Innovative Implementation of Compositional Delumping in Integrated Asset Modelling

New upscaling technique for compositional reservoir simulations of miscible gas injection

A reliable fluid characterization achieved through the use of an Equation of State (EOS) is vital to reproduce the phase behaviour of a reservoir fluid and consequently, to improve understanding of the reservoirs EOR potential. In this case, an EOS model was developed aimed at accurately reproducing the phase behaviour of two reservoir oils in the same field that contain, besides H2S, substantial amounts of organic sulphur components. The tuning of the EOS for both oils was more difficult to achieve than in cases without such level of sulphur components. The tuning of the Peng-Robinson EOS was achieved with 9 pseudo-components for both oils, following the selection of the lumping scheme considered adequate for reservoir simulation studies. Organic sulphur components are concentrated in the higher range of molecular weights and, consequently, were incorporated into the heavy fraction. The tuned EOS was then used on a 1 D reservoir simulation compositional model to match the available slim tube experiments, to verify the validity of the EOS tuning. Using the compositional simulation of slim tube displacements, minimum miscibility pressure (MMP) was estimated for both oils and different injection gases. Although a 9 component scheme is adequate for reservoir performance purposes, engineering of surface facilities requires more detail. Therefore, a delumping methodology was developed to allow engineers to estimate detailed fluid composition from the lumped components obtained from reservoir simulation studies. One of the objectives of this study is to obtain a table of variation of the detailed components distributions that can be used for delumping purposes at different stages of the gas injection project.

An efficient, non-iterative flash calculation algorithm has been developed for equation of state (EOS) compositional reservoir simulation problems. In this algorithm, the flash calculation is carried out by solving two sets of linear equations. The solution of these equations is the mole number of each component in each phase. The phase saturation can thus be obtained without solving the Rachford-Rice equation1. Tests against many PVT data sets show that the new approach gives the same results as those computed by the fugacity iteration based on the EOS. However, the computation time is reduced significantly. A series of reservoir simulations has been conducted by an EOS compositional reservoir simulator, UTCOMP, with this new flash approach. Comparisons are made with those performed by UTCOMP with a rigorous fugacity iteration of the Peng-Robinson5 EOS. The results obtained by these two methods are in very good agreement. However, a considerable reduction of the CPU time is generally accompanied by using the new approach. For the runs in which an immiscible displacement process takes place in the reservoir, the computational burden in the phase behavior calculations is reduced by around 90%. Correspondingly, the reduction of the total simulation time is as much as 65%. Besides, it is found that the number of components used affects on the CPU time of the simulations much less with the new approach. These observations may lead to the time used for an EOS compositional reservoir simulations with the new approach being comparable to a black-oil simulation without sacrificing computational accuracy. Also in this work an attempt is made to employ the non-iterative approach as a tool for providing the initial guess for the rigorous EOS iteration. A considerable reduction of the CPU time is still observed.

Pressure depletion in gas-condensate reservoirs create two-phase flow. It is pertinent to understand the behavior of gas-condensate reservoirs as pressure decline in order to develop proper producing strategies that would increase gas and condensate productivity.
 Eclipse 300 was used to simulate gas-condensate reservoirs, a base case model was created using both black-oil and compositional models. The effects of three Equation of States (EOS) incorporated into the models were analysed and condensate dropout effect on relative permeability was studied.
 Analysis of various case models showed that, gas production was maintained at 500MMSCF/D for about 18 and 12 months for black-oil and compositional models, respectively. However, the compositional model revealed that condensate production began after a period of two months at 50MSTB/D whereas for the black oil model, condensate production began immediately at 32MSTB/D. Comparison of Peng-Robinson EOS, Soave-Redlich-Kwong EOS and Schmidt Wenzel EOS gave total estimates of condensate production as 19MMSTB, 15MMSTB and 9MMSTB and initial values of gas productivity index as 320, 380 and 560, respectively. The results also showed that as condensate saturation increased, the relative permeability of gas decreased from 1 to 0 while the relative permeability of oil increased from 0.15 to 0.85.
 The reservoir simulation results showed that compositional model is better than black-oil model in modelling for gas-condensate reservoirs. Optimal production was obtained using 3-parameter Peng-Robinson and Soave-Redlich-Kwong EOS which provide a molar volume shift to prevent an underestimation of liquid density and saturations. Phase behaviour and relative permeability affect the behaviour of gas-condensate reservoirs.

Performance evaluation of miscible and near-miscible gas injection processes is available through conventional finite difference (FD) compositional simulation, which is widely used for solving large-scale multiphase displacement problems that always require large computation time. A step can be taken to reduce the time needed by considering low-resolution compositional simulation. The model can be adversely affected by numerical dispersion and may fail to represent geological heterogeneities adequately. The number of fluid components can possibly be reduced at the price of less accurate representation of phase behavior. Streamline methods have been developed in which fluid is transported along the streamlines instead of the finite difference grid. In streamline-based simulation, a 3D flow problem is decoupled into a set of 1D problems solved along streamlines, reducing simulation time and suppressing any numerical dispersion.Larger time steps and higher spatial resolution can be achieved in these simulationsparticularly when sensitivities runs are needed to reduce study uncertainties. Streamline-based reservoir simulation, being orders of magnitude faster than the conventional finite difference methods, may mitigate many of the challenges noted above. For gas injection, the streamlines approach could not provide a high resolution or adequate representation for the multiphase displacement. In this paper, the streamline simulations for both compositional and miscible gas injection were tested, in addition, the conventional gas injection scheme and detailed comparison between the FD simulation and the streamline approachareillustrated. Also, guidelines of how streamline can be potentially used to visualize the effect of gas displacement are presented in this paper.

Currently, for the most oil companies it is possible to identify a general tendency that the reservoir permeability for the new prospects is decline. Reserves quality deterioration increases the risk of the future oil production failure. This paper describes the organization of an integrated approach to source data analysis, simulation model constructon and miscible displacement design in a low-permeable reservoir.At the first stage, a reservoir fluid model is constructed. Based on the lack of a qualitative oil sample from the well (original sample was significantly depressurized during testing) and on the analysis of the fluid properties of the analog fields, the prediction of the main oil parameters was made. At the second stage, using static parameters of a low-permeable reservoir, probabilistic simulation models of the field was constructed, providing a coverage of the main spectrum of reservoir uncertainties. Further, import gas resources were estimated for the injection implementation.The main result is an integrated assessment of the impact of uncertainties for oil properties and gas import resources on key economic indicators for a miscible displacement at the field with a low-permeable reservoir using the reservoir simulation model. The analysis verified the key parameters that have the greatest impact on the economy of the project, determined the threshold values for the volume of gas imports, and based on the feasibility study made recommendations on the choice of the pilot area for injection. A program for conducting additional laboratory experiments was also prepared, this study will allow the narrowing of the oil composition uncertainties and its interaction with injection gas. In addition, a program is developed to carry out special core experiments on gas injection, which will make it possible to correct the behavior of phase permeabilities for a miscible displacement.This paper shows that on the basis of reservoir simulation an integrated approach organization for the verification, accounting and analysis of uncertainties on reservoir characteristics, fluid properties and volume of imported third-party gas allows to search for and select an economically viable and sustainable option to develop an oil field using miscible gas injection technology. An implementation of this approach provides a recovery factor increase from 7% to 25%.

Compositional reservoir simulation is a time bound activity demanding complex physics. We review the advantages of machine learning in complex compositional reservoir simulations to determine fluid properties, such as critical temperature and saturation pressure. A machine learning approach to predict critical temperatures during simulation based on the Heidemann-Khalil method is implemented, resulting in more accurate results with lower computational cost, outperforming the standard method and improving performance on a giant field model with compositional gradient and miscible gas injection. The fluid column grades from black oil to gas condensate; accurate phase behavior and miscibility modelling involves a significant number of components. This makes simulation performance one of the biggest challenges. Critical temperature of the mixture is commonly used to determine phase state (phase labeling), a crucial process of reservoir simulation. Mislabeling can result in incorrect physics and convergence issues, particularly in cells with gas-oil displacement. Simulators generally use the Li correlation to calculate the pseudo-critical temperature from the weighted average of component critical temperatures, leading to inaccuracies. The Heidemann-Khalil method is computationally costly, proportional to the cube of the number of components, prohibiting its use for complex compositional simulation models. We use machine learning to efficiently incorporate it into simulation. By using the machine learning approach, neural networks are trained based on a combination of feeds to reproduce the Heidemann-Khalil method with great precision. The accuracy of critical temperature and other fluid property determinations is thus improved, with machine learning ensuring a very low computational cost. Convergence is also improved. With the traditional Li method, especially at the beginning of gas injection, we faced numerical difficulties, and the runtime was slowed down. By implementing the machine learning based method, the convergence is smooth through the entire gas injection cycle, leading to a reduction of total iteration counts. We experience an overall four times speed-up of the simulation model, which greatly enhances the usage of this model in simulation studies. The use of machine learning methods to replace physics in the simulator is an evolving area. By showing a field example of a successful application that improves both accuracy and performance, we contribute to fostering research into new possibilities where physics-informed models will enhance simulation studies.

Integrated asset modelling is highly desirable for de-risking any development or optimization scenario of a hydrocarbon asset. It can exponentially improve the accuracy of decision, lower contingencies and increase profitability of an asset, thereby bringing additional value. In this study we describe an integrated modeling approach to evaluate different compression schemes in a tight clastic reservoir in Kirthar foldbelt of Pakistan. The modeling uses a data driven model that represents network pressure losses and generates decline type curves from the dynamic simulation model, and finally integrates the data driven model to generate forecasts for different compression scenarios. A sensitivity study on different compression schemes under variable operating conditions was conducted using steady state network simulations in OLGA and pressure drops recorded for each. The conventional approach of integrated asset modeling (integrating gathering network with reservoir simulation model) increases simulation run time and is not usually very efficient for screening large number of cases. We use regression analysis to create a data driven model that represents pressure drop in the network and can be used as proxy to the network model. The dynamic simulation model is used to generate decline curves which are then used in a decline curve-based proxy simulation of the reservoir. This technique allows reservoir simulations to be run many times faster and is particularly useful for evaluating the effect of reservoir and network coupling. It can handle models with thousands of wells, potentially reducing the run time from hours to minutes. The integrated asset model is generated by combining data driven model of network with decline curve-based proxy reservoir model using python scripts. This model was then used for evaluating different compression scenarios ranging from 850 psi to 100 psi plant arrival pressures. The new methodology described has enabled construction of an accelerated integrated model which runs in minutes contrary to many hours with conventional approach of dynamic simulation model coupled with network simulator. The integrated approach leads to accurate forecasts taking into account network pressure losses and makes it possible within a reasonable time for a large number of scenarios. As a result of the evaluation through this approach, the compression scheme with plant reception pressure of 200psi is the most optimum and selected for field development. This is the first implementation of such accelerated integrated asset modeling approach in the region and provides a quick and fast way of implementing integrated modeling and evaluating network changes including compression on the production forecast. It will be a useful case study and its accelerated approach will help promote integrated modeling simulations within the region.

PVT and Phase Behaviour of Petroleum Reservoir Fluids

Compositional reservoir simulation of gas injection processes is commonly performed with a reduced number of components to minimize computation time. However, in simulations of downstream processes, more detailed fluid descriptions are required to design and optimize separation processes. To connect up and down-stream processes, delumping approaches have been proposed to deconvolute the well/production streams from a reservoir simulator into a more detailed fluid description. The accuracy of delumping is closely tied to the lumping approach that is used to generate equation of state (EOS) input for the reservoir simulator. To this end, we have previously introduced a flow-based lumping scheme that includes information from the dynamics of a given displacement process into the grouping of detailed components. In this paper, we investigate the accuracy of the delumped production streams from various gas injection processes and compare results from different lumping schemes. We investigate the benefit of applying flow-based lumping schemes in the selection of pseudo-components for compositional reservoir simulation and find an improvement in terms of uniqueness and overall accuracy of the composition of the delumped well streams. Examples are provided from 3D displacement calculations of enhanced recovery by CO2 and separator gas injection for multicomponent reservoir fluids where a substantial amount of PVT data is available (CME, DDE, swelling test and slimtube MMP). Using a commercial reservoir simulator, the delumped compositions of the produced fluids are compared to detailed displacement calculations for three different lumping schemes and two different gas injection processes. We demonstrate that the proposed lumping/delumping strategy provides additional control of the predictive quality over currently available lumping/delumping approaches. In addition, the proposed workflow provides a unique answer to the question of how to lump a detailed fluid description for displacement calculations. A majority of oil fields currently under production as well as new developments are candidates for EOR processes such as gas floods. The approach presented in this paper are directly applicable to the study and design of such EOR processes an offers an improved workflow for integration of up and down-stream processes. This is, to our best knowledge, the first paper that presents a comparison of delumping results for flow-based lumping schemes in the context of gas injection processes in with emphasis on the accuracy of delumped production stream relative to other available lumping/delumping methods.

The reduction of greenhouse gas emissions in order to decelerate the global warming process could be achieved through the emerging process of geological CO2 storage. Also in terms of Enhanced Oil Recovery (EOR) the injection of CO2 as a pure component or as part of a mixture has proved to increase the productivity of oil and gas reservoirs. Optimization techniques have been applied independently to the reservoir and surface models, leading to non-optimal solutions due to the non-dynamic integration between models. A recent trend of the industry is the integration of sub-surface and surface simulators to have a better representation of the fluid production/injection, taking into account the constantly changing interaction between systems. The integrated approach has been used to integrate multiple reservoirs with common and advanced surface facilities to properly model the fluid flow behavior of the asset. Different injection variables, facilities, well completion, number of wells have been included in the analysis and numerical reservoir simulation models have been integrated with a network. As CO2 is captured, it is transported and re-injected to neighbor reservoirs, as an enhancement process for productivity or for storage purpose. After proving the feasibility of facilities for CO2 injection as EOR process or storage, the integrated approach has shown a more comprehensive solution that could be used for the design and further optimization of this type of projects. Analysis of reservoir properties as permeability, temperature, etc. is also taken into account in order to asses the viability of the CO2 optimal injection and storage strategy, while minimizing cost. Finally, integrated asset modeling also shows the flexibility to represent different types of settings such as CO2 source (reservoir and/or fossil fuel power plants), types of reservoirs and network scenarios.

Traditional reservoir simulators cannot capture the complicated interactions between fluid production and reservoir rock deformation during hydrocarbon recovery. In particular, in a recovery process when both phase behavior and deformation of reservoir rock play critical roles, a coupled geomechanics and compositional reservoir model can rigorously capture the above physical relations between solid and fluids and thereby perform more precise history matchings and predictions for better well planning and reservoir management decisions. This paper considers a fully coupled geomechanics and compositional modeling process. A novel solution procedure for the associated fully coupled Jacobian system is presented using the Schur complement technique. The proposed approach has the following features: (1) it has similar convergence properties as the fully implicit method, (2) combined with Krylov subspace iterative solvers, it can deal with symmetric and unsymmetric problems, and (3) it can be easily parallelized to solve large-scale reservoir simulation problems. In this work, a parallel, fully implicit, equation of state compositional reservoir simulator was fully coupled with a geomechanics FEM module. For each Newton iteration, an original two by two block Jacobian system is converted to a Schur complement system, and then the associated Schur complement system is solved using a BiCGstab(l) iterative method; subsequently all the unknowns for the reservoir simulator and the geomechanics module are obtained at the same time. The above partition solution procedure is equivalent to a block Gaussian elimination; hence its convergence is similar to a fully implicit coupled procedure. The proposed approach is implemented to solve two- and three-dimensional compositional reservoir simulation case studies considering geomechanics effects. We also compare the results between the fully coupled and the iteratively coupled approaches. It is shown that this approach presents a viable alternative in solution methods for solving general coupled physical problems.

The presence of hydrogen sulphide (H2S) in produced reservoir fluids mandates precautions in the design and operation of the surface facilities. The toxicity and corrosive nature of H2S, and the need to prevent both plugging of reservoir formations and increasing the sulphur content of the produced oil dictates the criticality of forecasting and monitoring the volumes and concentrations of H2S flowing through the whole asset. Ensuring the concentration is within acceptable operational limits is critical to safeguard the overall asset and the integrity of the surface pipeline network. The objective of this study was to utilize a history matched Digital Twin Integrated Asset Model (IAM) to predict the volumes and concentrations of H2S in a field located offshore Abu Dhabi by modeling the multi-stage separation, H2S removal, and re-injection facilities for gas injection and gas lifts. The field consists of multiple stacked carbonate reservoirs sharing the same surface facilities. The proposed modelling of H2S removal strategy involved a series of steps beginning with the sweetening of the produced associated gas for fuel gas requirements and mixing the extracted H2S volumes with the gas injection and gas lift streams. The sweetening process effectively mitigated any potential asset integrity issues arising due to corrosion of the power generation system and other surface facility assets. The stripped H2S gas, re-combined with the remaining produced gas, was used for gas-lifts and reinjected into the lower reservoirs for pressure maintenance and enhanced oil recovery (EOR). A next-generation surface-subsurface coupled simulator was utilized for the modeling of this field including the full asset surface pipeline network, the H2S removal plant, bypass lines and re-injection facilities for gas injection and gas-lifts. The Digital Twin IAM approach provided a robust method for tracking and predicting the concentration and volume of H2S in the produced gas over a period of 50 years. The simulation allowed tracking the H2S from its initial location in the reservoirs, into the production wells, then through the pipelines, all the way to the surface facilities where the sweetening of the produced is handled. Moreover, the use of the Digital Twin allowed the verification of the disposal plan of the extracted H2S, showing that mixing it with the re-injection gas stream is a feasible option. Recommendations based on the model were provided to the production and facilities team, leading to a robust long-term field development plan that ensures asset integrity.

Standard EOS-based phase equilibrium modeling in reservoir simulators involves computationally intensive and time-consuming iterative calculations for stability analysis and flash calculations. Therefore, speeding up stability analysis and flash calculations and improving robustness of the calculations are of utmost importance in compositional reservoir simulation. Prior knowledge of the tie lines traversed by the solution of a gas injection problem translates into valuable information with significant implications for speed and robustness of reservoir simulators. The solution of actual gas injection processes follows a very complex route due to dispersion, pressure variations, and multi-dimensional flow. The multiple mixing-cell (MMC) method originally developed to calculate minimum miscibility pressure of a gas injection process, accounts for various levels of mixing of the injected gas and initial oil. This observation suggests that the MMC tie lines developed upon repeated contacts may represent a significant fraction of the actual simulation tie lines encountered. We investigate this idea and use three tie-line-based K-value simulation methods for application of MMC tie lines in reservoir simulation. In two of the tie-line-based K-value simulation methods, we examine tabulation and interpolation of MMC tie lines in a framework similar to the compositional space adaptive tabulation (CSAT). In the third method, we perform K-value simulations based on inverse distance interpolation of K values from MMC tie lines. We demonstrate that for the displacements examined the MMC tie lines are sufficiently close to the actual simulation tie lines and provide excellent coverage of the simulation compositional route. The MMC-based methods are then compared to the computational time using other methods of phase equilibrium calculations including a modified application of CSAT (an adaptive tie-line-based K-value simulation), a method utilizing only heuristic techniques, and the standard method in an IMPEC-type (implicit pressure, explicit concentration) reservoir simulator. The results show that tabulation and interpolation of MMC tie lines significantly improves phase equilibrium and computational time compared to the standard approach with acceptable accuracy. The results also show that computational performance of the MMC-based methods with only prior tieline tables are very close to that of CSAT, which requires flash calculations during simulation. The K-value simulations using MMC-based tie-line interpolation methods improve the total computational time up to 51% in the cases studied with acceptable accuracy. The results suggest that MMC tie lines represent a significant fraction of the actual tie lines during simulation and can be used to significantly improve speed and robustness of phase equilibrium calculations in reservoir simulators.

This study addresses miscible solvent design for the planned hydrocarbon flood in the Virginia Hills Beaverhill Lake Unit in north central Alberta. The development a/miscibility between reservoir fluid and a series of increasingly rich solvents was studied in a rising bubble apparatus. In order to establish the validity of this relatively new technique of solvent design, the rising bubble test observations were compared to results of conventional slim tube tests previously conducted with the same soh1ents. A very good correlation was found between the minimum solvent enrichment/or miscibility determined by the two techniques. The rising bubble method is shown to be not only much faster but also superior in some respects to slim tube testing. Introduction The Virginia Hills Beaverhill Lake reservoir, a carbonate reef formation in north central Alberta, was discovered in 1957 and has been under water flood since 1963. Of the original estimated 60 × 106 m3 oil in place, about one-third has been recovered to date. Current plans are for implementation of a miscible enriched gas flood, with solvent injection targeted for 1990(1). The relatively high reservoir temperature (100 °C) and intermediate pressure (20 MPa to 27 MPa) are common to several Alberta pools where miscible gas injection is currently underway. The Virginia Hills development is a joint venture project with Shell Canada as the operator. The annual cost of the solvent (a blend of dry gas and natural gas liquids streams) required by the project is estimated to exceed $30 million. It is thus important to design the solvent containing the minimum amount of expensive natural gas liquids (NGL) yet be miscible so as to maximize the oil recovery. As the composition of the NGL stream available from the pipeline is expected to vary throughout the miscible flood project life, solvent enrichment levels have to be determined for a variety of NGL stream compositions. Given the time consuming and costly nature of the slim tube tests (and our unwillingness to rely entirely on minimum solvent enrichment predicted with the equation of state), the rising bubble apparatus (RBA) tests present an attractive alternative. However, while the slim tube test is universally accepted by the industry, the validity of this more recently emerged method of solvent design has to be established. A systematic comparison of the two techniques was therefore undertaken, the results of which are presented in this paper. Equation of state (EOS) calculations were used throughout the study to interpret, in qualitative terms, the oil-solvent phase behaviour and its trends. Background - Mass Transfer Mechanism in Virginia Hills Miscible Rich Gas Drive Table 1 shows the composition of the Virginia Hills recombined reservoir fluid (VH fluid). This is a light but somewhat waxy crude containing about 3070 (wt) asphaltenes. The compositions of typical dry gas and NGL streams available for blending the solvent are also shown in Table 1. Throughout this paper, the solvents are referred to by their NGL to dry gas molar ratio; 75/25 solvent thus contains 750/0 NGL and 250/0 dry gas.