Abstract

The talk will review relevant R&D topics that are still important today in areas such as preprocessing and non-matching interfaces for coupled flow and mechanics, i.e., the combination of computational geometry to reconstruct the geometry of reservoirs and the usage of different meshes for flow and mechanics via projection, and mortar interfaces to glue non-conforming subdomains. It then summarizes findings on reduced order modeling to hasten the resulting nonlinear computations for thermo-poroelasticity and applications to unconventional reservoirs coupled with mechanics. The talk ends by discussing optimization with the distributed Gauss-Newton method for assisted history matching (AHM) and machine learning with applications to well-location optimization (WLO). 

Biography

Horacio Florez is a computational scientist with a strong background in engineering mechanics, particularly in discretizations and parallel computing. He holds a master's and Ph.D. from The University of Texas at Austin. His career spans over twenty years in the oil and gas industry and academia, primarily working with finite elements applied to multi-physics. He has published more than 60 journal and conference papers and has contributed to computational geometry, mesh generation, coupled flow and geomechanics, reduced-order modeling, optimization, and machine learning. He is a Lead Petroleum Engineer at Saudi Aramco and currently researches coupled flow and geomechanics for CCS applications and optimization algorithms for assisted history matching.