Search

Reservoir pressure depletion is a critical aspect leading to declined oil production. Water flooding becomes mandatory to increase the recovery efficiency from certain reservoirs. Establishing well placements, injection rates, and injection schedules enables improved fluid displacement in such reservoirs. Standard reservoir simulators are widely used to provide near-optimal scenarios for maximizing oil recovery. Such reservoir simulations involved numerous iterations and required significant computational effort. Also, the conventional machine learning methods provide a relatively high uncertainty to predict the considered problem. In this context, a deep-learning surrogate model comprising long short-term memory networks and densely connected networks is proposed to predict cumulative oil production using various waterflood injection well patterns, injection rates, and injection period information. The relevance test clarifies that the oil recovery mainly controls by water injection duration and injection pattern. A total of 1485 data records are considered to validate the model's performance, demonstrating less than 3.5% absolute average relative error and a prediction versus actual value regression coefficient of 0.99989. In addition, the prediction accuracy of the proposed deep-learning model is also shown to outperform two well-known traditional machine-learning models (i.e. artificial neural networks and support vector regressions). [ABSTRACT FROM AUTHOR]

The numerical simulation of multiphase and multicomponent flows in oil reservoirs is a significant challenge, demanding robust and computationally efficient numerical formulations. Particularly, scenarios with high mobility ratios between injected and resident fluids can lead to Grid Orientation Effects (GOE), where numerical solutions strongly depend on the alignment between flow and computational grid and mobility ratio. This phenomenon relates to an anisotropic distribution in truncation error tied to the numerical approximation of the transport term. Although the oil industry commonly uses linear Two Point Flux Approximation (TPFA) for diffusive fluxes and the First Order Upwind (FOU) method for advective fluxes, both lack rotational invariance and TPFA struggles with non-k-orthogonal grids. This paper proposes a comprehensive cell-centered finite-volume formulation to simulate reservoir oil-water displacements, integrating the classical IMPES (Implicit Pressure Explicit Saturation) segregate approach with unstructured, non-k-orthogonal meshes. Diffusive flux discretization employs a Multipoint Flux Approximation with Harmonic Points (MPFA-H), capable of handling heterogeneous and strongly anisotropic media. A modified second-order Flux Corrected Transport (FCT) approach curbs artificial numerical diffusion for transport term discretization. Additionally, we incorporate a Flow-Oriented Scheme (FOS) for computing low-order and high-order approximations of the numerical fluxes to enhance multidimensional approximation and reduce GOE. The proposed strategy is validated through benchmark problems, yielding precise outcomes with reduced numerical diffusion and GOE effects, underscoring its efficiency for complex reservoir flow simulations. [ABSTRACT FROM AUTHOR]

The objective of this study is to propose a computationally inexpensive and effective approach that addresses the challenges faced with computationally expensive and time-consuming trial-and-error and direct optimization methods in well-control optimization. This approach involves combining proxy models such as artificial neural network (ANN) models with optimization algorithms to determine an optimal solution much faster. It was implemented in a heterogeneous oil reservoir undergoing waterflooding. The controllable parameters of the reservoir simulation model were identified as bottom-hole pressure for the producers and water injection rate for the injectors. Minimum and maximum values of each input parameter were defined based on reservoir conditions and used with a Box Behnken design (BBD) method to generate realizations for conducting reservoir simulations to obtain cumulative oil produced (COP) and cumulative water produced (CWP). The input and output data were normalized before being used for model development such that 70:15:15% of data was used for training, validation, testing, and all of the ANN model in which a coefficient of correlation (R) of 0.99756, 0.94354, 0.95813, and 0.98589 were obtained respectively. This indicates the accuracy, validity, and reliability of the model. The coefficient of determination (R2) for training, validation, testing, and all datasets as well as statistical error and trend analysis were used to validate the model. R2 values for each case were not less than 0.80, and the responses were reproduced by the ANN model with average relative error and root mean square error of not more than 0.7%. Weights and biases were extracted from the trained and validated ANN model to aid in outputting a visible ANN model that can be used for optimization studies. A multi-objective genetic algorithm was used to determine an optimal solution that maximized COP and minimized CWP. Average and optimized input data were used to run the developed ANN model. Results revealed that the optimized case outperformed the case for which average input values were used evidenced by the production of 34.198 MSm3 more oil and 14.297 MMSm3 less water. The findings of this study showed that using an ANN-MOGA approach will eliminate the computationally expensive, time-consuming, and inefficient trial-and-error approach for well-control optimization. Oil recovery was improved while water production was reduced resulting in low expenditure on treatment and disposal of produced water. [ABSTRACT FROM AUTHOR]

Consistent discretization methods are a natural fit for the novel cut-cell gridding technique for reservoir simulation, which preserves the orthogonality characteristic in the lateral direction. Both uniform (global) and novel hybrid (local) variants of consistent discretization methods are implemented and tested in the vicinity of fault representations on cut-cell grids. Novel consistent discretization methods, which do not require major intrusive changes to the solver structure of industrial-grade reservoir simulators, are investigated in this work. Cell-centered methods such as multi-point flux approximation (MPFA), average multi-point flux approximation (AvgMPFA), and nonlinear two-point flux approximation (NTPFA) methods fit naturally into the framework of existing industrial-grade simulators. Thus, cut-cell compatible variants of AvgMPFA and NTPFA and their novel hybridizations with TPFA are implemented and tested. An implementation of the relatively more computationally expensive MPFA is also made to serve as accuracy reference to AvgMPFA and NTPFA. AvgMPFA and NTPFA multiphase simulation results are compared in terms of accuracy and computational performance against the ones computed with reference MPFA and TPFA methods on a set of synthetic cut-cell grid models of varying complexity including conceptual models and a field-scale model. It is observed that AvgMPFA consistently yields more accurate and computationally efficient simulations than NTPFA on cut-cell grids. Moreover, AvgMPFA-TPFA hybrids run faster than NTPFA-TPFA hybrids when compared on the same problem for the same hybridization strategy. On the other hand, the computational performance of AvgMPFA degrades more rapidly compared to NTPFA with increasing "rings" of orthogonal blocks around cut-cells. Auspiciously, only one or two "rings" of orthogonal blocks around cut cells are sufficient for AvgMPFA to deliver high accuracy. [ABSTRACT FROM AUTHOR]

This paper presents an in-depth investigation of the biogeochemical modeling approaches applied to underground hydrogen storage. It delves into the intricate dynamics of hydrogen in the subsurface, focusing on small (pore-lab scale) and reservoir-scale models, highlighting the importance of capturing microbial, geochemical, and fluid flow dynamic interactions in porous media to simulate storage performance accurately. Small-scale models offer detailed insights into localized phenomena, such as microbial hydrogen consumption and mineral reactions, and can be verified and calibrated against laboratory data. Conversely, large-scale models are essential to assess the feasibility of a project and forecast the storage performance, but cannot be proven by real data yet. This work addresses the challenge of transitioning from fine-scale to reservoir models, integrating spatial heterogeneity and long-term dynamics while retaining biogeochemical complexity. Through the use of several simulation tools, like PHREEQC, Comsol, DuMuX, Eclipse, CMG-GEM, and others, this study explores how modeling approaches are evolving to incorporate multiphysics processes and biochemical feedback loops, which are essential for predicting hydrogen retention, flow, and potential risks. The findings highlight the strengths and limitations of current modeling techniques and suggest a workflow for exploiting at best existing modeling capabilities and developing reservoir models to support hydrogen storage appraisal and management. [ABSTRACT FROM AUTHOR]

Infill drilling is one of the most effective methods of improving the performance of polymer flooding. The difficulties related to infill drilling are determining the optimal numbers and placements of infill wells. In this study, an improved Archimedes optimization algorithm with a Halton sequence (HS-AOA) was proposed to overcome the aforementioned difficulties. First, to optimize infill well placement for polymer flooding, an objective function that considers the economic influence of infill drilling was developed. The novel optimization algorithm (HS-AOA) for infill well placement was subsequently developed by combining the AOA with the Halton sequence. The codes were developed in MATLAB 2023a and connected to a commercial reservoir simulator, Computer Modeling Group (CMG) STARS, Calgary, AB, Canada to carry out infill well placement optimization. Finally, the HS-AOA was compared to the basic AOA to confirm its reliability and then used to optimize the infill well placements for polymer flooding in a typical offshore oil reservoir. The results showed that the introduction of the Halton sequence into the AOA effectively increased the diversity of the initial objects in the AOA and prevented the HS-AOA from becoming trapped in the local optimal solutions. The HS-AOA outperformed the AOA. This approach was effective for optimizing the infill well placement for polymer flooding processes. In addition, infill drilling could effectively and economically improve the polymer flooding performance in offshore oil reservoirs. The net present value (NPV) of the polymer flooding case with infill wells determined by HS-AOA reached USD 3.5 × 108, which was an increase of 7% over that of the polymer flooding case. This study presents an effective method for optimizing infill well placement for polymer flooding processes. It can also serve as a valuable reference for other optimization problems in the petroleum industry, such as joint optimization of well control and placement. [ABSTRACT FROM AUTHOR]

The Bozi 3 reservoir is an ultra-deep condensate reservoir (−7800 m) with a high temperature (138.24 °C) and high pressure (104.78 MPa), leading to complex phase behaviors. Few PVT studies could be referred in the literature to meet such high temperature and pressure conditions. Furthermore, it is questionable regarding the applicability of existing condensate production techniques to such a high temperature and pressure reservoir. This study first characterized the phase behavior via PVT experiments and EOS tuning. The operating conditions were then optimized through reservoir numerical simulation. Results showed that: (1) the critical condensate temperature and pressure of Bozi 3 condensate gas were 326.24 °C and 43.83 MPa, respectively; (2) four gases (methane, recycled dry gas, carbon dioxide, and nitrogen) were analyzed, and methane was identified as the optimal injection gas; (3) gas injection started when the production began to fall and achieved higher recovery than gas injection started when the pressure fell below the dew-point pressure; (4) simultaneous injection of methane at both the upper and lower parts of the reservoir can effectively produce condensate oil over the entire block. This scheme achieved 8690.43 m3 more oil production and 2.75% higher recovery factor in comparison with depletion production. [ABSTRACT FROM AUTHOR]

Optimizing well placement is one of the primary challenges in oil field development. The number and positions of wells must be carefully considered, as it is directly related to the infrastructure cost and the profits over the field's life cycle. In this paper, we propose three estimation of distribution algorithms to optimize well placement with the objective of maximizing the net present value. The methods are guided by an elite set of solutions and are able to obtain multiple local optima in a single run. We also present an auxiliary regression model to preemptively discard candidate solutions with poor performance prediction, thus avoiding running computationally expensive simulations for unpromising candidates. The model is trained with the data obtained during the search process and does not require previous training. Our algorithms yielded a significant improvement compared to a state-of-the-art reference method from the literature, as evidenced by computational experiments with two benchmarks. [ABSTRACT FROM AUTHOR]

A physics-informed convolutional neural network (PICNN) is proposed to simulate two-phase flow in porous media with time-varying well controls. While most PICNNs in the existing literature worked on parameter-to-state mapping, our proposed network parameterizes the solutions with time-varying controls to establish a control-to-state regression. Firstly, a finite volume scheme is adopted to discretize flow equations and formulate a loss function that respects mass conservation laws. Neumann boundary conditions are seamlessly incorporated into the semi-discretized equations so no additional loss term is needed. The network architecture comprises two parallel U-Net structures, with network inputs being well controls and outputs being the system states (e.g., oil pressure and water saturation). To capture the time-dependent relationship between inputs and outputs, the network is well designed to mimic discretized state-space equations. We train the network progressively for every time step, enabling it to simultaneously predict oil pressure and water saturation at each timestep. After training the network for one timestep, we leverage transfer learning techniques to expedite the training process for a subsequent time step. The proposed model is used to simulate oil–water porous flow scenarios with varying reservoir model dimensionality, and aspects including computation efficiency and accuracy are compared against corresponding numerical approaches. The comparison with numerical methods demonstrates that a PICNN is highly efficient yet preserves decent accuracy. [ABSTRACT FROM AUTHOR]

High-pressure air injection (HPAI) is an enhanced oil recovery process in which compressed air is injected into deep, light oil reservoirs, with the expectation that the oxygen in the injected air will react with a fraction of the reservoir oil at an elevated temperature to produce carbon dioxide. The different chemical reactions taking place can be grouped into oxygen addition, thermal cracking, oxygen-induced cracking, and bond scission reactions. The latter reactions involve the combustion of a flammable vapor as well as the combustion of solid fuel, commonly known as "coke". Since stable peak temperatures observed during HPAI experiments are typically below 300 °C, it has been suggested that thermal cracking and combustion of solid fuel may not be important reaction mechanisms for the process. The objective of this work is to assess the validity of that hypothesis. Therefore, this study makes use of different oxidation and combustion HPAI experiments, which were performed on two different light oil reservoir samples. Modeling of those tests indicate that thermal cracking is not an important reaction mechanism during HPAI and can potentially be ignored. The work also suggests that the main fuel consumed by the process is a flammable vapor generated by the chemical reactions. This represents a shift from the original in situ combustion paradigm, which is based on the combustion of coke. [ABSTRACT FROM AUTHOR]

In this paper, an innovative approach for enhancing fluid transport modeling in porous media is presented, which finds application in various fields, including subsurface reservoir modeling. Fluid flow models are typically solved numerically by addressing a system of partial differential equations (PDEs) using methods such as finite difference and finite volume. However, these processes can be computationally demanding, particularly when aiming for high precision on a fine scale. Researchers have increasingly turned to machine learning to explore solutions for PDEs in order to improve simulation efficiency. The proposed method combines an adaptive multi-scale strategy with generative adversarial networks (GAN) to increase simulation efficiency on a fine scale. The devised model, called simulation enhancement GAN (SE-GAN), takes coarse-scale simulation results as input and generates fine-scale results in conjunction with the provided petrophysical properties. With this new approach, a deep learning model is trained to map coarse-scale results to fine-scale outcomes, rather than directly solving the fluid flow model. Case studies reveal that SE-GAN can achieve a significant improvement in accuracy while reducing computational time compared to the original fine-scale simulation solver. A comprehensive evaluation of numerical experiments is conducted to elucidate the benefits and limitations of this method. The potential of SE-GAN in accelerating the numerical solver for reservoir simulations is also demonstrated. [ABSTRACT FROM AUTHOR]

The Gas-Oil Ratio (GOR) is a crucial production parameter in oil reservoirs. An increase in GOR results in higher gas production and lower oil production, potentially leading to well shut-ins due to economic infeasibility. This study focuses on a real fractured oil field that requires urgent production operations to reduce the producing GOR. In this study, the static model for the field was developed using commercial software, involving steps such as data collection, fault modeling, meshing, and statistical analysis to prepare for dynamic simulation. The dynamic model incorporates geometry, gridding, and rock properties from the static model, utilizing a dual-porosity approach for the naturally fractured reservoir and the Peng-Robinson equation for fluid phase behavior. Initial reservoir conditions, production history, and rock-fluid interactions were defined, with relative permeability curves indicating a water-wet reservoir and low critical gas saturation affecting the GOR. To better understand the relationship between reservoir and production parameters, a detailed sensitivity analysis was performed using the Response Surface Methodology (RSM). Following the sensitivity analysis, a history matching process was conducted using the Designed Exploration and Controlled Evolution (DECE) optimizer to validate the model for future forecasts. Six operational scenarios were defined to decrease the production GOR and enhance final recovery from the field. The results indicate that the water injection scenario is effective in preventing the GOR increase by maintaining reservoir pressure, thereby sustaining production over a longer period. This scenario also improves oil recovery by approximately 6% compared to the base case. Finally, optimization was carried out using the DECE optimizer for each scenario to fine-tune the operational parameters. The goal was to maximize oil revenue for each scenario during the optimization process. This study stands out as one of the few that provides a comprehensive analysis of production behavior and development planning for a real fractured reservoir with high producing GOR. [ABSTRACT FROM AUTHOR]

Low salinity Waterflooding (LSWF) is a proven EOR technique which has emerged in the last few decades. In this, the salinity of the injected water is reduced. Studies have unveiled that the LSWF can improve about 5–38% of oil recovery. This work aims to study the effectiveness of LSWF by modeling Sea waterflooding (SWF) and LSWF using a commercial reservoir simulator and justifying the proposed mechanisms responsible for the incremental oil recovery due to LSWF. A critical parametric study has been undertaken, which led us to conclude that LSWF is a cost-effective, and environment-friendly method that results in additional oil recovery for sandstone reservoirs. [ABSTRACT FROM AUTHOR]

From June 2008 to August 2013, approximately 67 kt of CO2 was injected into a deep saline formation at the Ketzin pilot CO2 storage site. During injection, 3D seismic surveys have been performed to monitor the migration of sequestered CO2. Seismic monitoring results are limited by the acquisition and signal-to-noise ratio of the acquired data. The multiphysical reservoir simulation provides information regarding the CO2 fluid behavior, and the approximated model should be calibrated with the monitoring results. In this work, property models are delivered from the multiphysical model during 3D repeated seismic surveys. The simulated seismic data based on the models are compared with the real data, and the results validate the effectiveness of the multiphysical inversion method. Time-lapse analysis shows the trend of CO2 migration during and after injection. [ABSTRACT FROM AUTHOR]

East Natuna is well known for its huge natural gas reserves with a very high CO2 content. The appearance of CO2 content in an oil and gas field is always considered as waste material, and will severely affect the economic value of the field. The higher the content, the more costly the process, both technically and environmentally. In this research, the newly proposed reservoir management approach called CSSU (Carbon Sequestration, Storage, and Utilization) method is trying to be applied to change the paradigm of CO2 from waste material into economic material. The CSSU method is an integration of geological, geophysical, reservoir engineering, and engineering economics with the determination of technical and economic optimization of the use of CO2 produced as the working fluid in a power generation system that has been conditioned through an injection-production system in geological formations. Reservoir simulation modeling is done by three models, namely: Compositional, Compositional + Geomechanical Coupling, and Compositional + Geomechanical Coupling + Thermal. There is a difference in the the total injection between Compositional + Geomechanical Coupling and ordinary Compositional simulations of 1-2 % due to factors such as Modulus Young, Poisson's Ratio, Angle of Internal Friction, and Biot's Coefficient which affect the reservoir pore volume calculations and the total CO2 fluid injection calculation. The changes in geomechanical parameters will affect the CSSU techno-economic analysis where a 30 % change in the rock compressibility and poisson ratio parameters will effect changes in the electrical energy amounts being produced by 0.01 MW or 0.33 %, and in an economic value of 4 MMUS $ or 2.24 %. [ABSTRACT FROM AUTHOR]

We study linear models for the prediction of the initial guess for the nonlinear Newton-Raphson solver. These models use one or more of the previous simulation steps for prediction, and their parameters are estimated by the ordinary least-squares method. A key feature of the approach is that the parameter estimation is performed using data obtained directly during the simulation and the models are updated in real time. Thus we avoid the expensive process of dataset generation and the need for pre-trained models. We validate the workflow on a standard benchmark Egg dataset of two-phase flow in porous media and compare it to standard approaches for the estimation of initial guess. We demonstrate that the proposed approach leads to reduction in the number of iterations in the Newton-Raphson algorithm and speeds up simulation time. In particular, for the Egg dataset, we obtained a 30% reduction in the number of nonlinear iterations and a 20% reduction in the simulation time. [ABSTRACT FROM AUTHOR]

Reservoir simulation is crucial for understanding the flow response in underground reservoirs, and it significantly helps reduce uncertainties in geological characterization and optimize methodologies for field development strategies. However, providing efficient and accurate solutions for the strong heterogeneity remains challenging, as most of the discretization methods cannot handle this complexity. In this work, we perform a comprehensive assessment of various numerical linearization techniques employed in reservoir simulation, particularly focusing on the performance of the nonlinear solver for problem dealing with fluid flow in porous media. The primary linearization methods examined are finite difference central (FDC), finite forward difference (FDF), and operator-based linearization (OBL). These methods are rigorously analyzed and compared in terms of their accuracy, computational efficiency, and adaptability to changing reservoir conditions. The results demonstrate that each method has distinct strengths and limitations. The FDC method is more accurate particularly in complex simulations where strong heterogeneity are introduced but is generally slower in convergence. The OBL on the other hand, is more efficient and converges quickly, which makes it suitable for scenarios with limited computational resources and simple physics, while the FDF method provides a balanced combination of precision and computational speed, contingent upon careful step size management of the derivative estimations. This paper aims to guide the selection of appropriate linearization techniques for enhancing nonlinear solvers' accuracy and efficiency in reservoir simulation. [ABSTRACT FROM AUTHOR]

Carbon dioxide capture and storage (CCS) is a necessary requirement for high-emitting CO2 industries to significantly reduce volumes of greenhouse gases released into the atmosphere and mitigate climate change. Geological CO2 storage into depleted oil and gas fields is the fastest and most accessible option for CCS deployment allowing for re-purposing existing infrastructures and utilizing significant knowledge about the subsurface acquired during field production operations. The location of such depleted fields in the neighborhoods of high-emitting CO2 industries is an additional advantage of matured on-shore European fields. Considering these advantages, oil and gas operators are now evaluating different possibilities for CO2 sequestration projects for the fields approaching end of production. This article describes an integrated approach to reservoir simulations focused on evaluating a CO2 injection pilot at one of these matured fields operated by MND and located in the Czech Republic. The CO2 injection site in focus is a naturally fractured carbonate reservoir. This oil-bearing formation has a gas cap and connection to a limited aquifer and was produced mainly by pressure depletion with limited pressure support from water injection. The article summarizes the results of the efforts made by the multi-disciplinary team. An integrated approach was developed starting from geological modeling of a naturally fractured reservoir, integrating the results of laboratory studies and their interpretations (geomechanics and geochemistry), dynamic field data analysis (pressure transient analysis, including time-lapse) and history matching reservoir model enabling simulation of the pilot CO2 injection. The laboratory studies and field data analysis provided descriptions of stress-sensitive fracture properties and safe injection envelope preventing induced fracturing. The impact of potential salt precipitation in the near wellbore area was also included. These effects are considered in the context of a pilot CO2 injection and addressed in the reservoir simulations of injection scenarios. Single-porosity and permeability reservoir simulations with a dominating fracture flow and black-oil formulation with CO2 simulated as a solvent were performed in this study. The arguments for the choice of the simulation approach for the site in focus are shortly discussed. The reservoir simulations indicated a larger site injection capacity than that required for the pilot injection, and gravity-driven CO2 migration pathway towards the gas cap in the reservoir. The application of the approach to the site in focus also revealed large uncertainties, related to fracture description and geomechanical evaluations, resulting in an uncertain safe injection envelope. These uncertainties should be addressed in further studies in preparation for the pilot. The article concludes with an overview of the outcomes of the integrated approach and its application to the field in focus, including a discussion of the issues and uncertainties revealed. [ABSTRACT FROM AUTHOR]

Introduction Effective approaches and policies including identifying priorities and optimal water allocation techniques, especially in basins with different users are considered essential for sustainable development in each region. With 1100 m3 of renewable water per person per year, Iran is considered to be the most critical region in the world in terms of water resources. Unfortunately, most plans in the water sector of such countries are based on local economic growth, and no attention is paid to the amount of available water resources. Considering the issue of a water crisis and the droughts of the last few years, the issue of water resources management has gained high importance. To overcome the mentioned problems, it is inevitably essential to use newly developed water management techniques based on advanced approaches. Although optimization techniques are well-known tools in these issues, the simulation method is utilized as a helpful approach. To simulate water management in the basin, there are various available models. RIBASIM, MIKE BASIN, WEAP, and MODSIM models are famous and user-friendly ones in this collection. WEAP software is a comprehensive and advanced water resource system simulation tool widely used in watershed management and can consider physical and hydrological processes. The scenarios that can be investigated with this software include population growth, economic development, changing the policy of operating reservoirs, extracting more from underground water resources, saving water, allocating ecosystem needs, integrated use of surface and underground water, reuse of water, etc. Materials and Methods This study was conducted in the Nahand catchment area which is located in East Azerbaijan province. Nahand river is the main draining course of this catchment, on which a dam has been built to supply a part of Tabriz's drinking water. To control the performance indicators of the reservoir, several management and exploitation scenarios were developed and evaluated in the WEAP model. The WEAP model was presented in 1990 by the Stockholm Environment Institute (SEI). It is a comprehensive and advanced model for simulating water resource systems, which is extensively used in the management of water resources in watersheds. This model has provided a practical tool for water resource planning and policy analysis to put all the issues related to water resources and uses in a single environment. The WEAP model is capable of simulating issues related to consumption such as water consumption patterns, water reuse strategies, costs, and water allocation patterns, as well as issues related to resources such as river flow, groundwater resources, reservoirs, and water transmission lines. The inputs of the WEAP model include data on the population of Tabriz City, per capita consumption of drinking water per person, the amount of water wastage in the distribution network, the inlet discharge of the Nahand reservoir, the information of the Nahand dam, the amount of cultivated area, etc., and to evaluate the model R^2, RMSE, and MAE statistical indicators were used in two periods of calibration and validation. Then, various operating conditions were investigated by compiling the Reference (continuation of the status quo), SC1 (increase of input flow by 10%), and SC2 (decrease of input flow by 10%) scenarios. Besides, Reservoir performance indicators are used to measure its performance under different operating circumstances. Results and Discussion The simulation results of the studied area indicated that the WEAP model with evaluation criteria including R2, RMSE, and MAE in the calibration stage was 0.89, 1.16, and 1.01 MCM, respectively, and in the validation stage were 0.88, 6.22, and 6.01 MCM, respectively. The results also showed that the amount of water demand for the near future period (2021-2040) will increase due to the increase in population, and therefore, the resources in the basin will not be able to meet all assumed needs. The findings showed that the studied system for the near future period (2021-2040) under the reference (continuation of the status quo), SC1 (increase in flow by 10 %) and SC2 (decrease in flow by 10 %) scenarios from the drinking water supply point of view, will result in a shortage of 28.1, 7.3 and 44.3%, respectively, and from the supply of agricultural needs point of view will result in 31.4, 18.3 and 44.4%, respectively. Also, by evaluating the reservoir's performance indicators, it was found that under all assumed scenarios, the system will fail under the condition of supplying 100% and 80% of the needs, whereas the reservoir will be more sustainable by applying the SC1 scenario in comparison with the other two scenarios. Conclusion To choose the best management and exploitation scenarios, due to existing circumstances and limitations such as time limitation, cost, possible risks to the environment, etc., it is not possible to apply all scenarios in the basins and, thence, it is logical to choose the most suitable scenario. Therefore, software tools can help experts to make decisions by considering all limitations. By examining the results of the reservoir performance indicators, it can be seen that the reservoir will encounter failure in supplying 100 and 80% of the needs in the future period under all scenarios and the sustainability index of the reservoir (remedial stability) in supplying 100%. The needs under the Reference, SC1, and SC2 scenarios will reach 31, 49, and 22%, respectively, and in meeting 80% of the needs, the sustainability index will be slightly higher. [ABSTRACT FROM AUTHOR]

Underground hydrogen (H 2) storage (UHS) is crucial for the H 2 economy, offering scalable and long-term solutions vital for reliable and sustainable H 2 -based energy systems. Cushion gas is a common component in UHS designs, but its impact on H 2 storage performance is not yet fully understood. This study bridges this research gap by systematically investigating the effects of various cushion gas scenarios on UHS performance in porous rocks, specifically saline aquifers and depleted gas reservoirs, using reservoir simulations and deep learning. Firstly, we conducted 8,000 multiphase compositional reservoir simulations for UHS operations in two types of storage formations, considering four cushion gas scenarios: no cushion gas, methane (CH 4), nitrogen (N 2), and carbon dioxide (CO 2). These simulations cover a wide range of geological and operational parameters relevant to practical UHS projects. From these simulations, we derived key insights into the physical mechanisms governing UHS processes and proposed critical storage performance metrics, including H 2 withdrawal efficiency, produced H 2 purity, produced gas–water ratio, and well injectivity. Then, we developed a unified reduced-order model (ROM) using a deep neural network (DNN) based on the comprehensive simulation results. The DNN architecture, with an input layer, five hidden layers, and an output layer, takes 12 geological and operational parameters as inputs, and forecasts the four performance metrics. The ROM accurately predicts the cyclic evolution of performance metrics and is over 5,000,000 times faster than traditional physics-based simulations, allowing for thorough uncertainty quantification of UHS performance prediction under various geological and operational conditions. Key findings of this study include: (a) UHS in porous rocks is technically promising, with improving storage performance over cycles; (b) UHS in saline aquifers has higher withdrawal efficiency and purity but lower gas–water ratio and injectivity compared to depleted gas reservoirs; (c) In saline aquifers, cushion gas reduces withdrawal efficiency and purity but significantly enhances gas–water ratio and injectivity, making it particularly beneficial in scenarios with high water production risks; (d) In depleted gas reservoirs, cushion gas is less important under the current operational conditions, as it barely affects withdrawal efficiency and purity, and these reservoirs inherently exhibit high gas–water ratio and injectivity due to initial/leftover native gas (mostly CH 4); (e) The cushion gas type slightly affects storage performance, with N 2 slightly outperforming CH 4 and CO 2 in withdrawal efficiency and purity, CO 2 being better for gas–water ratio, and CH 4 for injectivity. The findings of this study support cushion gas designs in future UHS projects in both saline aquifers and depleted gas reservoirs. Moreover, the developed ROM offers a highly efficient tool for conducting extensive sensitivity analysis and uncertainty quantification, optimizing operational conditions, and facilitating the screening of potential field sites for future UHS projects. • Performed extensive reservoir simulations for underground hydrogen storage (UHS). • Analyzed the physical mechanisms affecting UHS performance. • Developed a unified reduced-order model (ROM) leveraging deep learning. • Conducted thorough uncertainty quantification to evaluate cushion gas effects. [ABSTRACT FROM AUTHOR]

This paper proposes an adaptive space-time algorithm based on domain decomposition for the large-scale simulation of a recently developed thermodynamically consistent reservoir problem. In the approach, the bound constraints are represented by means of a minimum-type complementarity function to enforce the positivity of the reservoir model, and a space-time mixed finite element method is applied for the parallel-in-time monolithic discretization. In particular, we propose a time-adaptive strategy using the improved backward differencing formula of second order, to take full advantage of the high degree of space-time parallelism. Moreover, the complicated dynamics with higher nonlinearity of space-time discretization require some innovative nonlinear and linear solution strategies. Therefore, we present a class of modified semismooth Newton algorithms to enhance the convergence rate of nonlinear iterations. Multilevel space-time restricted additive Schwarz algorithms, whose subdomains cover both space and time variables, are also studied for domain decomposition-based preconditioning. Numerical experiments demonstrate the robustness and parallel scalability of the proposed adaptive space-time algorithm on a supercomputer with tens of thousands of processor cores. [ABSTRACT FROM AUTHOR]

Recently, discontinuous polymer flooding has been proposed and successfully applied in some offshore oilfields. The performance of discontinuous polymer flooding depends on various operational parameters, such as injection timing, polymer concentrations, and crosslinker concentrations of four types of chemical slugs. Because the number of the operational parameters are large and they are nonlinearly related, the traditional reservoir numerical simulation might not simultaneously obtain the optimal results of these operational parameters. In this study, to simulate the discontinuous polymer flooding processes, a simulation model was built using a commercial reservoir simulator (CMG STARS), in which the mechanisms of the four types of chemical slugs were considered, such as polymer viscosification, adsorption, and degradation. Then, a PSO–ICA algorithm was developed by using the PSO algorithm to improve the exploration ability of the ICA algorithm. The codes were written with MATLAB and linked to CMG STARS to perform optimization processes. Finally, the PSO–ICA algorithm was compared with the ICA and PSO algorithms on benchmark functions to verify its reliability and applied to optimize a discontinuous polymer flooding process in a typical offshore oilfield in Bohai Bay, China. The results showed that the developed PSO–ICA algorithm had lower iteration numbers, higher optimization accuracy, and faster convergence rate than these of PSO and ICA, indicating that it was an effective method for optimizing the operational parameters of discontinuous polymer flooding processes. Compared to the continuous polymer flooding, the discontinuous polymer flooding had a higher oil production rate, a lower water cut, and a lower residual oil saturation. The net present value of the optimal scheme of discontinuous polymer flooding reached 7.49 × 108 $, which is an increase of 6% over that of the scheme of continuous polymer flooding. More research including selecting more reasonable parameters of the PSO–ICA algorithm to increase its optimization accuracy and convergence rate, comparing with other available optimization algorithms, and verifying the performance of the optimal scheme of discontinuous polymer flooding in the practical offshore oilfield will be required in the future. [ABSTRACT FROM AUTHOR]

Carbon capture and storage technologies play a crucial role in mitigating climate change by capturing and storing carbon dioxide emissions underground. Saline aquifers, among other geological formations, hold promise for long-term CO2 storage. However, accurately assessing their storage capacity and CO2 behavior underground necessitates advanced numerical simulation and modeling techniques. In this study, we introduce an approach based on a solubility thermodynamic model that leverages cubic equations of state offline from the simulator. This approach enables the precise prediction of CO2–brine equilibrium properties and facilitates the conversion of compositional data into black oil PVT data suitable for black oil simulations. By incorporating industry-scale saline aquifer properties, we simulate a carbon storage scheme using the black oil model technique, significantly reducing computation time by at least four times while preserving the essential physical phenomena observed in underground carbon storage operations. A comparative analysis between black oil and compositional simulations reveals consistent results for reservoir pressure, CO2 saturation distributions, and mass fraction of trapping mechanisms, with differences of less than 4%. This validation underscores the reliability and efficiency of integrating the black oil model technique into carbon storage simulations in saline aquifer formations, offering tangible benefits to industry operators and regulators by striking a balance between accuracy and efficiency. The capability of this approach to extend to temperatures of up to 300 °C and pressures of up to 600 bars broadens its applicability beyond conventional CCS applications, serving as a valuable tool for optimizing decision-making processes in CCS projects, particularly in scenarios where profitability may be marginal. [ABSTRACT FROM AUTHOR]

Carbon capture, utilization, and storage (CCUS) is a crucial part of the energy industry nowadays, aiming to reduce the overall carbon emission into the environment. One solution to CCUS is via the means of CO 2 enhanced recovery processes in a depleted oil reservoir. In such a case, life-cycle production optimization plays a crucial component, referring to optimizing a production-driven objective function via varying well controls during a reservoir's lifetime. One challenge is to obtain the optimal cash flow while trying to maintain the maximum CO 2 storage. Another challenge is the nonlinear constraints (such as field liquid production rate) which need to be honored due to the capacity of the processing facilities. This study presents an application of a stochastic gradient-based framework to solve the CO 2 storage multi-objective optimization problem. Our study focuses on carbon capture and storage via the means of nonlinearly constrained production optimization workflow for a CO 2 enhanced recovery process, in which we aim to bi-objectively maximize both the net-present-value (NPV) and the net present carbon tax credits (NPCTC). The main framework used in this work is line-search sequential quadratic programming (LS-SQP) with stochastic simplex approximated gradients (StoSAG). We demonstrate the performance and results of the algorithmic framework in a field-scale realistic problem. The case study being investigated is a multiphase flow Brugge model under CO 2 injection, simulated using a commercial compositional reservoir simulator. Results show that the LS-SQP algorithm with StoSAG gradients performs computationally efficiently and effectively in handling nonlinear state constraints imposed onto the problem. The workflow successfully solves both the single-objective and the multi-objective optimization problems with minimal and acceptable constraint violations. Various numerical settings have been experimented with to estimate the Pareto front for the bi-objective optimization problem, showing the trade-off between the two objectives NPV and NPCTC. We have demonstrated an approach to the carbon capture, utilization, and storage (CCUS) in the context of multi-objective production optimization of a CO 2 enhanced recovery process for a field-scale realistic reservoir model. The algorithmic framework used in this study has proven to be computationally effective on the problem and especially useful when utilized in conjunction with commercial flow simulators that lack the capability of computing adjoint-based gradients. [ABSTRACT FROM AUTHOR]

Reservoir simulation is traditionally based on the assumption that water is an inert phase, while hydrocarbon components split into oil and gas phases. This approach is usually reasonable when modeling conventional hydrocarbon recovery, but specific applications may require accounting for mass exchange between the water and hydrocarbon phases. We here present the extension of our Graphics Processing Units (GPUs) compositional reservoir simulator (Esler et al. SPE J. 27(01), 597–612, 2021) to support gas-water equilibrium. Specifically, the Søreide and Whitson equation of state (EoS) (Søreide and Whitson Fluid Phase Equilib. 77, 217–240, 1992) was implemented to compute mutual solubilities of hydrocarbon/brine mixtures. The impact of salinity on phase equilibrium is accounted for, with salt being treated as an active tracer. The simulator uses a mass-variables formulation, meaning that little modifications to the construction of transport equations and Jacobian assembly was required; most of the required code changes are localized in the EoS module for the computation of component fugacities, and phase properties such as partial molar fractions and partial molar volumes. Treating salt as an active tracer instead of defining a further pseudo-component has an important advantage with the Søreide and Whitson EoS. If salinity changes as in water vaporization processes, our choice ensures that flash iterations can still be cast as a Gibbs Minimization problem with salinity being a constant parameter. On the contrary, salinity would change as flash iterations progress, jeopardizing the thermodynamic consistency of the phase equilibria. The overall reservoir simulation system of equations is still accurate to first order in time, at the cost of possibly slight volume imbalances at the end of converged timesteps. In this paper, we focused on CO 2 sequestration in saline aquifers, where solubility trapping is a key mechanism. The accuracy of our implementation with respect to conventional CPU ones is first demonstrated on a synthetic box model. We then select an open-access aquifer model (Gassnova 2016) to illustrate its applicability in an industrial setting. Finally, we show how being able to seamlessly run high resolution models allows for modeling of convective mixing. A key conclusion of this work is that the extreme performance of GPU-based reservoir simulation naturally transfers to new fields of study, which is critical when modeling saline aquifers whose extent is an order of magnitude larger than that of typical oil and gas fields. [ABSTRACT FROM AUTHOR]

Monitoring is an important component of geological carbon storage operations because it provides data that can be used to estimate key quantities such as CO 2 plume location. The design of the monitoring strategy is complicated, however, because the monitoring plan must be established prior to the availability of extensive flow data. In this work, we present and apply a framework that integrates monitoring well optimization and (subsequent) history matching. The monitoring well optimization entails finding the locations of monitoring wells such that, with the data acquired at those locations, the expected uncertainty reduction in a particular flow quantity is maximized. This optimization requires the simulation of a large set of prior models, though these simulations need only be performed once for a given injection scenario. Once the monitoring wells are in place and CO 2 injection begins, history matching is performed using the monitoring data. This is accomplished here using an ensemble smoother with multiple data assimilation. The overall framework is applied to variogram-based geomodels that are representative of an actual storage project under development in the USA. Two injection scenarios are considered with two different (synthetic) 'true' models, which provide the observed data. History matched models are constructed using data from both optimally located and heuristically placed monitoring wells. Posterior uncertainty, evaluated in terms of the cumulative distribution function for a metric related to plume extent over the ensemble of history matched models, is shown to be minimized through use of optimized monitoring wells. These results demonstrate the importance of optimizing the monitoring plan, and the degree of uncertainty reduction that can be realistically achieved. [ABSTRACT FROM AUTHOR]

We perform a simulation study of hydrogen injection in a depleted gas reservoir, to assess the geomechanical impact of hydrogen storage, relative to other commonly injected gases (methane, CO 2). A key finding is that the differences in hydrogen density, compressibility, viscosity and thermal properties, compared to the other gases, result in significantly less thermal perturbation at reservoir level. The risks of fault reactivation and wellbore fractures due to thermally-induced stress changes are significantly lower when storing hydrogen compared to results observed in CO 2 scenarios. This implies that hydrogen injection and production has a much smaller geomechanical footprint, with benefits for operational safety. We also find that use of nitrogen cushion gas ensures efficient deliverability and phase separation in the reservoir. However, in this study, a large fraction of cushion gas was back-produced in each cycle, demonstrating the need for further studies of the surface processing requirements and economic implications. [Display omitted] • Geomechanical modelling assesses the risks associated with hydrogen storage. • Diffusivity, dissolution and methanogenesis have negligible effect on models. • Thermal perturbations associated with storage alter fracture pressures. • Significant mixing of hydrogen with cushion gas is observed. • Hydrogen storage presents less geomechanical risk than CO 2 or methane storage. [ABSTRACT FROM AUTHOR]

Evaluating uncertainty in CO 2 injection projections often requires numerous high-resolution geological realizations (GRs) which, although effective, are computationally demanding. This study proposes the use of representative geological realizations (RGRs) as an efficient approach to capture the uncertainty range of the full set while reducing computational costs. A predetermined number of RGRs is selected using an integrated unsupervised machine learning (UML) framework, which includes Euclidean distance measurement, multidimensional scaling (MDS), and a deterministic K-means (DK-means) clustering algorithm. In the context of the intricate 3D aquifer CO 2 storage model, PUNQ-S3, these algorithms are utilized. The UML methodology selects five RGRs from a pool of 25 possibilities (20% of the total), taking into account the reservoir quality index (RQI) as a static parameter of the reservoir. To determine the credibility of these RGRs, their simulation results are scrutinized through the application of the Kolmogorov–Smirnov (KS) test, which analyzes the distribution of the output. In this assessment, 40 CO 2 injection wells cover the entire reservoir alongside the full set. The end-point simulation results indicate that the CO 2 structural, residual, and solubility trapping within the RGRs and full set follow the same distribution. Simulating five RGRs alongside the full set of 25 GRs over 200 years, involving 10 years of CO 2 injection, reveals consistently similar trapping distribution patterns, with an average value of D max of 0.21 remaining lower than D critical (0.66). Using this methodology, computational expenses related to scenario testing and development planning for CO 2 storage reservoirs in the presence of geological uncertainties can be substantially reduced. [ABSTRACT FROM AUTHOR]

This paper reviews and evaluates work on the structural complexity of the potential carbon dioxide storage sites in the North Sea, including the nature of the reservoir structures, the reservoir rocks, the presence of inter-layers, faults, and fractures, and how these factors influence carbon dioxide capacity. In particular, the review emphasises the significance of studying caprocks in detail, not just the reservoir rock's carbon dioxide storage capacity. This work also particularly considers reservoir simulation work on North Sea sites and illustrates the importance of using fully coupled flow–geomechanical–geochemical modelling to ensure that complex feedback and synergistic effects are not missed. It includes comparisons with other sites where relevant. It also discusses recent challenges and controversies that have arisen from simulations of sequestration in North Sea reservoirs and the need for comprehensive field data to resolve these issues. [ABSTRACT FROM AUTHOR]

The constrained pressure residual (CPR) algorithm is a family of well-known and industry-standard preconditioners for large-scale reservoir simulation. The CPR algorithm is a twostage preconditioner to deal with different blocks stage-by-stage, and is often able to effectively improve the robustness behavior and the convergence speed of linear iterations. Nonetheless, as a linear preconditioner, it is hard for the traditional CPR method to be capable of action at the nonlinear level for effectively solving large sparse nonlinear systems of equations with high nonlinearity. In this paper, we present and study the extension of this linear method to the nonlinearly CPR (NCPR) case for solving fractured reservoir problems, to deal with the difficulty of the slow convergence or stagnation from the nonlinear level. In the proposed nonlinear preconditioning, a subspace nonlinear block system is first built and solved to remove the unbalanced nonlinearities of the pressure and the saturation fields, and the fast convergence can then be restored when a variant of semismooth Newton methods is called after the subspace nonlinear block system is solved. Experiments on two or three dimensional porous media applications are presented to demonstrate the applicability and parallel scalability of the aforementioned NCPR method. We also show that the proposed algorithm is superior to the commonly used nonlinear algorithm in terms of the robustness and the positivity-preserving property. [ABSTRACT FROM AUTHOR]

This study provides insights into the experience gained from the investigation of the dynamic behavior of a mature sour hydrocarbon reservoir modeling under an acid gas re-injection process production strategy. The primary objective was to analyze and evaluate the production behavior of proposed injection zones by assessing various injection scenarios and obtaining oil production over time. To achieve that, a workflow was developed to prioritize potential injection areas, select the optimal wells, determine the optimal operational parameters and optimize a pilot application design based on expected performance. Within this framework, the study encompasses diverse acid gas injection schemes on a pilot scale approach, including acid gas combined with waterflooding. The outcome of this analysis will eventually lead to the identification of the most promising and highest-performing injection scheme, elucidating the optimal range of operating parameters. This optimal combination forms the basis for the economic analysis of the venture and the subsequent detailed design of a full-scale application, where real-world implementation will validate the projected results. [ABSTRACT FROM AUTHOR]

The unstable supply of renewable energy due to seasonal dependency contradicts the periodic energy demand. As hydrogen presents high energy density and flow mobility, and low solubility and residual saturation, underground hydrogen storage (UHS) is a promising solution of scalable energy storage to rebalance energy demand and supply. Depleted gas reservoirs (DGR) are a suitable option for UHS due to their demonstrated integrity. In this study, we developed a numerical simulation model based on Tough + RealGasBrine (T + RGB) simulator to evaluate the feasibility of UHS in DGR. We calibrated different Equation-of-State (EOS) models for modeling the phase behavior of hydrogen-included mixtures with experimental data, and this further helps to examine the impact of various cushion gas pre-injection strategies on pressure maintenance and potential hydrogen leakage through cap-rock. After comparison among three EOS models, we adopted the Soave-Redlich-Kwong (SRK) EOS in our simulations as its high accuracy and computational efficiency. Afterward, we simulated one hydrogen injection-idle-withdrawal operation in a high-resolution mesh based on a synthetic heterogeneous anticline DGR. We observed that hydrogen displaces pre-existing methane and resides at the top of the storage zone due to gravity segregation. The average pressure and gas saturation of the whole simulation domain increase with the hydrogen injection. With regard to the hydrogen leakage, when the cap-rock permeability ranges from 10 − 5 to 10 − 3 mD, 0.05% of the injected hydrogen at maximum leaks into the cap-rock while about 1% of injected hydrogen is dissolved into the aqueous phase. Those results demonstrate that DGR is a viable option for UHS. However, only about 73% of injected hydrogen can be recovered if the bottom-hole pressure of the production well is 2 MPa below the reservoir pressure. The cushion gas becomes necessary for the UHS project to increase hydrogen recovery. Injecting cushion gas of nitrogen and carbon dioxide increases hydrogen recovery from the initial 73%–91% and 81%, respectively. The simulation results demonstrate that nitrogen exhibits better performance, in terms of higher hydrogen recovery factor and purity of producing gas, as the cushion gas. In this work, we quantitatively evaluated the hydrogen leakage problem, including leakage into cap-rock, dissolution in water, and mixing with other gas components, which is the first-of-its-kind analysis in literature to the author's best knowledge. The simulation results support the feasibility of UHS in the DGR. • Reservoir properties' evolution during the underground H 2 storage is presented via high-resolution simulations. • Hydrogen loss associated with leakage and dissolution are quantitatively identified. • Performance of the cushion gases, including nitrogen and carbon dioxide, is evaluated. [ABSTRACT FROM AUTHOR]

In binary multi-objective well placement optimization, multiple conflicting objective functions must be optimized simultaneously in reservoir simulation models containing discrete decision variables. Although multi-objective algorithms have been developed or adapted to tackle this scenario, such as the derivative-free evolutionary algorithms, these methods are known to generate a high number of duplicated strategies in discrete problems. Duplicated strategies negatively impact the optimization process since they: (i) degrade the efficiency of recombination operators in evolutionary algorithms; (ii) slow the convergence speed as they require more iterations to find a well-distributed set of strategies; and (iii) perform unnecessary re-evaluations of previously seen strategies through reservoir simulation. To perform multi-objective well placement optimization while avoiding duplicated strategies, this paper investigates the application of a newly proposed algorithm named MOEA/D-NFTS, with a modified diversity preservation mechanism that incorporates prior knowledge of the problem, on a multi-objective well placement optimization problem. The proposed methodology is evaluated on the UNISIM-II-D benchmark case, a synthetic carbonate black-oil simulation model in a well placement optimization problem using a binary strategy representation, indicating the presence or absence of a given candidate well position in the final strategy. The objective functions are the maximization of the Net Present Value, the maximization of the Cumulative Oil Production, and the minimization of Cumulative Water Production. The modified MOEA/D-NFTS performance is compared with a baseline algorithm without diversity preservation, and the evidence shows that the MOEA/D-NFTS produces statistically significant superior results, and is suitable for binary multi-objective well placement optimization. [ABSTRACT FROM AUTHOR]

Oil recoveries from medium and heavy oil reservoirs under natural recovery production are small because of the high viscosity of the oil. Normal water flooding procedures are usually ineffective, as the injected water bypasses much of the oil because of its high mobility. Thermal flooding processes are desirable but have many disadvantages from costs, effects on the environment, and loss of lighter hydrocarbons. Chemical flooding options, such as bio-polymer flooding options, are attractive, as they are environmentally friendly and relatively cheap to deploy and help to increase the viscosity of the injecting fluid, thereby reducing its mobility and increasing its oil recovery. The downside to polymer flooding includes reservoir temperature, salinity, molecular weight, and composition. Six weight percentages of two polymers (xanthan gum, XG, and gum arabic, GA) are dissolved in water, and their viscosity is measured in the laboratory. These viscosities are incorporated with correlations in the Eclipse software to create models with different polymer concentrations of (0.1% wt., 0.2% wt., 0.3% wt., 0.4% wt., 0.5% wt., and 1% wt.). A base case of natural recovery and water injection was simulated to produce an oil recovery of 5.9% and 30.8%, respectively, while at 0.1% wt. and 1% wt., respectively, oil recoveries of 38.8% and 45.7% (for GA) and 48.1% and 49.8% (for XG) are estimated. At 5% and 10% saline conditions, a drop in oil recovery of (4.6% and 5.3%) is estimated during GA flooding and (1.2% and 1.7%) for XG flooding at 1% wt., respectively. XG exhibits higher oil recoveries compared to GA at the same % wt., while oil recoveries during GA floodings are more negatively affected by higher saline concentrations. [ABSTRACT FROM AUTHOR]

In recent years, Machine Learning (ML) has become a buzzword in the petroleum industry, with numerous applications which guide engineers in better decision making. The most powerful tool that most production development decisions rely on is reservoir simulation with applications in multiple modeling procedures, such as individual simulation runs, history matching and production forecast and optimization. However, all of these applications lead to considerable computational time and computer resource-associated costs, rendering reservoir simulators as not fast and robust enough, and thus introducing the need for more time-efficient and intelligent tools, such as ML models which are able to adapt and provide fast and competent results that mimic the simulator's performance within an acceptable error margin. In a recent paper, the developed ML applications in a subsurface reservoir simulation were reviewed, focusing on improving the speed and accuracy of individual reservoir simulation runs and history matching. This paper consists of the second part of that study, offering a detailed review of ML-based Production Forecast Optimization (PFO). This review can assist engineers as a complete source for applied ML techniques in reservoir simulation since, with the generation of large-scale data in everyday activities, ML is becoming a necessity for future and more efficient applications. [ABSTRACT FROM AUTHOR]

This paper summarizes the results of the risk analysis and characterization of the CO2 and brine leakage potential of Farnsworth Unit (FWU) site wells. The study is part of the U.S. DOE's National Risk Assessment Partnership (NRAP) program, which aims to quantitatively evaluate long-term environmental risks under conditions of significant geologic uncertainty and variability. To achieve this, NRAP utilizes risk assessment and computational tools specifically designed to quantify uncertainties and calculate the risk associated with geologic carbon dioxide (CO2) sequestration. For this study, we have developed a workflow that utilizes physics-based reservoir simulation results as input to perform leakage calculations using NRAP Tools, specifically NRAP-IAM-CS and RROM-Gen. These tools enable us to conduct leakage risk analysis based on ECLIPSE reservoir simulation results and to characterize wellbore leakage at the Farnsworth Unit Site. We analyze the risk of leakage from both individual wells and the entire field under various wellbore integrity distribution scenarios. The results of the risk analysis for the leakage potential of FWU wells indicate that, when compared to the total amount of CO2 injected, the highest cemented well integrity distribution scenario (FutureGen high flow rate) exhibits approximately 0.01% cumulative CO2 leakage for a 25-year CO2 injection duration at the end of a 50-year post-injection monitoring period. In contrast, the highest possible leakage scenario (open well) shows approximately 0.1% cumulative CO2 leakage over the same time frame. [ABSTRACT FROM AUTHOR]

Grid-scale underground hydrogen storage (UHS) is essential for the decarbonization of energy supply systems on the path towards a zero-emissions future. This study presents the feasibility of UHS in an actual saline aquifer with a typical dome-shaped anticline structure to balance the potential seasonal mismatches between energy supply and demand in the UK domestic heating sector. As a main requirement for UHS in saline aquifers, we investigate the role of well configuration design in enhancing storage performance in the selected site via numerical simulation. The results demonstrate that the efficiency of cyclic hydrogen recovery can reach around 70% in the short term without the need for upfront cushion gas injection. Storage capacity and deliverability increase in successive storage cycles for all scenarios, with the co-production of water from the aquifer having a minimal impact on the efficiency of hydrogen recovery. Storage capacity and deliverability also increase when additional wells are added to the storage site; however, the distance between wells can strongly influence this effect. For optimum well spacing in a multi-well storage scenario within a dome-shaped anticline structure, it is essential to attain an efficient balance between well pressure interference effects at short well distances and the gas uprising phenomenon at large distances. Overall, the findings obtained and the approach described can provide effective technical guidelines pertaining to the design and optimization of hydrogen storage operations in deep saline aquifers. [Display omitted] • Cyclic storage performance is found by injection at the highest achievable rates. • The storage capacity, deliverability, and recovery efficiency increase cycle-wise. • Co-production of water during hydrogen withdrawal from the aquifer is marginal. • Pressure interference and gas up-rising emerge as main challenges in well planning. • The need for using additional wells can be avoided by optimizing the well distance. [ABSTRACT FROM AUTHOR]

Three different reservoir simulators that utilize both two-phase and three-phase microemulsion phase behavior models are used to model surfactant–polymer flooding to determine and compare their results. Different models are used in each simulator to describe the physical behavior of injected chemicals into the reservoir, which raises the need to benchmark their results. The physical behavior models of polymer and surfactant were constructed and verified on a 1D scale reservoir model and further verified in a 3D model. Finally, simulations were conducted in a field-scale reservoir containing 680,400 grids, where results were compared and analyzed. The 1D and 3D model results suggest an excellent match between the different simulators in modeling surfactant–polymer floods. In the case of the field-scale model, the simulators matched in terms of oil recovery and total volumes produced and injected, while having similar reservoir pressure profiles but with significant discrepancies in terms of injected and produced chemicals. These results indicate that despite the differences in the calculated injected and produced chemicals due to the different models in the simulators, the effect of surfactant–polymer floods on oil recovery, total injected and produced fluids, and average pressure profiles can be comparably modeled in all of the three simulators. [ABSTRACT FROM AUTHOR]

Numerical modeling of multiphase flow phenomena in fractured porous media has always been a hot topic in recent years. Projection-based embedded discrete fracture model (pEDFM) is a recently developed modeling framework that has been shown to resolve some limitations of the commonly used embedded discrete fracture model (EDFM). However, determining the continuous projection path of fracture networks and adding needed fracture-fracture (f-f) connections in the previous pEDFM workflows is too complicated to construct a generic and easy-to-program algorithm. Although there are several studies about applications of pEDFM to numerically model the flow problems in fractured porous media, a generic pEDFM workflow has not been truly formed. Therefore, this theoretically advanced numerical modeling framework of flow problems in fractured porous media has not been widely noticed and applied. To construct a generic pEDFM workflow, this paper first illustrates that the fracture projection path obtained by the micro-translation method can achieve acceptable simulation accuracy for flow across a high-conductivity fracture, then a generic and easy-to-program non-projective method is developed to model low-conductivity fractures. Numerical examples verify that the developed pEDFM workflow can effectively handle low-conductivity fractures (e.g. flow barriers) in general cases, thus avoiding the difficulty of developing a generic algorithm for constructing a continuous projection path for the low-conductivity fracture. It is also verified that the new pEDFM workflow with the addition of simple f-f connections can solve the cases that classical pEDFM solves with significant errors. Furthermore, it achieves similar or better computational accuracy than the pEDFM modified by adding complex f-f connections, thus avoiding the difficulty of developing a generic algorithm for constructing these additional f-f connections. Overall, the first generic pEDFM workflow is developed in this paper, which significantly improves the practicability of pEDFM and may induce more applications of pEDFM in numerical modeling of flow problems in fractured porous media. [ABSTRACT FROM AUTHOR]

Gas-oil gravity drainage is a recognized major contributor to production in fractured reservoirs. While various empirical and analytical methods have been proposed to model this process, many of them contain assumptions that are questionable or require parameters that are not accessible at the field level. The aim of this work is to provide new, easy-to-use scaling equations for estimating the recoverable oil through gravity drainage in naturally fractured reservoirs, considering the effects of resistance capillary pressure. To accomplish this, data from four oilfields undergoing gravity drainage, including rock properties (eight sets), block height (three sets), and fluid properties (four sets), were used to generate a wide range of recovery curves using a single porosity numerical simulation model. Aronofsky's and Lambert's functions were then utilized to match the generated recovery curves. Statistical analysis revealed that the Aronofsky's function is more accurate in replicating the recovery patterns, while the Lambert's function tends to overestimate the early-time oil recovery and underestimate the oil recovery at a later stage in the majority of cases. A sensitivity analysis was subsequently performed, revealing that parameters such as absolute permeability, viscosity of oil, height of block, gas and oil density, characteristics of relative permeability and capillary pressure curves and interfacial tension (IFT) influence the amount of time taken to achieve the final recovery. Of these parameters, absolute permeability has the most significant effect on the amount of time needed to attain the final recovery, while the effect of difference between oil and gas densities is the lowest. Consequently, two different expressions were developed using nonlinear multiple regression analysis of simulated gravity drainage data which can be combined with the Aronofsky model to substitute the rate convergence constant. The new scaling equations include the effects of capillary pressure and other relevant factors in gravity drainage simulations. Both forms show satisfactory accuracy, as evidenced by the statistical parameters obtained (R2 = 0.99 and MSE = 0.0019 for both established correlations). The new correlations were verified using a wide range of oilfield data and are expected to provide a better understanding of the recovery process in naturally fractured reservoirs. [ABSTRACT FROM AUTHOR]

There is a rising interest in developing robust and efficient sequential methods for reservoir simulation due to potential benefits from specialized nonlinear and linear solvers, flexible discretizations, and adaptivity. The recently proposed sequential fully implicit Newton (SFIN) method addresses a major bottleneck of the sequential strategies: the slow sequential coupling convergence when flow and transport problems are strongly coupled. However, the original SFIN algorithm requires fixed primary variables during the simulation. For the natural formulation widely used in reservoir simulation, primary variables will switch when a phase change happens. In this work, we proposed strategies to address the issue of inconsistent primary variables and extended the SFIN method to the natural black-oil formulation. Several challenging numerical cases are presented to demonstrate that the extended SFIN method can achieve significant sequential convergence acceleration and improved overall performance for natural formulation when phase changes and primary variable switch happen frequently. [ABSTRACT FROM AUTHOR]

Reservoir outflow is an important variable for understanding hydrological processes and water resource management. Natural streamflow variation, in addition to the streamflow regulation provided by dams and reservoirs, can make streamflow difficult to understand and predict. This makes them a challenge to accurately simulate hydrologic processes at a daily scale. In this study, three Machine Learning (ML) algorithms, Random Forest (RF), Support Vector Machine (SVM), and Artificial Neural Network (ANN), were examined and compared to model reservoir outflow. Past, current, and future hydrologic and meteorological data were used as model inputs, and the outflow of next day was used as prediction. Simulation results demonstrated that all three models can reasonably simulate reservoir outflow. For Carlyle Lake, the coefficient of determination and Nash–Sutcliffe efficiency were each close to one for the three models. The coefficient of determination, relative mean bias, and root mean square error indicated that the SVM performed better than the RF and ANN, but the SVM output displayed a larger relative mean bias than that from RF and ANN. For Lake Shelbyville, the ANN model performed better than RF and SVM when considering the coefficient of determination, Nash–Sutcliffe efficiency, relative mean bias, and root mean square error. The study results demonstrate that the three ML algorithms (RF, SVM, and ANN) are all promising tools for simulating reservoir outflow. Both the accuracy and efficacy of the three ML algorithms are considered to support practitioners in planning reservoir management. [ABSTRACT FROM AUTHOR]

Nowadays, oil and gas wells can produce excessive amounts of water in many fields. Excessive water production reduces oil and gas productivity by leading to fines and causing sand production and the corrosion of surface facilities. The combination of these unfavourable phenomena results in the early closure of wells as oil and/or gas production becomes unprofitable. In this study, a methodology for simulating water shut-off (WSO) treatments was tested on three synthetic reservoir models in order to conduct sensitivity studies on various parameters affecting the WSO treatment, such as polymer solution type, volume, and injection rate, for the purpose of evaluating different treatment scenarios, defining an optimal treatment design, and forecasting post-treatment well behaviour. Moreover, this protocol explains in detail the different stages of the reservoir simulation methodology for sensitivity analyses and describes the interpretation of WSO treatment results which enable the study of its application feasibility in a real field. [ABSTRACT FROM AUTHOR]

Subsurface energy storage in depleted petroleum reservoirs is a promising technique to balance and optimize the utilization of energy resources. In this work, we numerically explore the possibility of storing excessive hydrogen gas in depleted unconventional gas reservoirs. Our study is a multiscale analysis. From the molecular (pore) scale, we investigate the thermodynamics and transport mechanism of the hydrogen gas in the nanopores of the unconventional reservoirs. Then based on the results of the pore scale, we conduct reservoir-scale simulations to quantitatively investigate the preferred cycling pressure, the effective fraction of cushion gas and the amount of storage capacity of unconventional reservoirs. We have discovered that, compared to conventional gas reservoirs, hydrogen stored in unconventional reservoirs maintains higher purity because of the differential adsorption effect of the nanopores. This feature makes depleted unconventional gas reservoirs appealing candidates for underground storage of the hydrogen gas. • Potentials of hydrogen storage in unconventional reservoirs are investigated. • Molecular dynamics simulation and reservoir simulation are performed. • The preferential adsorption of methane facilitates the storage of hydrogen. • The hydraulically fractured region is the main contributor for hydrogen storage. • The existence of nanopores improves the purity of produced hydrogen. [ABSTRACT FROM AUTHOR]

Underground Hydrogen Storage (UHS) is an emerging large-scale energy storage technology. Researchers are investigating its feasibility and performance, including its injectivity, productivity, and storage capacity through numerical simulations. However, several ad-hoc relative permeability and capillary pressure functions have been used in the literature, with no direct link to the underlying physics of the hydrogen storage and production process. Recent relative permeability measurements for the hydrogen-brine system show very low hydrogen relative permeability and strong liquid phase hysteresis, very different to what has been observed for other fluid systems for the same rock type. This raises the concern as to what extend the existing studies in the literature are able to reliably quantify the feasibility of the potential storage projects. In this study, we investigate how experimentally measured hydrogen-brine relative permeability hysteresis affects the performance of UHS projects through numerical reservoir simulations. Relative permeability data measured during a hydrogen-water core-flooding experiment within ADMIRE project is used to design a relative permeability hysteresis model. Next, numerical simulation for a UHS project in a generic braided-fluvial water-gas reservoir is performed using this hysteresis model. A performance assessment is carried out for several UHS scenarios with different drainage relative permeability curves, hysteresis model coefficients, and injection/production rates. Our results show that both gas and liquid relative permeability hysteresis play an important role in UHS irrespective of injection/production rate. Ignoring gas hysteresis may cause up to 338% of uncertainty on cumulative hydrogen production, as it has negative effects on injectivity and productivity due to the resulting limited variation range of gas saturation and pressure during cyclic operations. In contrast, hysteresis in the liquid phase relative permeability resolves this issue to some extent by improving the displacement of the liquid phase. Finally, implementing relative permeability curves from other fluid systems during UHS performance assessment will cause uncertainty in terms of gas saturation and up to 141% underestimation on cumulative hydrogen production. These observations illustrate the importance of using relative permeability curves characteristic of hydrogen-brine system for assessing the UHS performances. • Experimentally measured hydrogen-brine relative permeability curves show distinct characteristics compared to that of other fluid systems. • UHS performances under different relative permeability curves are compared. • Relative permeability curves from other fluid systems are not suitable to act as proxy to hydrogen-brine system in reservoir-scale UHS simulation. [ABSTRACT FROM AUTHOR]

The injection of carbon dioxide (CO2) is an essential technology for maximizing the potential of hydrocarbon reservoirs while reducing the impact of greenhouse gases. However, because of the complexity of this injection, there will be many different chemical reactions between the formation fluids and the rock minerals. This is related to the clay content of sandstone reservoirs, which are key storage targets. Clay content and clay types in sandstone can vary substantially, and the influence of these factors on reservoir-scale CO2-water-sandstone interactions has not been managed appropriately. Consequently, by simulating the process of CO2 injection in two different clay-content sandstones (i.e., high- and low-clay content), we investigated the effect of the sandstone clay concentration on CO2-water-sandstone interactions in this article. High clay content (Bandera Grey sandstone) and low clay content (Bandera Brown sandstone) were considered as potential storage reservoirs and their responses to CO2 injection were computationally assessed. Our results indicate that the mineralogical composition of the sandstone reservoir significantly varies as a result of CO2-water-sandstone interactions. Clearly, the high clay-content sandstone (Bandera Grey) had a higher maximum CO2 mineral-trapping capacity (6 kg CO2/m3 sandstone) than Bandera Brown Sandstone (low clay content), which had only 3.3 kg CO2/m3 sandstone mineral-storage capacity after 400 years of storage. Interestingly, pH was decreased by ~3 in Bandera Grey sandstone and by ~2.5 in Bandera Brown sandstone. Furthermore, porosity increased in Bandera Grey sandstone (by +5.6%), more than in Bandera Brown Sandstone (+4.4%) after a 400-year storage period. Overall, we concluded that high clay-content sandstone shows more potential for CO2 mineral-trapping. [ABSTRACT FROM AUTHOR]

Waterflooding is one of the methods used for increased hydrocarbon production. Waterflooding optimization can be computationally prohibitive if the reservoir model or the optimization problem is complex. Hence, proxy modeling can yield a faster solution than numerical reservoir simulation. This fast solution provides insights to better formulate field development plans. Due to technological advancements, machine learning increasingly contributes to the designing and building of proxy models. Thus, in this work, we have proposed the application of the two-stage proxy modeling, namely global and local components, to generate useful insights. We have established global proxy models and coupled them with optimization algorithms to produce a new database. In this paper, the machine learning technique used is a multilayer perceptron. The optimization algorithms comprise the Genetic Algorithm and the Particle Swarm Optimization. We then implemented the newly generated database to build local proxy models to yield solutions that are close to the "ground truth". The results obtained demonstrate that conducting global and local proxy modeling can produce results with acceptable accuracy. For the optimized rate profiles, the R2 metric overall exceeds 0.96. The range of Absolute Percentage Error of the local proxy models generally reduces to 0–3% as compared to the global proxy models which has a 0–5% error range. We achieved a reduction in computational time by six times as compared with optimization by only using a numerical reservoir simulator. [ABSTRACT FROM AUTHOR]

In this paper, we introduce and study a highly scalable domain decomposition based solver for the fully implicit solution of subsurface flows in porous media with fractional derivative. In the proposed approach, a fully implicit scheme is proposed to relax the time step size by the stability condition, and the Newton–Krylov method is then applied to guarantee the nonlinear consistency of the resultant algebraic system. Moreover, to accelerate the convergence of the Newton iterations and be favorable for the scalability of the linear solver, the emphasis is placed on introducing the overlapping restricted additive Schwarz (RAS) preconditioners. Numerical experiments are presented to demonstrate the robustness and parallel scalability of the proposed fully implicit algorithm for some subsurface flow problems on a supercomputer. [ABSTRACT FROM AUTHOR]

Hydraulic fracturing is essential for assuring production from unconventional reservoirs with ultra-low permeability. The efficiency of hydraulic stimulation is strongly affected by geological discontinuities, such as faults, joints, and natural fractures. This study proposes robust numerical models for fully coupled hydromechanical simulation of the phenomena present in fracture propagation and fluid migration problems in fractured media. A novel mesh fragmentation technique with an intrinsic pore-cohesive zone approach is developed to simulate unrestricted hydraulic fracture propagation. The proposed method allows studying the effect of some primary parameters on hydraulic and natural fracture interaction. In a reservoir simulation, a 3D hydromechanical formulation for an enhanced dual porosity/dual permeability (EDPDP) model is combined with a discrete fracture model (DFM) to represent a fractured porous formation more realistically. The new model allows the study of the impacts of natural fractures with different orientations at multiple scales on the hydromechanical behavior of the reservoir. Finally, this research proposes a new methodology that integrates a robust fluid-driven fracture propagation model and reservoir simulation, improving the evaluation of production performance. We simulate several hydraulic fracturing scenarios for the assessment of cumulative reservoir production. We also study the effects of multiple length fractures on the hydraulically stimulated reservoir integrating EDPDP-DFM. The numerical results show that natural fractures form preferential paths of HF propagation, enhancing well–reservoir connectivity but reducing hydraulic fracture aperture by fluid leak-off. Fluid viscosity and injection rate control fracture opening, pressure, growth, and fluid leak-off. Finally, secondary fractures significantly impact the estimation of fluid drainage and pore pressure dissipation. Highlights: A new methodology to integrate a robust fracture propagation model and reservoir simulation. Discrete fracture and enhanced dual porosity-dual permeability models are combined to study the effects of fractures of multiple lengths on the hydraulically stimulated reservoir. A higher injection rate was more effective for lower fluid viscosity, enhancing fracture length and opening. Secondary fractures significantly impact the estimation of fluid drainage and pore pressure dissipation. [ABSTRACT FROM AUTHOR]

Geological CO2 sequestration (GCS) has been proposed as an effective approach to mitigate carbon emissions in the atmosphere. Uncertainty and sensitivity analysis of the fate of CO2 dynamics and storage are essential aspects of large-scale reservoir simulations. This work presents a rigorous machine learning-assisted (ML) workflow for the uncertainty and global sensitivity analysis of CO2 storage prediction in deep saline aquifers. The proposed workflow comprises three main steps: The first step concerns dataset generation, in which we identify the uncertainty parameters impacting CO2 flow and transport and then determine their corresponding ranges and distributions. The training data samples are generated by combining the Latin Hypercube Sampling (LHS) technique with high-resolution simulations. The second step involves ML model development based on a data-driven ML model, which is generated to map the nonlinear relationship between the input parameters and corresponding output interests from the previous step. We show that using Bayesian optimization significantly accelerates the tuning process of hyper-parameters, which is vastly superior to a traditional trial–error analysis. In the third step, uncertainty and global sensitivity analysis are performed using Monte Carlo simulations applied to the optimized surrogate. This step is performed to explore the time-dependent uncertainty propagation of model outputs. The key uncertainty parameters are then identified by calculating the Sobol indices based on the global sensitivity analysis. The proposed workflow is accurate and efficient and could be readily implemented in field-scale CO2 sequestration in deep saline aquifers. [ABSTRACT FROM AUTHOR]