Accelerating iterative solution methods using reduced‐order models as solution predictors

摘要: We propose the use of reduced-order models to accelerate solution systems equations using iterative solvers in time stepping schemes for large-scale numerical simulation. The acceleration is achieved by determining an improved initial guess process based on information vectors from previous steps. algorithm basically consists two projection steps: (1) projecting governing onto a subspace spanned low number global empirical basis functions extracted step solutions, and (2) solving this reduced space back original, high dimensional one. applied simulation two-phase flow through heterogeneous porous media. In particular we considered implicit-pressure explicit-saturation (IMPES) investigated scope pressure equation, which far most time-consuming part any IMPES scheme. substantial reduction iterations associated solution. Our largest test problem involved 93 500 variables, case obtained maximum computing 67%. method particularly attractive problems with time-varying parameters or source terms. Copyright © 2006 John Wiley & Sons, Ltd.