Date of Graduation
5-2025
Document Type
Thesis
Degree Name
Bachelor of Science in Chemical Engineering
Degree Level
Undergraduate
Department
Chemical Engineering
Advisor/Mentor
Monroe, Jacob I.
Abstract
This study investigates the use of derivative-informed Gaussian Process (GP) models to estimate thermodynamic behavior across temperature and density by building a Helmholtz-based equation of state. Argon, a stable monatomic gas, was chosen as a case study within the vapor region. The GP model was trained using values of experimentally measurable properties found by taking first and second derivatives of the original potential function. Results show that while the GP model offered uncertainty quantification and informed thermodynamic behavior, it predicted values that deviated from the ground truth depending on the property. The model exhibited high confidence in regions with substantial error due to an over-reliance on the mean function. Further analysis revealed that the mean function dominated the predictions while the kernel function minimally contributed. Recommendations are provided to weaken the mean function and optimize the kernel hyperparameters.
Keywords
Gaussian Process Model; Mean Function; Equations of State; Thermodynamic Derivatives; Helmholtz Free Energy; Uncertainty
Citation
Lee, A. T. (2025). Using Gaussian Process Regression to Learn Thermodynamic Equations of State with Uncertainty Quantification. Chemical Engineering Undergraduate Honors Theses Retrieved from https://scholarworks.uark.edu/cheguht/218
Included in
Computational Chemistry Commons, Computational Engineering Commons, Numerical Analysis and Computation Commons, Numerical Analysis and Scientific Computing Commons, Physical Chemistry Commons, Statistical Methodology Commons, Thermodynamics Commons
