Advancing predictive models for particulate formation in turbulent flames via massively parallel direct numerical...

Area of Science:

• Chemical Engineering
• Combustion Science
• Computational Fluid Dynamics

Background:

• Global energy consumption drives continued fossil fuel use, necessitating improved combustion efficiency and reduced pollutant emissions for climate change mitigation.
• Accurate and comprehensive data are crucial for advancing turbulent combustion modeling, which is essential for designing cleaner combustion devices.
• Pollutant formation is a complex process involving turbulent mixing, chemical reactions, and multi-scale interactions.

Purpose of the Study:

• To demonstrate how direct numerical simulation (DNS) databases can illuminate physico-chemical mechanisms in turbulent flames.
• To identify specific modeling needs for pollutant formation, particularly soot, in combustion processes.
• To support the development of more accurate computational fluid dynamics (CFD) models for cleaner combustion technologies.

Main Methods:

• Utilized direct numerical simulation (DNS) with detailed chemical kinetics and physical process descriptions.
• Analyzed the intricate evolution of soot formation within turbulent flames.
• Leveraged DNS databases to assess and identify deficiencies in current combustion models.

Main Results:

• DNS analysis provided insights into the complex mechanisms governing soot formation in turbulent flames.
• The study highlighted critical areas where existing combustion models require improvement.
• Identified specific physico-chemical processes that need to be better represented in CFD models.

Conclusions:

• Direct numerical simulation (DNS) is a powerful tool for understanding complex combustion phenomena like soot formation.
• The findings underscore the need for enhanced turbulent combustion models informed by high-fidelity simulation data.
• Improved combustion models are vital for developing efficient and low-emission energy technologies.