Date of Award

Fall 2025

Document Type

Open Access Thesis

Department

Mechanical Engineering

First Advisor

Sang Hee Won

Abstract

Large, complex chemical kinetic models have long since been limited by the computational cost it takes to evaluate said model. Over time many reduction methods have been developed to combat this computational cost. However, very little has been done towards evaluating the limitations of reductions as a way of understanding the original model’s characteristics. This paper aimed to use a path flux analysis-based reduction algorithm to rapidly develop multiple reduced mechanisms for a 96 species FFCM-2 methane chemical kinetic model as well as an 858 species n-Dodecane chemical kinetic model. The goal of formulating these reduced mechanisms was to identify the key species that support or inhibit autoignition predictions. Each mechanism was evaluated at low to high temperature ranges (800 – 1600 K), low to high pressure ranges (1 – 60 atm), and lean to rich fuel mixtures (𝜙𝜙=0.5 to 𝜙𝜙=3.0). The resulting reduction evaluations showed that low temperature radicals such as CH3O2 dominated the smaller methane model. However, when the initial fuel molecule (CH4) and the main combustion products (CO2, CO, and H2O) are properly emulated, the generated reduced model is able to accurately predict ignition delay times for all conditions without significant errors. This methodology was later expanded to a larger, more complex n-dodecane chemical kinetic model. To which, up to 8 core species were determined to assist in total model fidelity with an additional 6 core species defining ignition delay time. Reductions regarding this larger model were limited in some capacity to available resources which allows for further studies to be recommended.

Rights

© 2025, Christian Robert Smith

Download

Share

COinS