We recently published a study proposing a control-theory–based approach to minimize induced seismicity during fluid injection in underground reservoirs used for geothermal energy and subsurface storage. The method relies on a physics-based seismicity model combined with robust tracking control to regulate seismicity rates and fluid pressure while maintaining the fluid circulation required for energy production.
Numerical simulations demonstrate that this approach can safely manage seismic risk under uncertain reservoir conditions and variable energy demand, offering a mathematically grounded alternative to empirical mitigation strategies.
For more details: https://www.sciencedirect.com/science/article/pii/S2352380824000376?via%3Dihub

#InducedSeismicity #GeothermalEnergy #ControlTheory #EnergyGeotechnics #GeoEnergy #SubsurfaceEngineering, @diego gutièrrez oribio @European Research Council (ERC) @ENSTA

We recently published a study exploring how control theory can be used to optimize hydraulic stimulation while mitigating induced seismicity in geothermal systems using a single-well. Using a robust control we show how injection schedules can be adapted in real-time to increase reservoir stimulation efficiency while keeping seismic risk within regulatory limits.
The framework is demonstrated on the 2018 Otaniemi geothermal project, highlighting how control can outperform traditional traffic-light approaches in both safety and performance.
For more details: https://www.sciencedirect.com/science/article/pii/S0375650525001476?via%3Dihub

#GeothermalEnergy #InducedSeismicity #ControlTheory #EnergyTransition #Geomechanics #SustainableEnergy, @taeho kim @diego gutiérrez oribio, @mateo acosta, @jean philippe avouac, @European Research Council (ERC), @ENSTA

We recently published a study introducing a POD–TANN framework for multiscale modeling in geomechanics. By combining Proper Orthogonal Decomposition with thermodynamics-based neural networks, the approach enables accurate, energetically consistent macroscopic models directly derived from high-fidelity simulations, with a strong reduction in computational cost.
For more details: https://onlinelibrary.wiley.com/doi/epdf/10.1002/nag.3891

@giovanni piunno, @cristina jomni, @European Research Council (ERC), @ENSTA, @CentraleNantes
#ComputationalMechanics, #TANN, #ReducedOrderModeling, #Geomechanics, #Homogenization, #Macroelements

Traditional time integrators struggle with nonsmooth systems, often slowing down or losing accuracy. Our new approach uses Reinforcement Learning to automatically choose optimal time steps, adapting even when the dynamics are nonsmooth. Tested on three challenging problems, a sliding-mode control system, a diode circuit, and a seismic fault model, our RL-based integrator significant speed-ups, reaching up to 10× faster in the hardest cases. A promising new direction for efficient time integration across complex dynamical systems.
For more details: https://link.springer.com/article/10.1007/s44379-025-00048-6

INJECT is an ERC-funded project focused on preventing human-induced seismicity as part of the fight against climate change. Through this site, we’ll share the latest news, research findings, innovative tools, and valuable resources related to our mission.

Explore our website to learn more about the project, the scientific challenges we address, our team, open positions, and more. Stay tuned as we share our curiosity and discoveries in this evolving field!