AI-powered geochemistry with Repsol

SLB and Repsol Technology Lab have collaborated to develop an innovative AI-driven reactive transport model (RTM-AI) in the Intersect™ reservoir simulator. It delivers rapid, accurate geochemical modeling, enabling faster, better informed decisions for energy transition projects, including carbon capture, utilization, and storage (CCUS).

Addressing computational barriers to scaling CCUS

As the global energy transition progresses new approaches to energy production and storage are urgently needed, including for CCUS. However, traditional reservoir modeling can struggle with the complexity and computational demands involved. For instance, geochemical interactions—such as mineral dissolution and precipitation—are critical for understanding subsurface dynamics for applications like CCUS and geothermal energy. Modeling these reactive transport processes is inherently complex and often computationally expensive. The high computational cost can limit the feasibility of detailed subsurface characterization studies and hinder operational decisions, creating a significant barrier to scaling new energy initiatives. Therefore, SLB and Repsol came together to develop a solution that integrates advanced geochemical modeling into reservoir simulations, without sacrificing speed or accuracy.

Advancing CCUS through AI‑enhanced subsurface modeling

Together, Repsol and SLB created a reactive transport model with artificial intelligence (RTM-AI) solution, an innovative framework that is now embedded in the Intersect™ reservoir simulator. Leveraging state-of-the-art machine learning (ML), the RTM-AI solution efficiently models the non-linear and coupled effects of reactive transport processes. This accelerates the evaluation of complex geochemical interactions, such as mineral dissolution and precipitation, while maintaining high accuracy. Initially applied to dolomitization studies, the approach is now being extended to other critical applications that advance carbon storage technologies, including the assessment of long-term stability of injected CO₂ in subsurface formations. The result is a scalable, efficient solution that empowers stakeholders to make timely, data-driven decisions for more sustainable energy projects.