General implicit coupling framework for multi-physics problems

Ruslan Rin, Pavel Tomin, Timur Garipov, Denis Voskov, Hamdi Tchelepi

Research output: Chapter in Book/Conference proceedings/Edited volumeConference contributionScientificpeer-review

12 Citations (Scopus)


We present a new framework for solving coupled multi-physics problems. The objective is to develop a platform where different coupling strategies for the simulation of complex physical processes can be employed with great flexibility in order to find an optimal - in terms of robustness and computational efficiency - strategy for a given problem. The new simulator is modular; each module represents a particular physics process, such as compositional, thermal, poromechanics, reactions, wells, and surface facilities. The platform provides seamless coupling between the physics modules without resorting to conditional branches and intermediate interface and treats terms that are coupled across multiple physics modules efficiently. The different modules can be coupled with each other in a sequential or fully coupled manner, and different solution strategies can be applied to different modules. This allows investigation of complex coupling strategies that have not been studied before. Examples of target problems include modeling compositional-thermal EOR processes in both conventional and unconventional resources with tight coupling to nonlinear poromechanics. The paper addresses the design of the framework and provides details of its implementation.

Original languageEnglish
Title of host publicationSociety of Petroleum Engineers - SPE Reservoir Simulation Conference 2017
PublisherSociety of Petroleum Engineers
Number of pages16
ISBN (Print)9781510838864
Publication statusPublished - 1 Jan 2017
EventSPE Reservoir Simulation Conference 2017 - Montgomery, United States
Duration: 20 Feb 201722 Feb 2017


ConferenceSPE Reservoir Simulation Conference 2017
Country/TerritoryUnited States


Dive into the research topics of 'General implicit coupling framework for multi-physics problems'. Together they form a unique fingerprint.

Cite this