A Workflow for Dynamic Screening Assisted, Automated Fractured Reservoir Modelling

The main objective of this work is to come up with an automated reservoir modelling workflow that allows for the partial circumvention of the labour intensive task of manual model building for fractured reservoirs. This is done by using an iterative optimisation algorithm (PSO) to link up an automated and parameterised reservoir modelling workflow with flow diagnostics, a fast proxy for fluid flow simulations. The goal here is to not only generate an ensemble of reservoir models in a fast manner, but also to carry a sufficient degree of geological realism throughout the entire modelling workflow. This is tested on a synthetic field that is based upon a conceptual model for fracture distributions in folded carbonate rocks.

The results indicate that this workflow is capable of delivering a diverse ensemble of reservoir models that match the targeted dynamic response very well. Nevertheless, it can be argued whether or not the models themselves preserve enough geological realism. To counteract that a penalising term for overly complex models and a metric for geological realism need to be included.