To main content

Dynamic coarsening for efficient simulation of geothermal energy applications

Abstract

Geothermal reservoir simulations are computationally demanding due to the need to resolve highly localized heat transport near wells alongside large-scale pressure dynamics. This presentation introduces a dynamic coarsening framework that significantly improves computational efficiency by adapting spatial resolution to the evolving thermal state of the system. The method constructs a hierarchy of coarse grids from an underlying fine-scale geomodel and dynamically assembles an adaptive grid at each timestep based on indicators of local solution complexity. High resolution is retained near active thermal regions, while less critical areas are coarsened. A key contribution is an energy-conservative mapping between grids that ensures accurate transfer of thermal quantities during adaptation, together with a sequential splitting scheme that allows pressure and heat transport to be solved on different spatial resolutions. Results from heterogeneous benchmark models and a real geothermal storage site demonstrate that the approach can reduce transport degrees of freedom by up to 80–90% while maintaining high accuracy. The framework is flexible, applicable to complex geomodels, and well suited for accelerating simulations of low- to medium-enthalpy geothermal systems.

Category

Conference lecture

Language

English

Affiliation

  • SINTEF Digital / Mathematics and Cybernetics

Presented at

EAGE GET 2026

Place

Hannover

Date

02.11.2026 - 06.11.2026

Organizer

EAGE

Date

04.11.2026

Year

2026

View this publication at Norwegian Research Information Repository