Speaker
Description
Recently a novel method has been developed to determine relative permeability for gas-liquid systems e.g. for the underground storage of hydrogen from a hybrid experimental/modelling workflow. It follows the philosophy of measuring time sequences of pore scale fluid distributions by in-situ micro-CT imaging for gas-liquid systems which due to their rich physics (immiscible displacement combined with dissolution, diffusion and ripening effects) are inaccessible to numerical modelling. Relative permeability is then computed by numerical Stokes flow simulations on the imaged 3D pore scale fluid distributions for the gas and liquid phases. This method has the advantage of a much larger accessible mobile saturation range than traditional relative permeability measurements.
One of the complications in this methodology is that for instance in imbibition, where the wetting (aqueous) phase displaces the non-wetting (gas) phase, at already relatively low water saturation the gas phase is not permanently connected anymore at individual micro-CT snapshots but is transported by processes such as ganglion dynamics, which is not captured by the Stokes flow simulations which provide relative permeability only for a connected pathway. This has been overcome by restricting the Stokes flow simulations to sub-domains of the sample in which gas clusters percolate between inlet and outlet over most of the imaged time sequence. The respective workflow involves to a significant degree manual steps which limits the practical applicability but also restricts uncertainty analysis. The Stokes flow simulations which thanks to the highly optimized LIR solver in GeoDict solver run in 10 minutes or less are not a limiting factor. The key limiting factor is actually the selection of the sub-domains involves visual inspection in 3D combined with connected objects analysis followed by the Stokes flow simulations.
Here we present the development of a workflow where the selection of sub-domains is performed by an algorithmic process which then allows automation. The aim of the workflow is to integrate the whole post-processing which includes sub-domain selection and Stokes flow simulation in a fully automated workflow where all computational steps are stored in a data base with respective report and export functionality.
| Country | The Netherlands |
|---|---|
| Student Awards | I would like to submit this presentation into both awards |
| Acceptance of the Terms & Conditions | Click here to agree |
