Discover our new research paper, “Insights of Using Control Theory for Minimizing Induced Seismicity in Underground Reservoirs”!

The main obstacle to exploiting the earth’s underground renewable energy resources is earthquakes caused by fluid injections. In our study, we applied control theory at the reservoir level for the first time to prevent induced seismicity while ensuring fluid circulation to meet energy demands. Despite using a simplified reservoir model, our robust controller can handle the uncertainties and heterogeneities inevitable in real-world scenarios. The derivation was challenging, involving the coupling of a PDE with a nonlinear ODE to estimate seismicity rates.

Our research is advancing rapidly beyond this publication, promising even more innovative solutions. Stay tuned for future developments!

🔗 Read the full paper here

#RenewableEnergy #GeothermalEnergy #CarbonCapture #HydrogenStorage #ControlTheory #Sustainability #Research #SeismicityControl #ERC_Research

Major milestone! Our ERC-funded research project “Controlling earthQuakes” (CoQuake) concluded on November 30th, 2023. We’ve developed innovative methods to control seismic activity through fluid injections, presenting new possibilities for earthquake prevention.

The CORDIS article below highlights how we achieved controlled aseismic slip by fluid injections and other major scientific results, which I am proud to share along with their broader implications for science and society.

Many thanks to all the brilliant people I had the chance to collaborate with. These fascinating results open new perspectives for renewable energy production, induced seismicity prevention, and fundamental scientific breakthroughs.

To be continued, see ERC-INJECT project!

Ever wondered why materials fail? It’s often due to localized deformation in ultra-thin zones, a headache for classical continuum theories. They tend to give us unphysical results and numerical quirks like mesh dependency and zero dissipation.

Many believe multiphysical processes could be the solution – things like thermo-hydro-chemo-mechanical couplings. But in our latest research, we did the math to see if they really can elliminate mesh dependency issues.

Surprisingly, our findings challenge the common belief. Multiphysics processes don’t always solve the problem! Intrigued?
Explore the details here: https://doi.org/10.1016/j.ijmecsci.2024.109265

Congratulations, George, on earning the 2023 Pierre Londe PhD Prize from the French Society of Rock Mechanics! I am very proud of you! It was a real pleasure collaborating with you and Phil to tackle challenging problems on EQ control and fault mechanics. Your dedication and expertise shine through and this recognition is well-deserved. 👏

We are excited to present our latest research paper, which deepens into the possibility of creating a slow aseismic response in an earthquake model by fluid injections. The implications of this research are promising, offering enhanced earthquake control strategies.
Our approach involves employing a wave partial differential equation with uncertain multiplicative parameters and additive boundary and in-domain terms. Through the design of a robust homogeneous boundary feedback controller using a Lyapunov approach, we aim to drive the slip and slip-rate of the wave equation to follow a slow reference. This controlled response allows for the dissipation of stored energy, contributing to a slow-aseismic behavior.
Explore the details here: https://www.sciencedirect.com/science/article/pii/S0167691123001184

Our research dives deep into the physics behind fault friction. Using the Cosserat theory and accounting for large shear deformations, we have gained valuable insights. Our numerical results provide further evidence for the rate and state friction law, enhancing our understanding of earthquake nucleation and seismic energy release. We also made an exciting discovery—traveling shear bands within faults, resulting in oscillating frictional responses. These findings challenge conventional models and expand our knowledge of frictional weakening during earthquakes.

https://www.sciencedirect.com/science/article/abs/pii/S0020740323000863

📢 Calling all Geomechanics enthusiasts! Join us at the ALERT Geomaterials Doctoral School 2023 from September 28th to 30th, 2023, for an exciting exploration of the intersection between Machine Learning and Geomechanics.

🤖🏗️ Unleashing the Power of Machine Learning in Geomechanics 🌍💡

Are you curious about Machine Learning and its applications in solving real-life challenges in geomechanics and solid mechanics? This doctoral school is designed to provide you with a comprehensive understanding of Machine Learning, its key methods, and how it can be leveraged to tackle complex problems in the field.

📚 Through a combination of lectures and practical exercises, you will delve into regression and classification ML methods, supervised and unsupervised techniques, Artificial Neural Networks, deep learning, model reduction techniques, and even delve into reinforcement learning.

By attending this doctoral school, you will:

1️⃣ Gain a solid understanding of Machine Learning and its potential in geomechanics.

2️⃣ Study the most important ML methods used in the field.

3️⃣ Explore the fundamental mathematical and geometric principles behind ML techniques.

4️⃣ Get hands-on experience using ML through practical examples, while also developing an appreciation for physics- and geomechanics-based ML methods.

Don’t miss this unique opportunity to expand your knowledge and skills at the forefront of Machine Learning in Geomechanics. Join us at the ALERT Geomaterials Doctoral School 2023 and unlock the potential of this transformative technology in your field.

Register now and secure your spot! For more details, visit our website https://alertgeomaterials.eu/2023/06/registration-for-the-alert-workshop-and-school-2023-is-open/.

#ALERTGeomaterials2023 #MachineLearning #Geomechanics #SolidMechanics #Education #Research #DoctoralSchool

https://www.sciencedirect.com/science/article/pii/S0022509623000492

Machine learning (ML) approaches have shown high potential for replacing classical constitutive models for the description of the mechanical behavior of inelastic, heterogeneous materials. However, existing ML aproaches are incremental in time mostly and, therefore, they fail to provide a continuous-time constitutive description that is independent of the time step used for in training data generation.

To deal with this problem we propose a new approach which allows to decouple the material representation from the incremental formulation. Inspired by the Thermodynamics-based Artificial Neural Networks (TANN) and the theory of the internal variables, the evolution TANN (eTANN) are continuous-time and, therefore, independent of time incrementation. Key feature of the proposed approach is the identification of the evolution equations of the internal variables in the form of ordinary differential equations, rather than in an incremental discrete-time form. In this work, we focus attention to juxtapose and show how the various general notions of solid mechanics are implemented in eTANN.

Building on previous works, we propose a methodology that allows to identify, from data and first principles, admissible sets of internal variables from the microscopic fields in complex materials. The capabilities as well as the scalability of the proposed approach are demonstrated through several applications involving a broad spectrum of complex material behaviors, from plasticity to damage, viscoplastitcity and double scale homogenization of heterogeneous inelastic materials.

We are delighted and honored that our recently published article in JGR Solid Earth “A Discrete Elements Study of the Frictional Behavior of Fault Gouges” was selected as an Editor’s research Highlight and featured in Eos science news magazine published by AGU: https://eos.org/editor-highlights/upscaling-slip-and-friction-from-grains-to-the-fault-core.

IEEE Transactions on Control Systems Technology
arXiv

This paper addresses the possibility of using robust control theory for preventing earthquakes through fluid injections in the earth’s crust. The designed robust controllers drive aseismically a seismic fault to a new equilibrium point of lower energy by tracking a slow reference signal. The control design is based on a reduced-order nonlinear model able to reproduce earthquake-like instabilities. The designed controllers generate a continuous control signal to stabilize the tracking error despite the presence of uncertainties related to the frictional and mechanical properties of the underlying physical process and external perturbations. The developed controllers are tested extensively and compared on the basis of numerical simulations and experiments in the laboratory. The present work opens new perspectives for the application of robust nonlinear control theory to complex geosystems, earthquakes and the production of renewable energies.