Multiphase Thermo-Poromechanics on Corner Point Geometry Grids using Finite Volume Method
This paper proposes and validates a finite volume method using the Multi-Point Stress Approximation with weak symmetry (MPSA-W) to compute geomechanics directly on industry-standard Corner Point Geometry grids, thereby eliminating the need for error-prone grid conversions and improving accuracy in modeling CO2 injection scenarios with faults and pinch-outs.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
The earth beneath our feet is not a static stage; it is a living, breathing system that reacts to every change we impose upon it. When we pump fluids out of deep underground reservoirs to extract oil or gas, or when we inject new fluids back in, the rock layers shift, compress, and sometimes crack. This movement, known as geomechanics, is a critical factor in modern energy projects. If the ground moves unexpectedly, it can damage the well, trigger small earthquakes, or even compromise the safety of storing carbon dioxide deep underground to fight climate change. To predict these movements, scientists build digital models of the earth, but these models are only as good as the tools used to calculate them. For decades, the standard tool for simulating how fluids move through rock has been different from the tool used to calculate how that rock deforms. This mismatch has forced engineers to constantly reshape their digital maps, a tedious and error-prone process that often distorts the very features they are trying to study, such as the jagged lines of underground faults.
A team of researchers from TotalEnergies, SINTEF Digital, and CHLOE has developed a new way to solve this problem, allowing them to calculate rock deformation directly on the same complex maps used for fluid flow. Their work focuses on a specific type of digital grid called Corner Point Geometry, which is the industry standard for representing the messy, irregular reality of underground rock layers, including the sharp angles of faults and the thinning layers where rock disappears. Traditionally, to study how these grids react to pressure and temperature changes, engineers had to convert the grid into a different, simpler shape that older computer methods could understand. This conversion often stretched or squashed the grid cells, especially near faults, leading to inaccurate predictions. The researchers propose a new approach that skips this conversion entirely, using a mathematical technique called the Multi-Point Stress Approximation to calculate the forces and movements directly on the original, complex grid.
The team tested this new method against established computer programs that use the older, conversion-based approach, as well as against known mathematical solutions for simple scenarios. They began with basic models to ensure the new method worked correctly on regular grids, grids with faults, and grids where rock layers pinch out and vanish. In these tests, the new method produced results that matched the known solutions with high precision, showing only tiny differences even on the most distorted grids. They then moved to more realistic scenarios, simulating the injection of cold water into a hot reservoir and the injection of cold carbon dioxide into a depleted gas field in the Netherlands. In these complex simulations, the new method continued to agree closely with the traditional software, successfully capturing how the rock responded to changes in pressure and temperature.
One of the most significant findings was the ability to handle non-conforming grids, where the layers of rock do not line up perfectly from one side of the model to the other, a common feature in real geological formations that includes faults. The traditional software struggled with these irregular shapes, often requiring the grid to be smoothed out or simplified, which could hide important details. The new method, however, handled these irregularities naturally, producing results that were nearly identical to the reference software in areas where the grids matched, and providing stable solutions in areas where the traditional software could not run at all. The researchers also explored how tightly the fluid flow and rock movement need to be calculated together. They found that for many scenarios, a simpler, step-by-step calculation was sufficient, saving significant computing time without sacrificing accuracy.
The study concludes that this new method is a robust and reliable alternative to the traditional tools used in the oil and gas industry. By allowing engineers to run geomechanical simulations directly on the complex grids they already use for fluid flow, the new approach eliminates the need for risky grid conversions and reduces the time spent preparing models. This is particularly valuable for projects involving geological carbon storage, where the safe injection of cold carbon dioxide into hot, deep reservoirs requires precise predictions of how the rock will behave. The researchers validated their work on a real-world field model from the Netherlands, demonstrating that the method can handle the extreme variations in rock properties found in nature, including areas with very soft rock and areas with extremely stiff salt domes. While the method showed a minor discrepancy in one specific, highly unstable corner of a test model, the overall results confirmed that it provides a clear, accurate, and efficient path forward for simulating the earth's response to industrial activity.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.