Author ORCID Identifier

https://orcid.org/0009-0000-8119-3008

Semester

Summer

Date of Graduation

2026

Document Type

Thesis

Degree Type

MS

College

Statler College of Engineering and Mineral Resources

Department

Mechanical and Aerospace Engineering

Committee Chair

David Mebane

Committee Member

Chetan Kulkani

Committee Member

Adam Halasz

Committee Member

Kostas Sierros

Abstract

One of the major hurdles for electric or hybrid aviation is the comparative energy density between lithium-ion batteries and jet fuel. To help obtain maximal power-to-weight ratio in electric or hybrid powertrains the battery pack design should be optimized to be as light as possible while meeting thermal safety and power requirements. The proposed solution is an optimization routine beginning with a cell-level parameter estimation of lithium-ion battery cells, where data driven Bayesian Gaussian Processes replace empirically or analytically derived parameters in a first principles electrochemical model in PyBaMM, as functions of the transient state of the cell. The resulting parameter functions provide uncertainty quantification, interpretability, and higher accuracy than constant parameter estimation techniques, validated on Samsung 25R lithium-ion 18650 2500 mAh battery cells. The cell models are linked in an electrothermal 4-series 2-parallel pack-level model over a simulated flight path. This framework allows for pack-level optimization using OpenMDAO for pack dimensions and cell placement. The cell-level diffusivity estimations resulted in better pack-level accuracy with minimal speed difference. This translated to a weight optimized pack design that meets power and thermal safety requirements. This framework allows for rapid design iteration, running on a work station in hours, and can drive design decisions and design of experiments for future aircraft

Download

Share

COinS