Data-driven models and digital twins for sustainable combustion technologies

Area of Science:

• Combustion science and engineering
• Sustainable energy technologies
• Data-driven modeling and artificial intelligence

Background:

• Industrial combustion systems are critical for high-density energy but are complex and computationally expensive to simulate.
• Existing high-fidelity simulations are often impractical for real-world industrial applications due to computational costs.
• Decarbonization goals necessitate innovative solutions for cleaner energy production.

Purpose of the Study:

• To highlight the essential role of data in developing sustainable combustion technologies.
• To explore the potential of data-driven approaches and artificial intelligence (AI) in addressing combustion system complexities.
• To discuss challenges and propose solutions for creating predictive digital twins of industrial combustion systems.

Main Methods:

• Leveraging data-driven methodologies and AI to overcome limitations of traditional physics-based simulations.
• Focusing on the development of digital twins that accurately represent industrial combustion system behavior.
• Addressing challenges in data availability, data fidelity, and integration with numerical simulations.

Main Results:

• Data-driven approaches and AI offer viable pathways for modeling complex combustion systems.
• Renewable synthetic fuels can be effectively integrated using these advanced techniques.
• The development of continuously updating digital twins is crucial for practical applications.

Conclusions:

• Data and AI are pivotal for advancing sustainable combustion and achieving decarbonization targets.
• Overcoming challenges in data and simulation is necessary for the widespread adoption of digital twins in industry.
• Future research should focus on enhancing data fidelity and integrating AI with physics-based models for robust combustion solutions.