The goal of this research was to develop a general-purpose, adaptive reservoir simulator for high-resolution simulations. Specifically, research was to develop a new-generation chemical flooding simulator capable of efficiently and accurately simulating oil reservoirs with at least a million gridblocks in less than one day on parallel computers.

Program
This project was selected in response to DOE's Oil Exploration and Production solicitation DE-PS26-01NT41048, focus area Reservoir Efficiency Processes. The goal of the solicitation focus area is to address the need to access oil not recoverable by conventional methods. Technologies may address gas flooding, heavy oil recovery, the use of chemicals, reservoir simulation, or microbes.

University of Texas
Austin, TX

Increased domestic oil production using advanced technologies of improved oil recovery (IOR) processes involves numerical modeling of such processes to minimize the risk involved in development decisions. The oil industry is requiring much more detailed analyses, with increasing demand for reservoir simulations with geological, physical, and chemical models of much greater detail than in the past. The computational work required to produce accurate reservoir simulations is very large for these problems, and thus there is a great need for both parallel computing and implicit algorithms.

The premise of this research is that a general-purpose reservoir simulator for several IOR processes should be developed so that high-resolution simulations of a variety of very large and difficult problems can be achieved using state-of-the-art computing and computers. Such a simulator is not currently available to the industry. The goal of this proposed research is to develop a new-generation chemical flooding simulator that is capable of efficiently and accurately simulating oil reservoirs with at least a million gridblocks in less than one day on massively parallel computers or on a cluster of PCs. The simulator also includes the implicit compositional EOS model to simulate gas processes and a nonorthogonal grid option so that geological features such as curved reservoir boundaries and impermeable barriers like shales and faults can be modeled.

Project Results
A fully implicit, parallel compositional reservoir simulator called GPAS is developed with the capability of both equation-of-state gas compositional and surfactant/oil/brine phase behavior. A simulation of over 1 million gridblocks of a surfactant/polymer flood was successfully performed using more than 100 processors.

Benefits
There is a greater demand for parallel computing since the oil industry is requiring reservoir simulations with geological, physical, and chemical models of much more detail.

Project Summary
Project researchers have:

• Developed a fully implicit, parallel, compositional chemical flooding simulator and validated it against other simulators.
• Determined that the results of parallel runs are almost identical to those on a single processor.
• Shown that the simulator scales well using a cluster of workstations.
• Successfully performed field-scale, high-resolution surfactant/polymer flood simulations with over 1 million gridblocks.

(August 2005)
Researchers are currently collaborating with an oil company to use the GPAS simulator in its field operations and designs of surfactant/polymer flooding.

Publications
Final Report [PDF-1.60MB] - January, 2005

John, A., Han, C., Delshad, M., Pope, G.A., and Sepehrnoori, K., A New-Generation Chemical Flooding Simulator, SPE 89436, accepted for publication in SPE Reservoir Evaluation and Engineering, 2005.

Han, C., Delshad, M., Sepehrnoori, K., and Pope, G.A., A Fully Implicit, Parallel, Compositional Chemical Flooding Simulator, SPE 97217, presented at the SPE Annual Technical Conference and Exhibition, Dallas, TX, October 9-12, 2005.

Six DOE reports.

$930,801

$234,689 (20% of total)

NETL - Sue Mehlhoff (sue.mehlhoff@netl.doe.gov or 918-699-2044)
U. of Texas - Gary Pope (gpope@mail.utexas.edu or 512-471-3235)