Parallelization on Unstructured Grids

Usadi, Adam K.; Mishev, Ilya D.; Shaw, Jason D.; Wiegand, Klauss

doi:10.2523/106063-ms

Cited by 5 publications

(2 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Some oil companies developed their own internal unstructured simulators tailored to their needs (Khan & Dawson, 2004;Usadi et al, 2007). Only within the last decade, with the development of the latest generation of simulators, have unstructured grids became more widely available for reservoir flow simulation.…”

Section: Motivationmentioning

confidence: 99%

“…Empower, the ExxonMobil flow simulator capable of ustructured grid simulation, adopts a global flow based method for upscaling (Khan & Dawson, 2004;Usadi et al, 2007). Single phase simulations are performed on the so called computational gridan unstructured grid with scale close to the refined structured gridfor several boundary conditions.…”

Section: Permeability and Transmissibility Upscaling On Unstructured mentioning

confidence: 99%

See 1 more Smart Citation

Gridding and Scaling Strategies for Unstructured Reservoir Flow Simulation

MODENESI¹

View full text Add to dashboard Cite

I would first like to thank my advisor Ivan Menezes, for the opportunity of working in this project. His professionalism and enthusiasm were unfaltering for the whole period. I am also grateful to my co-advisor Leonardo Duarte, who developed the reservoir simulator used for this research. His support with several issues was indispensable for the developed code to work. I am grateful to Pontifical Catholic University of Rio de Janeiro-PUC-Rio, for giving me the opportunity of enrolling in this master's program. In addition, this study was financed in part by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior-Brasil (CAPES)-Finance Code 001, to whom I also thank. My sincere thanks to the company for which I work, Petrobras, for conceding time for me to dedicate to this master's program. The advice and guidance offered by several of my colleagues, in particular Daniel Miranda and Marcos Barbosa, was essential to help to shape this dissertation. Finally, I'd like to acknowledge my wife and parents for their support and inspiration while working on this dissertation.

show abstract

Section: Motivationmentioning

confidence: 99%

Section: Permeability and Transmissibility Upscaling On Unstructured mentioning

confidence: 99%

Gridding and Scaling Strategies for Unstructured Reservoir Flow Simulation

MODENESI¹

View full text Add to dashboard Cite

show abstract

A General Unstructured Grid, EOS-Based, Fully-Implicit Thermal Simulator for Complex Reservoir Processes

Subramanian

Dratler

et al. 2007

SPE Reservoir Simulation Symposium

View full text Add to dashboard Cite

This paper describes a general unstructured grid, EOS-based, fully-implicit thermal simulator for complex reservoir processes. Under the unstructured grid framework, the simulator uses Newton's method to solve component material balance equations, energy balance equation and volume balance equation for component moles, energy and pressure, where chemical reactions and/or external heat sources/sinks are treated in source terms. Because of the similarity among component material balance equations and the energy balance equation, energy is treated as a "component" to achieve a uniform formulation with common code for all simulations (black oil, compositional and thermal). The thermal simulator was validated using analytical models as well as other thermal simulators. Application of the thermal simulator includes studies of grid-orientation problems and to design and optimise Cold Lake heavy oil development. Introduction Modern reservoir management requires a simulator to accurately represent reservoir details using fine-scale geologic features, complex well paths, and modeling of large-scale interactions between multiple fields. Unstructured gridding makes it possible to capture and honor more geologic and engineering detail in reservoir simulation models with greater exactness than Cartesian-based reservoir grids. However, industry has been generally reluctant to apply this capability to practical reservoir simulation due in part to concerns about potential loss in computational efficiency. Many papers have been published under Cartesian-based framework1–5. Few papers are available to address reservoir simulation issues under general unstructured grid framework6–11. In 2001, Beckner, et al.6 presented ExxonMobil's new unstructured grid reservoir simulation system, which discussed field examples involving complex geologic features (e.g. non-vertical faults and stratigraphic pinchouts) and multiple reservoirs connected to a common production infrastructure. It was reported that the simulator significantly reduces simulation cycle-time through ease-of-use and integration with geologic models. One example indicated that about 100,000 blocks can represent the equivalent of 1.6 million rectangular grid blocks for achieving accurate geologic features. One of the most important aspects for unstructured reservoir simulator development is the linear solver for solving large linear system created by an unstructured grid which cannot be efficiently solved by conventional solution methods normally applied to simulators using rectangular gridding. Beckner, et al.8 in 2006 reported their collaborative effort on developing an unstructured linear solver library called SparSol, which includes a numerical library of scaling and reordering methods, preconditioners and iterative methods. Results show that SparSol is faster than several popular, freely-available packages for the set of matrices tested. Usadi, et al.9 reported their experiences with parallelizing an unstructured reservoir simulator on SMP machines. Included in their paper are how parallelization is performed at a high level through several variants of data partitioning adapted to the specific algorithmic needs, that solver convergence rate can be strongly dependent on simulator determined data partitioning, and that well management performance can be strongly dependent on the way reservoir engineers have applied their constraints and field production analysis. The next few sections of this paper present an overview of the simulation equations, solution procedures, and some discussions. Examples are given to illustrate simulator results, to demonstrate how the simulator can reduce grid-orientation problems and how the equation of state (EOS) method can be used in thermal simulation. The final section contains the conclusions resulting from this work.

show abstract

A General Unstructured-Grid, Equation-of-State-Based, Fully Implicit Thermal Simulator for Complex Reservoir Processes

Subramanian

Dratler

et al. 2009

SPE Journal

View full text Add to dashboard Cite

Summary This paper describes a general unstructured-grid, equation-of-state- (EOS) based, fully implicit thermal simulator for complex reservoir processes. Under the unstructured grid framework, the simulator uses Newton's method to solve component material-balance equations, energy-balance equation, and volume-balance equation for component moles, energy, and pressure, where chemical reactions and external heat sources/sinks are treated in source terms. Because of the similarity among component material-balance equations and the energy-balance equation, energy is treated as a "component" to achieve a uniform formulation with common code for all simulations (black-oil, compositional, and thermal). The thermal simulator was validated using analytical models and other thermal simulators. The thermal simulator is used to study grid-orientation problems and to design and optimize Cold Lake heavy-oil development. Introduction Modern reservoir management requires a simulator to represent reservoir details accurately using fine-scale geologic features, complex well paths, and modeling of large-scale interactions among multiple fields. Unstructured gridding makes it possible to capture and honor more geologic and engineering detail in reservoir-simulation models with greater exactness than Cartesian-based reservoir grids. However, industry generally has been reluctant to apply this capability to practical reservoir simulation partly because of concerns about potential loss in computational efficiency. Many papers have been published under Cartesian-based framework (Mifflin et al. 1991; Watts 1986; Coats 1980; Watts et al. 2005). Few papers are available to address reservoir simulation issues under general unstructured-grid framework (Naccache 1997; Beckner et al. 2001, 2006; Heinemann et al. 1991; Usadi et al. 2007; Karypis and Kumar 1998).

show abstract

Parallelization on Unstructured Grids

Cited by 5 publications

References 14 publications

Gridding and Scaling Strategies for Unstructured Reservoir Flow Simulation

Gridding and Scaling Strategies for Unstructured Reservoir Flow Simulation

A General Unstructured Grid, EOS-Based, Fully-Implicit Thermal Simulator for Complex Reservoir Processes

A General Unstructured-Grid, Equation-of-State-Based, Fully Implicit Thermal Simulator for Complex Reservoir Processes

Contact Info

Product

Resources

About