Accelerating Oil-Water Subsurface Flow Simulation Through Reduced-Order Modeling and Advances in Nonlinear Analysis

Advisors

Louis Durlofsky, degree supervisor
Hamdi Tchelepi, degree supervisor
Oleg Volkov, degree committee member

Abstract

Reservoir simulation is an important tool for understanding and predicting subsurface flow and reservoir performance. In applications such as production optimization and history matching, thousands of simulation runs may be required. Therefore, proxy methods that can provide approximate solutions in much shorter times can be very useful. Reduced-order modeling (ROM) methods are a particular type of proxy procedure that entail a reduction of the number of unknown variables in the nonlinear equations. This dissertation focuses on two of the most promising proper orthogonal decomposition (POD)-based ROM methods, POD-TPWL and POD-DEIM. A separate (non-ROM) technique to accelerate nonlinear convergence for oil-water problems is presented in the appendix.