Data-Driven Model Reduction, Scientific Frontiers, and Applications ()
- Texas A&M University
- College Station, TX
- Joe C. Richardson Petroleum Engineering Building (RICH) 910
- Eduardo Gildin, Harold Vance Department of Petroleum Engineering
- Model Reduction of Coupled Flow and Geomechanics: Ideas from Structural Mechanics
Abstract
Global model reduction has been applied extensively in reservoir simulation to mitigate the high computational cost of multiphase flow but, its application to combined multiphysics, as in the case of flow and geomechanics, has been minimal. In this talk, I will revisit some ideas from model reduction of structural mechanics problems, such as modal decomposition, static and dynamic condensation and will generalize its application to a projection framework (POD-DEIM). I will show as well how the POD-DEIM results focused on linear and nonlinear thermo-poroelasticity, yield substantial single and double digits speedups, up to 50X if we combine with multi-threading assembling and perform MOR on both flow and geomechanics.