AI-based Reservoir Modelling Workflows - an Illustrative Overview

Modern AI developments open outstanding opportunities to tackle generic reservoir modelling tasks with the methods designed to handle diverse and uncertain data. Effective AI application to reservoir modelling workflows relies on the ability to ensure interpretability of the machine learning model outcomes. This can be achieved by embedding the domain context into the AI model structure, so the data are no longer treated as merely digital values but as the variables with physical meaning and interpretation in the subsurface context.

This overview demonstrates a few examples of how AI applications resolve several steps of reservoir modelling workflow:
- AI seismic segmentation and geobody interpretation with unsupervised learning.
- Constrain geological conceptual modelling with learning sedimentological structure sequence patterns from outcrops.
- Populate facies in meandering fluvial reservoir models based on learning from depositional process modelling with generative adversarial networks (GANs).

The work will demonstrate how to gain better understanding and representation of associated geological uncertainty when geological domain knowledge is embedded into the AI algorithms’ structure.