Agenda
Monday, March 27
14:00 - 17:00
Tuesday, March 28
07:00 - 17:00
08:00 - 10:00
01
Advances in Algorithms and Implementations for High-Performance Computing
Grand Ballroom A
Session Chairpersons
Abdulrahman Mohammad Manea - Saudi Aramco PE&D, Rami Younis - University of Tulsa
Novel algorithms and careful adaptations of existing ones are necessary to exploit the full potential of evolving high-performance computing (HPC) architectures. This session combines a range of innovations in implementation methods, spanning generalized approaches for heterogeneous HPC architectures to specialized implementations for GPU architectures. Additionally, the session presents algorithmic advancements spanning compositional flash computations to numerical solution methods.
Time | Paper # | Presentation |
---|---|---|
0800-0830 |
212205
![]() |
Implementing a Hardware Agnostic Commercial Black-oil Reservoir Simulator |
0830-0900 |
212248
![]() |
An Adaptive Coloring Scheme for Graphics Processing Unit Preconditioners |
0900-0930 |
212198
![]() |
A GPU-accelerated Simulator for Challenging Extreme-Scale Geomechanical Models |
0930-1000 |
212183
![]() |
Graphics Processing Unit Performance Scalability Study on a Commercial Black-oil Reservoir Simulator |
08:00 - 10:00
02
Model-Based and Data-Driven Optimization
Grand Ballroom B
Session Chairpersons
Eduardo Gildin - Texas A&M University, Mary Fanett Wheeler - The University of Texas At Austin
Forecasting hydrocarbons production is paramount for steering business strategies and decisions to maximize economic gains but, at the same time, minimize risks associated with environmental and socio-economic issues that may arise during the life-cycle of a reservoir. Numerical optimization has been used to develop decision making strategies associated with unknown geological media, geopolitical and economical variability experienced during much longer periods of time. This session showcases novel algorithms using classical optimization methods and newly developed artificial intelligence frameworks for production optimization and efficient model calibration under uncertain environments.
Time | Paper # | Presentation |
---|---|---|
0800-0830 |
212178
![]() |
Nonlinearly Constrained Life-Cycle Production Optimization Using Sequential Quadratic Programming (SQP) with Stochastic Simplex Approximated Gradients (StoSAG) |
0830-0900 |
212237
![]() |
Practical Closed-Loop Reservoir Management Using Deep Reinforcement Learning |
0900-0930 |
212207
![]() |
Efficient Adaptation and Calibration of Adjoint-based Reduced-order Coarse-grid Network Models |
0930-1000 |
212188
![]() |
A Fast History Matching and Optimization Tool and its Application to a Full Field with More Than 1000 Wells |
Alternate |
212212
![]() |
A Quasi-newton Method For Well Location Optimization Under Uncertainty |
10:00 - 10:30
10:30 - 12:15
P01
Opening Plenary Session: What is the Future of Reservoir Simulation and Modeling in the Evolving Energy Landscape?
Grand Ballroom B
Moderator(s)
Sebastien Matringe - Hess Corporation
Speaker(s)
James M Hacker - ExxonMobil, Jack Hunter Norbeck - Stanford University, Moon Chaudhri - Chevron, Shashi Menon - SLB, Ola Terjeson Miljeteig - Equinor
The energy industry is undergoing a transition, where the ability to provide reliable resources for the world economy combines with the target of achieving net zero emissions within few decades.
The key objective of oil & gas asset developments, from unconventional to enhanced oil recovery, will be to create the capability to minimize greenhouse gas emissions. At the same time, CO2 sequestration will become a new critical activity for many, if not most, of the oil and gas companies, and many of them are already largely investing significant resources in pilot and demonstration projects. On the other hand, technologies matured in the oil & gas industry may be effectively used to scale up geothermal energy production and storage.
The panel discussion will address the expected transformation of our industry with the ambition of highlighting the demand that reservoir simulation and computational modeling science should meet to provide tools, solutions, and competencies suitable for the ongoing energy industry evolution.
12:15 - 13:30
PL01
Tuesday Poster Luncheon
Grand Ballroom C
Poster authors will present, Tuesday and Wednesday, at 12:45pm during the luncheon.
Time | Paper # | Presentation |
---|---|---|
1245-1330 |
212166
![]() |
A Mineral-Composition Dependent Fracture Numerical Model of Thermally Treated Shale Gas Reservoirs |
1245-1330 |
212211
![]() |
Use of Clustering Techniques for Automated Lumping of Components in Compositional Models |
1245-1330 |
212185
![]() |
Parameter Inversion in Geothermal Reservoir Using Markov Chain Monte Carlo and Deep Learning |
1245-1330 |
212165
![]() |
High Performance Computing and Speedup Techniques in Geochemical Modeling of Matrix Acidizing |
1245-1330 |
212163
![]() |
Troll Reservoir Simulation Development, From Well Long-Term Tests to Full FMU Simulations |
1245-1330 |
212236
![]() |
New Fast Simulation Of 4d (x, Y, Z, T) Co2 Eor By Fourier Neural Operator Based Deep Learning Method |
1245-1330 |
212212
![]() |
A Quasi-newton Method For Well Location Optimization Under Uncertainty |
1245-1330 |
212164
![]() |
The Hybrid-dimensional Darcy's Law: A Non-conforming Reinterpreted Discrete Fracture Model (RDFM) for the Compressible Miscible Displacement and Multicomponent Gas Flow in Fractured Media |
1245-1330 |
212174
![]() |
Formation Fracturing by High-energy Impulsive Mechanical Loading |
1245-1330 |
212253
![]() |
Water Injection Optimization to Prevent Reservoir Souring |
1245-1330 |
212189
![]() |
Modeling Fault Reactivation Triggered by Fluid Injection |
1245-1330 |
212252
![]() |
Matrix Formulation for Simultaneous Calculations of Pressure and Temperature in Wells and Pipelines |
1245-1330 |
212184
![]() |
Implementation of Adaptive Localization for Enhancing Ensemble-Based History Matching in Hydrocarbon Reservoir Management |
1245-1330 | 212167 | Deep Learning-based Multiresolution Parameterization for Spatially Adaptive Model Updating |
13:30 - 15:30
03
Discretization Methods for Flow and Geomechanics
Grand Ballroom A
Session Chairpersons
Yu-Shu Wu - Colorado School of Mines, Mohammad Karimi-fard - Stanford University
This session focuses on discretization techniques for flow and transport as well as thermo-poroelasticity problems. In particular, different types of finite-volume and finite-element approaches will be discussed.
Time | Paper # | Presentation |
---|---|---|
1330-1400 |
212213
![]() |
Consistent Discretization Methods for Reservoir Simulation on Cut-cell Grids |
1400-1430 |
212238
![]() |
An Enriched Galerkin Discretization Scheme for Two Phase Flow on Non-orthogonal Grids |
1430-1500 |
212206
![]() |
Pressure Jump Stabilization for Compositional Poromechanics on Unstructured Meshes |
1500-1530 |
212240
![]() |
Using Enriched Galerkin as an Energy and Mass Conservative Scheme for Simulating Thermoporoelasticity Problems |
Alternate |
212164
![]() |
The Hybrid-dimensional Darcy's Law: A Non-conforming Reinterpreted Discrete Fracture Model (RDFM) for the Compressible Miscible Displacement and Multicomponent Gas Flow in Fractured Media |
13:30 - 15:30
04
Physics Informed Neural Network
Grand Ballroom B
Session Chairpersons
Jincong He - Chevron Corporation, Eduardo Gildin - Texas A&M University
Neural networks have emerged as powerful proxies for modeling complex input-output relationships for reservoir management problems. Recent works on infusing physical principles into the training process have shown potentials to further improve the performance of neural networks. This session showcases the application of these physics-informed neural networks in various problems including production forecast, uncertainty quantification, and phase behavior modeling.
Time | Paper # | Presentation |
---|---|---|
1330-1400 | 212202 | A Physics-informed Neural Network for Temporospatial Prediction of Hydraulic-geomechanical Processes |
1400-1430 | 212201 | Reservoir Connectivity Identification and Robust Production Forecasting Using Physics Informed Machine Learning |
1430-1500 |
212255
![]() |
Uncertainty Quantification For Transport In Porous Media Using Parameterized Physics Informed Neural Networks |
1500-1530 |
212187
![]() |
A Data-driven Deep Learning Framework for Microbial Reaction Prediction for Hydrogen Underground Storage |
15:30 - 16:00
16:00 - 18:00
05
Fractures, Faults and Induced Seismicity
Grand Ballroom A
Session Chairpersons
Joshua Alexander White - Lawrence Livermore National Laboratory, Adolfo Antonio Rodriguez - OpenSim Technology
Understanding fluid flow and geomechanical behavior in fractured and faulted formations is critical to safe and efficient reservoir operations. This session will explore recent advances in multi-physics modeling of these complex systems.
Time | Paper # | Presentation |
---|---|---|
1600-1630 |
212171
![]() |
An Integrated Modeling Framework for Simulating Complex Transient Flow in Fractured Reservoirs with 3D High-quality Grids |
1630-1700 |
212170
![]() |
Resolving Ambiguity in 2008-2015 Irving-Dallas Seismicity by Coupling Geomechanical Models at Fort Worth Basin and Barnett Reservoir Scales |
1700-1730 |
212247
![]() |
Coupled Geomechanics and Fluid Flow Modeling for Petroleum Reservoirs Accounting for Multi-Scale Fractures |
1730-1800 |
212234
![]() |
Role of Inelasticity in Production-induced Subsidence and Fault Reactivation in the Groningen Field |
Alternate |
212174
![]() |
Formation Fracturing by High-energy Impulsive Mechanical Loading |
16:00 - 18:00
06
Data Analytics and Machine Learning
Grand Ballroom B
Session Chairpersons
Hector Manuel Klie - DeepCast, Eduardo Gildin - Texas A&M University
Data Analytics and Machine Learning, as a subject, has driven a paradigm shift and significant transformations of the oil and gas industry in recent years. More efficient algorithms, novel reduced order proxy models, have been deployed to gain efficiency, improve production forecasts, and maximize recoveries. These new efficient algorithms alleviate computational burden of conventional reservoir simulations by running more scalable and faster models while preserving accuracy. This session explores new approaches to physic-based and data driven models developed to aim at reducing computational costs, improving accuracy of prediction while preserving physical constraints, conservation laws ad constitutive relations with data.
Time | Paper # | Presentation |
---|---|---|
1600-1630 | 212193 | A Novel Machine-Learning Assisted Phase-Equilibrium Calculation Model for Liquid-Rich Shale Reservoirs |
1630-1700 | 212217 | Application of Deep Neural Networks to the Operator Space of Nonlinear Partial Differential Equations for Physics-Based Proxy Modelling |
1700-1730 |
212204
![]() |
Guided Deep Learning Manifold Linearization of Porous Media Flow Equations |
1730-1800 | 212230 | Data-Space Inversion for Rapid Physics-Informed Direct Forecasting in Unconventional Reservoirs |
18:00 - 19:00
Wednesday, March 29
07:30 - 17:00
08:30 - 12:00
07
Multiscale Methods, Mesh Adaptivity and Upscaling
Grand Ballroom A
Session Chairpersons
Vegard Kippe - Equinor ASA, Knut-Andreas Lie - SINTEF
This session explores the most recent advances in the commercial implementation of multiscale methods, which have been widely researched over the past two decades. You will also learn about new methods for (dynamic) mesh adaptivity and for upscaling highly detailed representations of reservoir heterogeneity.
Time | Paper # | Presentation |
---|---|---|
0830-0900 |
212229
![]() |
Strongly Coupled Prolongation in Multiscale Pressure Solver for High-Contrast Heterogeneous Reservoir Simulation |
0900-0930 |
212244
![]() |
Bridging Computational Stratigraphy and Reservoir Simulation for Geologically Realistic High-resolution Reservoir Modeling |
0930-1000 |
212239
![]() |
Dynamic Mesh Adaptivity and Novel Stopping Criterion Guided by a Posteriori Error Estimates for Coupled Geomechanics Using Mixed Finite Element Method for Flow |
1030-1100 |
212194
![]() |
An Adaptive Grid Refinement Method for Flow-based Embedded Discrete Fracture Models |
1100-1130 |
212231
![]() |
Multiscale Reservoir Simulation of High-Resolution Models |
1130-1200 |
212215
![]() |
Two-Step Upscaling of Sub-Seismic Geo-Heterogeneity with Flow-Rate- and Direction Dependent Saturation Functions |
08:30 - 12:00
08
Integrated Reservoir Modeling and Case Studies
Grand Ballroom B
Session Chairpersons
Santosh Kumar Verma - ExxonMobil Upstream Research Co., Kyrre Bratvedt - SLB
Reservoir simulation often needs to be tailored to specific applications to better represent the underlying physical processes. Papers in this session provide case studies of integrating reservoir simulation with wellbore models, CO2 flooding, gridding methods, as well as insight into the development of complex simulation systems.
Time | Paper # | Presentation |
---|---|---|
0830-0900 |
212200
![]() |
Accurate Production Forecast and Productivity Decline Analysis Using Coupled Full-field and Near-wellbore Poromechanics Modeling |
0900-0930 |
212246
![]() |
High Performance Integrated Asset Modeling: A Giant Gas Field Case Study |
0930-1000 |
212224
![]() |
Integrating Geomechanics Studies To Shale And Tight Phased Development - Application To Delaware Basin |
1030-1100 |
212226
![]() |
Integrating Pipe Fractional Flow Theory with Fully Compositional Wellbore Models |
1100-1130 |
212259
![]() |
Use of "Look-Ahead" Reservoir Models to Optimize Reservoir Performance |
1130-1200 |
212173
![]() |
Successful Application of an Ensemble Modeling Workflow for a Deepwater Field in US Gulf of Mexico |
Alternate |
212253
![]() |
Water Injection Optimization to Prevent Reservoir Souring |
Alternate |
212189
![]() |
Modeling Fault Reactivation Triggered by Fluid Injection |
10:00 - 10:30
12:00 - 13:30
PL02
Wednesday Poster Luncheon
Grand Ballroom C
Poster authors will present, Tuesday and Wednesday, at 12:45pm during the luncheon.
Time | Paper # | Presentation |
---|---|---|
1245-1330 |
212166
![]() |
A Mineral-Composition Dependent Fracture Numerical Model of Thermally Treated Shale Gas Reservoirs |
1245-1330 |
212211
![]() |
Use of Clustering Techniques for Automated Lumping of Components in Compositional Models |
1245-1330 |
212185
![]() |
Parameter Inversion in Geothermal Reservoir Using Markov Chain Monte Carlo and Deep Learning |
1245-1330 |
212165
![]() |
High Performance Computing and Speedup Techniques in Geochemical Modeling of Matrix Acidizing |
1245-1330 |
212163
![]() |
Troll Reservoir Simulation Development, From Well Long-Term Tests to Full FMU Simulations |
1245-1330 |
212236
![]() |
New Fast Simulation Of 4d (x, Y, Z, T) Co2 Eor By Fourier Neural Operator Based Deep Learning Method |
1245-1330 |
212212
![]() |
A Quasi-newton Method For Well Location Optimization Under Uncertainty |
1245-1330 |
212174
![]() |
Formation Fracturing by High-energy Impulsive Mechanical Loading |
1245-1330 |
212253
![]() |
Water Injection Optimization to Prevent Reservoir Souring |
1245-1330 |
212189
![]() |
Modeling Fault Reactivation Triggered by Fluid Injection |
1245-1330 |
212252
![]() |
Matrix Formulation for Simultaneous Calculations of Pressure and Temperature in Wells and Pipelines |
1245-1330 |
212184
![]() |
Implementation of Adaptive Localization for Enhancing Ensemble-Based History Matching in Hydrocarbon Reservoir Management |
1245-1330 | 212167 | Deep Learning-based Multiresolution Parameterization for Spatially Adaptive Model Updating |
13:30 - 17:00
09
Linear and Nonlinear Solvers
Grand Ballroom A
Session Chairpersons
Denis Viktorovich Voskov - Delft University of Technology, Leonardo Patacchini - Stone Ridge Technology
As the resolution and complexity of reservoir models keep inexorably increasing, so does the need for efficient and scalable solvers, able to efficiently track the dynamics entailed by multiple physics at different scales. This session will feature novel algorithmic developments in general purpose and specialized linear solvers, nonlinear formulations and nonlinear solvers.
Time | Paper # | Presentation |
---|---|---|
1330-1400 |
212172
![]() |
Constrained Pressure Residual Preconditioner Including Wells for Reservoir Simulation |
1400-1430 |
212176
![]() |
A Global-convergent Newton Optimization Algorithm for the Phase Behavior Calculations with Capillary Pressure Effect for Tight Reservoir Fluids |
1430-1500 |
212261
![]() |
Smooth Formulation for Three-phase Black-oil Simulation with Superior Nonlinear Convergence |
1530-1600 |
212199
![]() |
An Adaptive Newton-ASPEN Solver for Complex Reservoir Models |
1600-1630 |
212219
![]() |
Dissipation-based Nonlinear Solver for Efficient Implicit Simulation of Compositional and Discrete Fracture Models |
1630-1700 |
212175
![]() |
Learning to Solve Parameterized Single-cell Problems Offline to Expedite Reservoir Simulation |
Alternate |
212252
![]() |
Matrix Formulation for Simultaneous Calculations of Pressure and Temperature in Wells and Pipelines |
13:30 - 17:00
10
C02 Storage and Enhanced Geothermal System Modeling
Grand Ballroom B
Session Chairpersons
Gilles Darche - TotalEnergies SE, Vijay Kumar Shrivastava - Computer Modelling Group Ltd.
The session will focus on simulation of CO2 storage, and enhanced geothermal system modeling. The thermal, hydrodynamic, and geomechanical effects associated with these processes will be presented including a well placement strategy for optimizing and controlling geological CO2 storage.
Time | Paper # | Presentation |
---|---|---|
1330-1400 |
212241
![]() |
A Sequentially Coupled THM Model for Fractured Enhanced Geothermal Systems Using XFEM And Hybrid EDFM And MINC Models |
1400-1430 |
212251
![]() |
Phase-field Simulation of Near-wellbore Nucleation and Propagation of Hydraulic Fractures in Enhanced Geothermal Systems (EGS) |
1430-1500 |
212228
![]() |
Integrated Framework for Optimization of Horizontal/Deviated Well Placement and Control for Geological CO2 Storage |
1530-1600 |
212254
![]() |
A New Phase-labeling Method Based on Machine Learning for CO2 Applications |
1600-1630 |
212182
![]() |
A Unified Thermal-Reactive Compositional Simulation Framework for Modeling CO2 Sequestration at Various Scales |
1630-1700 |
212235
![]() |
Coupled CO2 Injection Well Flow Model to Assess Thermal Stresses Under Geomechanical Uncertainty |
15:00 - 15:30
17:15 - 18:45
P02
The 11th SPE Comparative Solution Project: A CO2 Storage Inspired Comparative Solution Project
Grand Ballroom B
Moderator(s)
Knut-Andreas Lie - SINTEF
Speaker(s)
Anthony Robert Kovscek - Stanford University, Jan M. Nordbotten - University of Bergen
This session is the announcement of, and call for participation in, the 11th SPE comparative solution project (CSP). The 11th SPE CSP It is motivated on the simulation challenges associated with CO2 storage operations in geological settings of realistic complexity. The CSP contains three versions: Version 11A is a 2D geometry at the laboratory scale, inspired by a recent CO2 storage forecasting and validation study. For Version 11B, the 2D geometry and operational conditions from 11A are rescaled to field conditions characteristic of the Norwegian Continental Shelf. Finally, for Version 11C, the geometry of version 11B is extruded to a full 3D field model. The CSP has a two-year timeline, being launched at the 2023 SPE Reservoir Simulation Conference, and culminating at the 2025 SPE Reservoir Simulation Conference.
19:00 - 20:00
Thursday, March 30
07:30 - 10:00
08:30 - 12:00
11
Complex Physics
Grand Ballroom A
Session Chairpersons
Alberto Cominelli - Eni E&P, Faruk Omer Alpak - Shell International E&P Co.
Reservoir simulation technology has been expanding its capability of rigorously solving increasingly complex multi-physics problems posed by developments on hydrocarbon recovery as well as energy transition themes. Advancements on thermal/compositional simulation, reactive transport including novel ways of accounting for geochemical effects, and multicontinuum transport modeling will be focus areas for the presentations in this session. Algorithmic developments, effective workflows, and results of state-of-the-art studies with applications to hydrocarbon recovery as well as subsurface storage will be shared by participants from leading industrial and academic institutions.
Time | Paper # | Presentation |
---|---|---|
0830-0900 |
212257
![]() |
Salt Precipitation and Water Evaporation Modelling in a Black-Oil Reservoir Simulator |
0900-0930 |
212233
![]() |
Experimental and Simulation Studies of Cryogenic Effects in the Near-wellbore Region |
0930-1000 |
212208
![]() |
Advanced Modeling of Diffusion and Convection in Multiphase Compositional Simulation for Tight Porous Media |
1030-1100 |
212221
![]() |
A Numerical Study of Selective Inorganic Precipitation in Fractures to Geoengineer Resilient and Secure Underground Storage Sites |
1100-1130 |
212222
![]() |
Investigating Acidizing in Carbonate Reservoirs: Global Sensitivity Analysis |
1130-1200 |
212218
![]() |
Molecular Simulation of Competitive Adsorption of Hydrogen and Methane: Analysis of Hydrogen Storage Feasibility in Depleted Shale Gas Reservoirs |
08:30 - 12:00
12
Data Assimilation
Grand Ballroom B
Session Chairpersons
Pallav Sarma - Tachyus, Hector Manuel Klie - DeepCast
Data assimilation is an approach that seeks to effectively combine observations and model outputs with the purpose of improving the quality of predictions. In reservoir simulation, data assimilation encompasses a series of key workflows that include history matching, state estimation and uncertainty quantification. In this session, we’ll discuss the latest developments and applications of data assimilation on a variety of use cases that stress the conception of novel capabilities. Topics to be reviewed include extraction of new features embedded in the data, efficient handling of hyper-parameters via deep learning, acceleration of gradient-based calculations and, estimations involving multiple objectives.
Time | Paper # | Presentation |
---|---|---|
0830-0900 |
212190
![]() |
Stein Variational Gradient Descent for Reservoir History Matching Problems |
0900-0930 |
212196
![]() |
Feature Extraction in Time-lapse Seismic Using Deep Learning for Data Assimilation |
0930-1000 |
212232
![]() |
On the Formulation of the Ensemble History-Matching Problem |
1030-1100 |
212242
![]() |
A Practical Approach to Select Representative Deterministic Models Using Multi-Objective Optimization from an Integrated Uncertainty Quantification Workflow |
1100-1130 |
212169
![]() |
Efficient Inverse Modeling Framework for Energy Transition Applications Using Operator-Based Linearization and Adjoint Gradients |
1130-1200 |
212180
![]() |
Performance Benchmarking of Different Methods to Solve Gauss-Newton Trust Region Subproblems |
Alternate |
212184
![]() |
Implementation of Adaptive Localization for Enhancing Ensemble-Based History Matching in Hydrocarbon Reservoir Management |
Alternate | 212167 | Deep Learning-based Multiresolution Parameterization for Spatially Adaptive Model Updating |
10:00 - 10:30