GENERAL PURPOSE COMPOSITIONAL SIMULATION FOR MULTIPHASE REACTIVE FLOW WITH A FAST LINEAR SOLVER
Open Access
- Author:
- Qiao, Changhe
- Graduate Program:
- Mathematics
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- May 13, 2015
- Committee Members:
- Jinchao Xu, Dissertation Advisor/Co-Advisor
Russell Taylor Johns, Dissertation Advisor/Co-Advisor
Li Li, Dissertation Advisor/Co-Advisor
Ludmil Tomov Zikatanov, Committee Member
Chun Liu, Committee Member - Keywords:
- PennSim
FASP
compositional simulation
reservoir simulation
reactive transport - Abstract:
- Reservoir simulation is an important tool for petroleum engineers to predict oil production and optimize the management of oil fields. Fluids in oil and gas reservoirs are complex. In many cases, reservoir fluids consist of three or more phases and contain a large number of species that can react with each other. Many existing models (black oil, compositional and reactive transport models) were developed to simulate fluids with a specific fluid characterization. In this dissertation, a general modeling framework is developed that allows for simulation of different fluid types and processes. Based on the general framework we developed a new in-house simulator that we named PennSim. Reservoir simulators spend over 90% CPU time on the linear solver for reservoir models with over one million grid blocks. The performance of iterative linear solvers depends on the choice of the preconditioner, the design of which depends on the knowledge of the PDEs. We designed advanced multistage preconditioners based on the PDE structure. New decoupling strategies were proposed for the constraint pressure residual method. Advantageous performance is demonstrated and over three times speed-up is shown. We applied the general modeling framework to design novel models to explore the mechanisms of enhanced oil recovery (EOR). With coupled surface complexation reaction and multiphase flow, we proposed a numerical model for low salinity waterflooding for carbonates. The wettablility alteration was modeled by detailed description of fluid and surface reactions. Our model is the first predictive model for low salinity EOR in carbonates. We also designed the coupled compositional and reactive model for water-alternating CO2 injection. Injectivity changes were predicted for injection water with different compositions. This dissertation aims to cover the generic modeling, numerical simulation and fast iterative solvers for multiphase multicomponent flow and transport in porous media. PennSim was developed and its manual is available in supplemental materials.