Efficient development and uncertainty propagation of aviation fuel chemical kinetic models

Wiersema, Paxton

Permalink

https://hdl.handle.net/2142/132750 Copy

Description

Title

Efficient development and uncertainty propagation of aviation fuel chemical kinetic models

Author(s)

Wiersema, Paxton

Issue Date

2025-10-29

Director of Research (if dissertation) or Advisor (if thesis)

Lee, Tonghun

Doctoral Committee Chair(s)

Lee, Tonghun

Committee Member(s)

• Mayhew, Eric
• Glumac, Nick
• Ansell, Phillip

Department of Study

Mechanical Sci & Engineering

Discipline

Mechanical Engineering

Degree Granting Institution

University of Illinois Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Keyword(s)

combustion, chemical kinetics, SAF

Abstract

This work presents a comprehensive methodology for the rapid development of reduced chemical kinetic mechanisms for jet fuels, with a focus on sustainable aviation fuels (SAFs). The framework is designed to balance model accuracy, computational efficiency, and experimental requirements for quick development times and application in complex environments like computational fluid dynamics (CFD). The method uses experimental ignition delay data from a shock tube and a rapid compression machine as primary validation targets, which are then used for data-driven optimization of chemical kinetic mechanisms. To efficiently propagate and reduce model uncertainty, a hybrid response surface network and stochastic gradient descent ensemble (HRSN-SGDE) technique is developed, enabling the rapid generation and analysis of a large distribution of optimized mechanisms. An approach is then introduced to generate a base mechanism from fuel composition data and a detailed surrogate mechanism library, which estimates averaged molecular properties and pyrolysis products for a lumped fuel decomposition mechanism structure. The methodology is demonstrated on four diverse fuels, and the uncertainty is propagated through a series of simulation outputs, including ignition delay, flame speed, and species concentrations. Finally, this uncertainty is propagated into a diffusion flame simulation to show uncertainty effects in more complex environments. The results show that while ignition delay measurements effectively constrain the models to match global performance, significant uncertainty can persist in other outputs, such as species concentration profiles, highlighting key areas for future improvements to enhance the reliability of chemical kinetic models for next-generation aviation fuels.

Graduation Semester

2025-12

Type of Resource

Thesis

Handle URL

https://hdl.handle.net/2142/132750

Copyright and License Information

Copyright 2025 Paxton Wiersema

Owning Collections