Day 1 Tue, March 28, 2023最新文献

{"title":"Pressure Jump Stabilization for Compositional Poromechanics on Unstructured Meshes","authors":"Ryan M. Aronson, François P. Hamon, N. Castelletto, Joshua A. White, H. Tchelepi","doi":"10.2118/212206-ms","DOIUrl":"https://doi.org/10.2118/212206-ms","url":null,"abstract":"\u0000 While commonly used in practice for large-scale simulation of coupled subsurface flow and displacement, discretizations in which the solid matrix displacement is represented using linear, nodal elements and flow variables are represented as piecewise constants over each cell are not inherently inf-sup stable. This means that when undrained and incompressible conditions are approached, spurious pressure oscillations will appear in the numerical solution. This is particularly relevant in simulations of carbon sequestration, where the caprock above the injection location should be nearly impermeable. In this work we extend the idea of pressure jump stabilization to the compositional poromechanics setting in order to suppress these spurious oscillations. We apply this method to simulations of CO2 injection into a synthetic aquifer which is represented using a fully unstructured mesh. The results show that the stabilization is effective at smoothing the pressure field without adversely affecting the prediction quality of other quantities of interest.","PeriodicalId":225811,"journal":{"name":"Day 1 Tue, March 28, 2023","volume":"105 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124093033","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Guided Deep Learning Manifold Linearization of Porous Media Flow Equations","authors":"M. Dall'Aqua, E. Coutinho, E. Gildin, Zhenyu Guo, Hardikkumar Zalavadia, S. Sankaran","doi":"10.2118/212204-ms","DOIUrl":"https://doi.org/10.2118/212204-ms","url":null,"abstract":"\u0000 Integrated reservoir studies for performance prediction and decision-making processes are computationally expensive. In this paper, we develop a novel linearization approach to reduce the computational burden of intensive reservoir simulation execution. We achieve this by introducing two novel components: (1) augment the state-space to yield a bi-linear system, and (2) an autoencoder based on a deep neural network to linearize physics reservoir equations in a reduced manifold employing a Koopman operator. Recognizing that reservoir simulators execute expensive Newton-Raphson iterations after each timestep to solve the nonlinearities of the physical model, we propose \"lifting\" the physics to a more amenable manifold where the model behaves close to a linear system, similar to the Koopman theory, thus avoiding the iteration step. We use autoencoder deep neural networks with specific loss functions and structure to transform the nonlinear equation and frame it as a bilinear system with constant matrices over time. In such a way, it forces the states (pressures and saturations) to evolve in time by simple matrix multiplications in the lifted manifold. We also adopt a \"guided\" training approach: our training process is performed in three steps: we initially train the autoencoder, then we use a \"conventional\" MOR (Dynamic Mode Decomposition) as an initializer for the final full training when we use reservoir knowledge to improve and to lead the results to physically meaningful output.\u0000 Many simulation studies exhibit extremely nonlinear and multi-scale behavior, which can be difficult to model and control. Koopman operators can be shown to represent any dynamical system through linear dynamics. We applied this new framework to a two-dimensional two-phase (oil and water) reservoir subject to a waterflooding plan with three wells (one injector and two producers) with speed ups around 100 times faster and accuracy in the order of 1-3 percent on the pressure and saturations predictions. It is worthwhile noting that this method is a non-intrusive data-driven method since it does not need access to the reservoir simulation internal structure; thus, it is easily applied to commercial reservoir simulators and is also extendable to other studies. In addition, an extra benefit of this framework is to enable the plethora of well-developed tools for MOR of linear systems. This is the first work that utilizes the Koopman operator for linearizing the system with controls to the author's knowledge. As with any ROM method, this can be directly applied to a well-control optimization problem and well-placement studies with low computational cost in the prediction step and good accuracy.","PeriodicalId":225811,"journal":{"name":"Day 1 Tue, March 28, 2023","volume":"4 16","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"113955028","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Data-Driven Deep Learning Framework for Microbial Reaction Prediction for Hydrogen Underground Storage","authors":"Klemens Katterbauer, Abdallah Al Shehri, A. Qasim, A. Yousif","doi":"10.2118/212187-ms","DOIUrl":"https://doi.org/10.2118/212187-ms","url":null,"abstract":"\u0000 As the use of hydrogen gas (H2) as a renewable energy carrier experiences a major boost, one of the key challenges for a constant supply is safe and cost-efficient storage of surplus H2 to bridge periods with low energy demand. For this purpose, underground gas storage (UGS) in salt caverns, deep aquifers and depleted oil-/gas reservoirs has been proposed, which provide suitable environments with potentially high microbial abundance and activity. Subsurface microorganisms can use H2 in their metabolism and thus may lead to a variety of undesired side effects such as H2 loss into formation, H2S build up, methane formation, acid formation, clogging and corrosion.\u0000 We present a new AI framework for the determination of metabolism processes of subsurface microorganisms in hydrogen underground storage. The AI framework enables to determine the potential microbial related processes and reactions in order to optimize storage strategies as well as incorporate potential remediating actions to ensure efficient and safe underground hydrogen storage and minimize related side effects.\u0000 We evaluated the framework on investigating potential microbial reactions for hydrogen storage in the Pohokura gas field in New Zealand. The framework evaluates reservoir parameters, such as temperature, pressure, salinity and hydrogen injection volumes as well as duration, and then classifies which reactions may take place as well as indicates the likelihood of the reaction taking place. For the deep learning framework, an optimized random forest algorithm was implemented and compared to other multi-class classification problems. The results demonstrated the ability to determine the microbial reactions that may take place with hydrogen storage reservoir as well as its severity, which enhances the optimization of injection strategy as well as suitability of a reservoir.\u0000 This framework represents an innovative approach to microbial reaction prediction for underground hydrogen storage. The framework allows potential reservoirs to be efficiently evaluated and optimization strategies to be utilized in order to maximize the efficiency of underground hydrogen storage.","PeriodicalId":225811,"journal":{"name":"Day 1 Tue, March 28, 2023","volume":"72 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121419275","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Consistent Discretization Methods for Reservoir Simulation on Cut-Cell Grids","authors":"F. Alpak, M. Jammoul, M. Wheeler","doi":"10.2118/212213-ms","DOIUrl":"https://doi.org/10.2118/212213-ms","url":null,"abstract":"\u0000 Discretization methods have been developed to accompany a novel cut-cell gridding technique for reservoir simulation that preserves the orthogonality characteristic in the lateral direction. A major drawback of the cut-cell gridding method is that polyhedral cells emerge near faults that have relatively small volumes. Pragmatic but non-rigorous approximation methods have been developed in the past to merge these cells with their neighbors so that the grid representation fits the two-point flux approximation (TPFA) framework. In this work, we take a different approach and investigate the global and local applications of select consistent discretization methods in the vicinity of fault representations on cut-cell grids.\u0000 We develop and test consistent discretization methods that are of low computational cost and do not require major intrusive changes to the solver structure of commercial reservoir simulators. 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. Therefore, we develop and test variants of the AvgMPFA and NTPFA methods that are specifically designed to operate on cut-cell grids. An implementation of the well-established but computationally expensive MPFA method is also made for cut-cell grids to serve as a reference to computations with AvgMPFA and NTPFA. All investigated methods are implemented within the framework of a full-physics 3D research simulator with a general compositional formulation, which encompasses black-oil models.\u0000 We use a set of synthetic cut-cell grid models of varying complexity including conceptual models and a field-scale model. We compare the novel cut-cell adapted AvgMPFA and NTPFA simulation results in terms of accuracy and computational performance against the ones computed with reference MPFA and TPFA methods. We observe that AvgMPFA consistently yields more accurate and computationally efficient simulations than NTPFA on cut-cell grids. Moreover, AvgMPFA hybrids run faster than NTPFA 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 owing to its relatively wider stencil. Auspiciously, only one or two \"rings\" of orthogonal blocks around cut cells are sufficient with AvgMPFA to deliver high accuracy.","PeriodicalId":225811,"journal":{"name":"Day 1 Tue, March 28, 2023","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126103306","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Coupled Geomechanics and Fluid Flow Modeling for Petroleum Reservoirs Accounting for Multi-Scale Fractures","authors":"Dawei Wu, Y. Di, Zhijiang Kang, Yu-Shu Wu","doi":"10.2118/212247-ms","DOIUrl":"https://doi.org/10.2118/212247-ms","url":null,"abstract":"\u0000 Accurate modeling of fractured reservoirs is very challenging due to the various scales of fractures. The fracture networks may be too complex to be represented using the equivalent continuum model (ECM) or dual porosity-dual permeability (DPDK) model, yet too computational costly to be modeled using the discrete fracture (DFM) or embedded discrete fracture (EDFM) models. This paper proposes a hybrid model that integrates ECM, DPDK, and an integrally embedded discrete fracture model (IEDFM) to account for multi-scale fractures. The hybrid model is applied to investigate the coupled geomechanics-fluid flow process in fractured reservoirs.\u0000 In the hybrid model, small-scale fractures are upscaled into effective matrix permeability tensor using ECM, medium-scale fractures are considered as an individual continuum using DPDK, and large-scale fractures are explicitly represented using IEDFM. The multiphase flow in effective matrix and fracture continua is modeled using the multi-point flux approximation (MPFA), and fluid exchanges between the anisotropic continua and the large-scale fracture control volumes are precisely calculated using the IEDFM. Empirical models are used to calculate the displacement of natural fractures, and analytical models are used to calculate the aperture changes of hydraulic fractures. The overall deformation of a fractured rock is described using an equivalent method. The coupled geomechanics-fluid flow system is discretized by the finite element-finite volume method (FV-FEM) and solved using the fixed-stress split iterative coupling approach.\u0000 Several examples are presented to demonstrate the applicability of the proposed method. The hybrid model is first employed to simulate water flooding process in a naturally fractured reservoir with multi-scale fractures. Effects of different scales of fractures, geomechanics coupling and capillary pressure are investigated. A case of producing from horizontal well in a hydraulic fractured tight oil reservoir is then studied, where the hydraulic fractures are modeled explicitly using IEDFM and the stimulation areas around hydraulic fractures are modeled using DPDK. Effects of stimulation area size on the pressure depletion and on the stress evolution process in the reservoir are investigated.","PeriodicalId":225811,"journal":{"name":"Day 1 Tue, March 28, 2023","volume":"96 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128119823","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Nonlinearly Constrained Life-Cycle Production Optimization Using Sequential Quadratic Programming (SQP) With Stochastic Simplex Approximated Gradients (StoSAG)","authors":"Q. Nguyen, M. Onur, F. Alpak","doi":"10.2118/212178-ms","DOIUrl":"https://doi.org/10.2118/212178-ms","url":null,"abstract":"\u0000 Life-cycle production optimization is a crucial component of closed-loop reservoir management, referring to optimizing a production-driven objective function via varying well controls during a reservoir's lifetime. When nonlinear-state constraints (e.g., field liquid production rate and field gas production rate) at each control step need to be honored, solving a large-scale production optimization problem, particularly in geological uncertainty, becomes significantly challenging. This study presents a stochastic gradient-based framework to efficiently solve a nonlinearly constrained deterministic (based on a single realization of a geological model) or a robust (based on multiple realizations of the geologic model) production optimization problem. The proposed framework is based on a novel sequential quadratic programming (SQP) method using stochastic simplex approximated gradients (StoSAG). The novelty is due to the implementation of a line-search procedure into the SQP, which we refer to as line-search sequential quadratic programming (LS-SQP). Another variant of the method, called the trust-region SQP (TR-SQP), a dual method to the LS-SQP, is also introduced. For robust optimization, we couple LS-SQP with two different constraint handling schemes; the expected value constraint scheme and minimum-maximum (min-max) constraint scheme, to avoid the explicit application of nonlinear constraints for each reservoir model. We provide the basic theoretical development that led to our proposed algorithms and demonstrate their performances in three case studies: a simple synthetic deterministic problem (a two-phase waterflooding model), a large-scale deterministic optimization problem, and a large-scale robust optimization problem, both conducted on the Brugge model. Results show that the LS-SQP and TR-SQP algorithms with StoSAG can effectively handle the nonlinear constraints in a life-cycle production optimization problem. Numerical experiments also confirm similar converged ultimate solutions for both LS-SQP and TR-SQP variants. It has been observed that TR-SQP yields shorter but more safeguarded update steps compared to LS-SQP. However, it requires slightly more objective-function evaluations. We also demonstrate the superiority of these SQP methods over the augmented Lagrangian method (ALM) in a deterministic optimization example. For robust optimization, our results show that the LS-SQP framework with any of the two different constraint handling schemes considered effectively handles the nonlinear constraints in a life-cycle robust production optimization problem. However, the expected value constraint scheme results in higher optimal NPV than the min- max constraint scheme, but at the cost of possible constraint violation for some individual geological realizations.","PeriodicalId":225811,"journal":{"name":"Day 1 Tue, March 28, 2023","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121698279","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"New Fast Simulation of 4D (x, y, z, t) CO2 EOR by Fourier Neural Operator Based Deep Learning Method","authors":"Jianqiao Liu, Hongbin Jing, Huanquan Pan","doi":"10.2118/212236-ms","DOIUrl":"https://doi.org/10.2118/212236-ms","url":null,"abstract":"\u0000 The training speed is slow for the convolutional neural network (CNN)-based physics-informed neural network (PINN) in surrogate models and it is difficult to be applied to large-scale engineering problems. The Fourier Neural Operator (FNO) network can speed up 100 times faster than the PINN according to current literature. But the current FNO only handles the 3D (x, y, t) spatial-temporal domain. In this work, we developed a new framework to simulate the 4D (x, y, z, t) subsurface flow problems using the FNO network and the domain decomposition method. After numerical simulation runs, the obtained results of subsurface flow field distributions in 4D spatial-temporal domain (x, y, z, t) are decomposed into multiple 3D spatial-temporal domains (x, y, t) in the z dimension. Then, multiple FNO networks are used to train 3D spatial-temporal domain (x, y, t) in parallel to predict the distributions of the flow field in subsequent time steps. Finally, the predicted results of the 4D spatial-temporal solution in subsequent time steps are obtained by re-coupling the trained 3D (x, y, t) results in the z dimension. In this way, our new framework successfully extends FNO-network from 3D (x, y, t) to 4D (x, y, z, t) to predict field distributions in subsurface flow. The new framework was successfully applied to some very complex cases of CO2 injection for enhanced oil recovery (EOR) in compositional simulations. The predicted accuracy is enough for the method to be applied to simulate the complex CO2 EOR in fractured systems. The computational speed in 4D (x, y, z, t) can be as fast as it does in 3D (x, y, t) through parallel training. The tested results show that our new framework can efficiently simulate the EOR processes by injecting CO2 into complex fracture reservoirs. For the first time, we developed a new methodology that successfully extends the current FNO network from 3D (x, y, t) to 4D (x, y, z, t). Our framework paves way for the fast FNO network to solve the large-scale spatial-temporal domain of reservoir engineering systems.","PeriodicalId":225811,"journal":{"name":"Day 1 Tue, March 28, 2023","volume":"111 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133280913","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Using Enriched Galerkin as an Energy and Mass Conservative Scheme for Simulating Thermoporoelasticity Problems","authors":"A. G. Almetwally, R. Podgorney, M. Wheeler","doi":"10.2118/212240-ms","DOIUrl":"https://doi.org/10.2118/212240-ms","url":null,"abstract":"\u0000 Accurate simulation of the thermoporoelasticity problems is beneficial for the exploitation activities of aquifers, geothermal, and hydrocarbon reservoirs. Simulating such problems using a finite-element Continuous Galerkin scheme (CG) lacks local energy/mass conservation. Despite being a conservative scheme, Discontinuous Galerkin (DG) is computationally expensive with much higher degrees of freedom (DoFs). This paper presents the Enriched Galerkin scheme (EG) implementation for thermoporoelasticity problems to ensure local energy/mass conservation with fewer DoFs.","PeriodicalId":225811,"journal":{"name":"Day 1 Tue, March 28, 2023","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126949763","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"High Performance Computing and Speedup Techniques in Geochemical Modeling of Matrix Acidizing","authors":"Wan Wei, A. Sanaei, Fabio Bordeaux Rego, K. Sepehrnoori","doi":"10.2118/212165-ms","DOIUrl":"https://doi.org/10.2118/212165-ms","url":null,"abstract":"\u0000 Matrix acidizing is a stimulation treatment during which acid is injected below formation fracture pressure. The purpose of acidizing is to enlarge pore space or create channels through dissolution of plugging particles and formation minerals near the wellbore. Simulation of acidizing process is computationally expensive, especially for geochemical simulation which considers full-species transport and complex reactions. In this paper, geochemical modeling of acidizing process is implemented through coupling two simulation models. One is UTCOMP (a 3D reservoir simulator) which is responsible for calculations of fluid flow and solute transport. The other is IPhreeqc (a geochemical package) which is responsible for calculations of kinetic and equilibrium reactions among minerals and aqueous species. Acidizing simulation through the coupled model UTCOMP-IPhreeqc is computationally expensive, and geochemical calculations through IPhreeqc are the computational bottleneck. To improve the computational efficiency, geochemical calculations which take up the majority of the computational time are parallelized. And speedup techniques are implemented to reduce the number of IPhreeqc calls through monitoring the amount change of geochemical components. We have validated the coupled model UTCOMP-IPhreeqc through comparison with the analytical solution in previous work. Parallel performance is measured by comparing total CPU time, CPU time spent on geochemical calculations, and speedup ratios among simulation runs using different processor numbers. For heterogeneous matrix, different dissolution patterns are generated under different injection rates, and the computational time varies depending on the total injection time and the average time step size. For different dissolution patterns, the overall speedup ratio is up to 6.69 when using 16 processors, reducing 85% of CPU time compared with the case using a single processor. The speedup ratio for geochemical calculations is up to 14.21 when using 16 processors, saving 93% of CPU time compared with the case using a single processor. Besides parallel computing, the speedup techniques also improve the computational efficiency, and obtain optimal performance for wormhole dissolution patterns in which most of the geochemical reactions occur in a localized volume. The computational time is reduced to 49% maintaining 96% accuracy compared with the case without using speedup techniques. The coupled model UTCOMP-IPhreeqc has the modeling ability of full-species transport and complex reactions. On this basis, the presented model significantly improves the computational efficiency of UTCOMP-IPhreeqc through parallel computing and speedup techniques reducing the computational time of geochemical calculations.","PeriodicalId":225811,"journal":{"name":"Day 1 Tue, March 28, 2023","volume":"47 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122798337","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Novel Machine-Learning Assisted Phase-Equilibrium Calculation Model for Liquid-Rich Shale Reservoirs","authors":"Fangxuan Chen, Sheng Luo, S. Wang, H. Nasrabadi","doi":"10.2118/212193-ms","DOIUrl":"https://doi.org/10.2118/212193-ms","url":null,"abstract":"\u0000 In composition reservoir simulation, fluid phase behavior is determined by vapor-liquid equilibrium (VLE) calculations. VLE calculations can consume more than half of the CPU time of compositional reservoir simulations. To accelerate the VLE calculations, machine learning (ML) technique is introduced. In this work, we developed a novel ML-assisted VLE calculation model for shale reservoirs. Our model has two main innovations compared with previous ML-assisted VLE calculation models. Firstly, the extended Peng-Robinson equation of states (PR-C EOS) is incorporated for VLE calculation. Previous models used the conventional Peng-Robinson equation of states (PR EOS), which becomes inaccurate when the pore diameter reduces to the scale of nanometers. With PR-C EOS, fluid characteristics can be accurately modeled under nano-scale conditions, making our model applicable to shale reservoirs. Secondly, in our model, a general set of pseudo components is selected to cover different fluid types. Previous models are designed for a specific type of hydrocarbon mixture. There are two parts to our model: stability analysis and flash calculation. In the stability analysis, the multi-layer perceptron (MLP) is trained to predict whether the fluid is in single-phase or two-phase condition. The equilibrium ratios are estimated using a physics-informed neural network (PINN) in the flash calculation. The application of ML techniques accelerates the CPU time by two orders of magnitude without losing too much accuracy. This work provides the framework of incorporating ML into VLE calculation and develops a ML-assisted VLE calculation model that is suitable for various hydrocarbon mixtures in shale reservoirs.","PeriodicalId":225811,"journal":{"name":"Day 1 Tue, March 28, 2023","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129162025","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}