Local and global uncertainty analyses of a methane flame model

Area of Science:

• Combustion chemistry
• Chemical kinetics
• Thermodynamics

Background:

• Laminar methane flame models are crucial for understanding combustion processes.
• Quantifying uncertainties in these models is essential for accurate predictions.

Purpose of the Study:

• To perform local and global uncertainty analyses on a laminar methane flame model.
• To investigate the impact of uncertainties in kinetic and thermodynamic data on key flame characteristics.

Main Methods:

• Utilized the Leeds methane oxidation mechanism for a flat, premixed, stationary, laminar flame.
• Employed Morris, Monte Carlo (Latin hypercube sampling), and improved Sobol' methods for global uncertainty analysis.
• Assigned probability density functions to 175 reaction rate coefficients and 37 species' enthalpies of formation.

Main Results:

• Determined approximate probability density functions and standard deviations for laminar flame velocity, maximal flame temperature, and radical concentrations (H, O, OH, CH, CH2).
• Identified specific rate parameters and enthalpies of formation as major contributors to model result uncertainty.
• Quantified minimum and maximum result values across physically realistic parameter combinations.

Conclusions:

• Model uncertainties are predominantly driven by a small subset of kinetic and thermodynamic parameters.
• Inaccurate knowledge of kinetic data is the primary source of uncertainty, though thermodynamic data also contributes significantly.