Parallel, Multigrid Finite Element Simulator for Fractured/Faulted and Other Complex Reservoirs Based on CCA
DEFC2604NT15531
Project Goals
The goals of this project were to develop blackoil, thermal, and compositional reservoir simulators for complex fractured, faulted reservoirs. Development of new well models for representing complicated wells and parallel implementation were the other project objectives.
Performer
University of Utah, Salt Lake City, UT
Benefits
The main goal of this project was to overcome difficulties associated with the representation and simulation of complex fractured/faulted systems with complicated wells. Significant advances were made in achieving this goal with the development of modular simulators that are able to access the most modern computational algorithms. Having a tool to better manage fractured/faulted reservoirs and reservoirs with complex wells will lead to improved oil and natural gas production from these reservoirs.
Background
Finitedifference methods, most commonly used in the current stateoftheart reservoir simulators, are not suitable for representing reservoirs with complex geometry, particularly faulted and fractured systems, and for modeling complex well geometries. Threedimensional, threephase black oil finiteelement simulators developed at the University of Utah provide true unstructured threedimensional gridding to model complex domains. Modular standardization, parallel protocols, new well models, and new thermal and compositional simulation possibilities implemented in this project have provided enhanced capability and functionality to model complex reservoir systems.
Accomplishments
A simulation framework was finalized, wherein the “physical models” (blackoil, thermal, and compositional) and the “discretization methods” (controlvolume finiteelement, mixed finiteelement) were completely decoupled. Three distinct discretization methods—the finite difference (FD), the controlvolume finiteelement (CVFE), and the mixed finiteelement (MFE), with their own advantages—were created to study complex domains with discrete fractures and faults. A verification hierarchy was created to ensure that the solutions from the models were accurate. One of the outcomes was that a singlephase analytical model for hydraulic fractures was used to validate the controlvolume finiteelement simulation results. A methodology to grid complex, threedimensional objects with faults and fractures was identified. New thermal and compositional modules operating with FD, CVFE, and MFE were developed. Thermal simulations of steamassisted gravity drainage were performed. New well models were developed for complex wells. Parallel simulation protocols for unstructured grids were created.
Benchmarking studies were performed on basement reservoirs, where it was shown that the performance of simulators developed agreed well with results from other simulators such as Eclipse™. Intuitive well models of complicated wells for the mixed finiteelement simulator were developed. A thermal simulation model formulation was completed and implemented for hot water flooding. The simulator structure was modified to use not only the conventional reservoir simulation boundary conditions (bottomhole pressure and rate constraints) but also constant flux boundary conditions and Dirichlet (constant pressure) specifications.
The project has made it possible to simulate complicated domains with fractures and faults and systems with complex wells such multilaterals.
Major accomplishments were:
 Decoupling of physical and numerical models, allowing rapid simulator development.
 Incorporation of the most modern solvers in two different frameworks—PETSC and TRILLINOS.
 Direct compatibility with sophisticated fracture generation and meshing programs.
 Threephase finiteelement simulations of systems of complex geometry.
 Two distinct numerical algorithms appropriate for specific applications—CVFE and MFE.
 Benchmarked well models—allowing simulation of complex multilateral wells in complicated geometry.
 Benchmarked hydraulic fracture simulations using compressible singlephase flow analytical results.
 Thermal simulation of complex geologic fractured systems with complex wells now possible.
 Thermalcompositional simulation of reservoirs using the CVFE formulation appropriate for complex geometries.
 Parallel formalism for unstructured grids.
 Based upon reservoir and fracture description, the simulator could be employed to optimize oil recovery.
 A fully implicit, general compositional thermal model using a new equation lineup concept was developed. A study of two different applications, one with the controlvolume finite element discretization and the other with finite difference showed that a given physical model can be integrated with different “Discretization Method” modules to solve the reservoir problem.
 Incorporation of a discretefracture representation provides an alternative to dual porosity models for simulating thermal recovery in fractured reservoirs.
 Developed a multiphase, multidimensional fractured reservoir simulation workflow.
This project has created tools to better manage fractured/faulted reservoirs and reservoirs with complex wells. Better management of these reservoirs will lead to significant improvement in oil and gas production from these reservoirs.
Current Status (February 2009)
This project has been completed. The final report is listed below under "Additional Information".
Project Start: September 9, 2004
Project End: August 31, 2008
Anticipated DOE Contribution: $800,000
Performer Contribution: $200,000 (20 percent of total)
Contract Information
NETL – Gary Covatch (gary.covatch@netl.doe.gov or 3042854589)
University of Utah  Milind Deo (mddeo@eng.utah.edu or 8015817629)
Additional Information
Final Project Report [PDF20.9MB]
Publications
Fu, Yao, Yang, YiKun and Deo, M. D., 2005, ThreeDimensional, ThreePhase DiscreteFracture Reservoir Simulator Based on Control Volume Finite Element (CVFE) Formulation, SPE 93292, Society of Petroleum Engineers, Dallas, Texas.
Yang, Yikun and Deo, M. D., 2006, Modeling of Multilateral and Maximum ReservoirContact Wells in Heterogeneous Porous Media, SPE 99715, Society of Petroleum Engineers, Dallas, Texas.
The two heterogeneous domains used to demonstrate the capability of being able to model and simulate heterogeneity in the plane of the fractures. In the left panel, three permeabilities (low =10 md, medium =100 md, and high =1000 md) were distributed randomly and evenly (33 percent each). In the right panel, domain with 80 percent low and 20 percent highpermeability distribution is shown.
The domain and the grid system used to demonstrate the applicability of the simulator to model multisegmented wells.
A faultedfractured domain simulation showing advantages and disadvantages about producing from the fault zone.
Domain decomposition of a complex geometrical system and speedup observed for 64 processors.
