Enabling CO2 capture and storage using AI
Lead Research Organisation:
Imperial College London
Department Name: Chemical Engineering
Abstract
Abstracts are not currently available in GtR for all funded research. This is normally because the abstract was not required at the time of proposal submission, but may be because it included sensitive information such as personal details.
Organisations
Publications
Cai S
(2024)
Advancing neural network-based data assimilation for large-scale spatiotemporal systems with sparse observations
in Physics of Fluids
Chen B
(2024)
Solving the discretised multiphase flow equations with interface capturing on structured grids using machine learning libraries
in Computer Methods in Applied Mechanics and Engineering
Maes J
(2024)
Dispersivity calculation in digital twins of multiscale porous materials using the micro-continuum approach
in Journal of Environmental Chemical Engineering
Naderi S
(2024)
A discrete element solution method embedded within a Neural Network
in Powder Technology
Wu X
(2023)
A long short-term memory neural network-based error estimator for three-dimensional dynamically adaptive mesh generation
in Physics of Fluids
| Description | The ECO-AI project has made progress in developing artificial intelligence (AI) technologies to make carbon capture and storage (CCS) more affordable and efficient - a critical step in fighting climate change. Our key achievements include: • Smarter CO2 Capture: We've created AI systems that can rapidly identify and test new chemical solvents that could dramatically reduce the energy needed to capture carbon dioxide from industrial emissions. This is like having a "virtual laboratory" that can test thousands of potential solutions much faster than traditional methods. • Advanced Storage Modeling: We've developed AI tools that can accurately predict how CO2 will behave when stored underground in geological formations. These tools help identify optimal storage sites and injection strategies while minimizing risks of leakage, making the entire process safer and more reliable. • Multi-scale Analysis: Our AI models can now bridge the gap between microscopic rock properties and large-scale storage behaviors, allowing for more accurate predictions of CO2 movement in diverse geological settings. This is like connecting what happens at the level of individual grains of sand to what happens across an entire underground reservoir. • Accelerated Innovation: We've created models that can predict how technological innovations might impact the feasibility and cost of CCS projects across different industrial sectors, helping to inform policy decisions and investment priorities. The project has produced several open-source software libraries and datasets that are currently publicising to other researchers in the field and our industrial partners. If adopted, these will accelerate the path to commercial deployment of more efficient CCS technologies. |
| Exploitation Route | The AI solvers, ML models and new datasets will serve for future researchers working in the area. Some work is still on-going, but these initial tools have demonstrated the suitability and potential of the use of AI to make CCS more efficient and affordable. Furthermore, our approaches to physics-aware AI and uncertainty quantification can be adopted by researchers working on other complex environmental systems. |
| Sectors | Chemicals Energy Environment |
| Description | • Industry Adoption: Several industrial partners, including energy companies, are now testing our AI frameworks for site selection and risk assessment in their CCS planning processes. • Policy Influence: Our sectoral analysis of CO2 reduction targets and innovation trajectories is informing discussions with policymakers about effective market interventions to accelerate CCS deployment. • Academic Cross-fertilization: The AI methods developed for CCS are being adapted for other environmental challenges, including groundwater remediation and subsurface energy storage. • Skills Development: The ECO-AI hackathons have trained a team of early career researchers in AI applications for environmental science, building capacity for the next generation of interdisciplinary scientists. |
| First Year Of Impact | 2025 |
| Sector | Chemicals,Energy,Environment |
| Impact Types | Societal Policy & public services |
| Description | ECO-AI Local Workshop |
| Form Of Engagement Activity | Participation in an activity, workshop or similar |
| Part Of Official Scheme? | No |
| Geographic Reach | Local |
| Primary Audience | Professional Practitioners |
| Results and Impact | First ECO-AI workshop for Imperial and Heriot-Watt teams to network, exchange ideas, and plan future work together. It was a very successful first meeting with the team coming together in person on one site. |
| Year(s) Of Engagement Activity | 2024 |
| Description | Participation in Second ECO-AI Local workshop |
| Form Of Engagement Activity | Participation in an activity, workshop or similar |
| Part Of Official Scheme? | No |
| Geographic Reach | Local |
| Primary Audience | Study participants or study members |
| Results and Impact | Second ECO-AI workshop for Imperial and Heriot-Watt teams, run at Imperial. A mid-point workshop for teams to come togehter, provide updates on progress, plan work ahead identifying collaboration opportunities between team members. |
| Year(s) Of Engagement Activity | 2024,2025 |