). It can be used as an efficient alternative to expensive detailed flow simulations. It presents an important potential for assessing the uncertainty due to multiscale heterogeneity on effective mass transfer and the resulting recovery performance, as well as assisting decisions-making for future pilot and field development.

Abstract Fractured reservoirs are highly heterogeneous and can be characterized by the probability distributions of fracture properties in a discrete fracture network model. The relationship between production performance and the fracture parameters is vastly nonlinear, rendering the process of adjusting model parameters to match both the static geological and dynamic production data challenging. This creates a need for a comprehensive history matching workflow for fractured reservoirs, which considers different local as well as global fracture parameters and leads to multiple equally-probable realizations of the discrete fracture network model parameters for uncertainty quantification. This paper presents an integrated approach for the history matching of fractured reservoirs. This new methodology includes generating multiple discrete fracture models, upscaling them for numerical multiphase flow simulation, and updating the fracture properties using dynamic flow responses such as continuous rate and pressure measurements. Available geological and tectonic information such as well-logs, seismic, and structural maps are incorporated into commercially available DFN modeling and simulation software to infer the probability distributions of relevant fracture parameters (including aperture, length, connectivity, and intensity) and to generate multiple discrete fracture network model realizations. The fracture models are further upscaled into an equivalent continuum dual-porosity model in the software using either analytical approaches or dynamic methods. The upscaled models are subjected to the flow simulation, and their production performances are compared to the true recorded responses. An automated history matching algorithm is implemented to reduce the uncertainties of the fracture properties. Components of vectors representing the principal flow directions and average fracture orientations are obtained by means of eigenvector decomposition of the permeability tensor and are optimized in the algorithm. In addition, both global fracture intensity and the local grid based intensity, which highly affect the fluid flow pattern and rate in different regions of the reservoir, are adjusted. A case study with various fracture sets is presented. The initial realizations were generated by means of Monte Carlo simulations, using the observed fractures at the well locations. Fracture intensity, orientation, and conductivity of different fracture sets were the uncertain parameters in our studies. Using the proposed methodology, parameters of different fracture sets were satisfactorily updated. Implementation of this automated history matching approach resulted in multiple equally probable discrete fracture network models and their upscaled flow simulation models that honor the geological information, and at the same time they match the dynamic production history.

Abstract Fractured reservoirs are highly heterogeneous and can be characterized by the probability distributions of fracture properties in a discrete fracture network model. The relationship between production performance and the fracture parameters is vastly nonlinear, rendering the process of adjusting model parameters to match both the static geological and dynamic production data challenging. This creates a need for a comprehensive history matching workflow for fractured reservoirs, which considers different local as well as global fracture parameters and leads to multiple equally-probable realizations of the discrete fracture network model parameters for uncertainty quantification. This paper presents an integrated approach for the history matching of fractured reservoirs. This new methodology includes generating multiple discrete fracture models, upscaling them for numerical multiphase flow simulation, and updating the fracture properties using dynamic flow responses such as continuous rate and pressure measurements. Available geological and tectonic information such as well-logs, seismic, and structural maps are incorporated into commercially available DFN modeling and simulation software to infer the probability distributions of relevant fracture parameters (including aperture, length, connectivity, and intensity) and to generate multiple discrete fracture network model realizations. The fracture models are further upscaled into an equivalent continuum dual-porosity model in the software using either analytical approaches or dynamic methods. The upscaled models are subjected to the flow simulation, and their production performances are compared to the true recorded responses. An automated history matching algorithm is implemented to reduce the uncertainties of the fracture properties. Components of vectors representing the principal flow directions and average fracture orientations are obtained by means of eigenvector decomposition of the permeability tensor and are optimized in the algorithm. In addition, both global fracture intensity and the local grid based intensity, which highly affect the fluid flow pattern and rate in different regions of the reservoir, are adjusted. A case study with various fracture sets is presented. The initial realizations were generated by means of Monte Carlo simulations, using the observed fractures at the well locations. Fracture intensity, orientation, and conductivity of different fracture sets were the uncertain parameters in our studies. Using the proposed methodology, parameters of different fracture sets were satisfactorily updated. Implementation of this automated history matching approach resulted in multiple equally probable discrete fracture network models and their upscaled flow simulation models that honor the geological information, and at the same time they match the dynamic production history.

Abstract Heavy oil and bitumen recovery cost are excessive mainly due to high energy requirement to generate heat and its environmental impacts. Steam Assisted Gravity Drainage (SAGD) is an example of this case. The determination of optimal operating conditions, such as injection rates and well locations, based on reservoir and fluid characteristics is essential in the design of field applications. Many Steam Assisted Gravity Drainage (SAGD) optimization studies published in the literature combined numerical simulation with graphical or analytical techniques for design and performance evaluation. There have been limited efforts that integrated the simulation exercise with global optimization algorithms. Some studies focused on optimization of cumulative steam-to-oil ratio (cSOR) in SAGD by altering steam injection rates, while others focused on optimization of cumulative net energy-to-oil ratio (cEOR) in solvent-additive SAGD by altering injection pressures and fraction of solvent in the injection stream. Typical scoring functions were the net present value per hectare of land (NPV/ha) by controlling steam and solvent rates. Several studies also considered total project net present value calculation by changing total project area, capital cost intensities, solvent prices, discount rate, and risk factors to determine the well spacing and drilling schedule. Optimization techniques commonly used in those studies were scattered search, simulated annealing, and genetic algorithm (GA). In continuation of these efforts, we focused on optimizing the SAGD process and its extension to solvent-additive SAGD and several optimization techniques including simulated annealing and genetic algorithm were tested and compared. Additional procedures were incorporated to improve the implementation configuration and initial population or seed. The objective function was defined to obtain the lowest cumulative steam-oil ratio (cSOR) and highest recovery factor. It was used later as a scoring function by changing solvent-to-steam ratio and steam injection rates. The results in this paper can be implemented directly in the efforts of minimization of cost and environmental impacts while accelerating the recovery in SAGD.

Abstract Heavy oil and bitumen recovery cost are excessive mainly due to high energy requirement to generate heat and its environmental impacts. Steam Assisted Gravity Drainage (SAGD) is an example of this case. The determination of optimal operating conditions, such as injection rates and well locations, based on reservoir and fluid characteristics is essential in the design of field applications. Many Steam Assisted Gravity Drainage (SAGD) optimization studies published in the literature combined numerical simulation with graphical or analytical techniques for design and performance evaluation. There have been limited efforts that integrated the simulation exercise with global optimization algorithms. Some studies focused on optimization of cumulative steam-to-oil ratio (cSOR) in SAGD by altering steam injection rates, while others focused on optimization of cumulative net energy-to-oil ratio (cEOR) in solvent-additive SAGD by altering injection pressures and fraction of solvent in the injection stream. Typical scoring functions were the net present value per hectare of land (NPV/ha) by controlling steam and solvent rates. Several studies also considered total project net present value calculation by changing total project area, capital cost intensities, solvent prices, discount rate, and risk factors to determine the well spacing and drilling schedule. Optimization techniques commonly used in those studies were scattered search, simulated annealing, and genetic algorithm (GA). In continuation of these efforts, we focused on optimizing the SAGD process and its extension to solvent-additive SAGD and several optimization techniques including simulated annealing and genetic algorithm were tested and compared. Additional procedures were incorporated to improve the implementation configuration and initial population or seed. The objective function was defined to obtain the lowest cumulative steam-oil ratio (cSOR) and highest recovery factor. It was used later as a scoring function by changing solvent-to-steam ratio and steam injection rates. The results in this paper can be implemented directly in the efforts of minimization of cost and environmental impacts while accelerating the recovery in SAGD.