This research tackles the conceptual and computational challenges associated with continuum-based reduced chemical kinetics by utilizing data-driven, physics-informed modeling approaches. By leveraging these methods, we propose a novel framework to identify appropriate parameter sets for the model, requiring minimal intuition or prior knowledge of the system. Our systematic approach integrates machine learning, manifold learning, and data mining to uncover reduced chemical kinetic models and effectively design species interaction rules.
This research aims to develop a reduced-order model for simulating thermal runaway in lithium-ion batteries to better understand the fundamental mechanisms in a sing cell under extreme conditions. The key components of the cell are modeled to simplify the system, and their chemical reactions are modeled to identify the chemical paths. Numerical simulations for thermal runaway are conducted using a multi-phase diffuse interface model and adjoint-based data assimilation for parameter calibration. The goal is to enhance current modeling approaches and improve the prediction of thermal runaway events.
This research aims to understand the interfacial interactions in terms of energy transfer processes in energetic matrials with a particular emphasis on how shock waves and frictional heating affect the material's behavior under extreme conditions. In the shock-induced energy deposition study, the focus is on understanding how a shock blast interacts with a hyperelastic solid, leading to deformation and energy absorption. In the frictional heating study, the focus shifts to how mechanical vibrations, specifically from an ultrasonic horn, induce localized heating in a polymer-bonded explosive, potentially triggering thermal runaway reactions.
© Florida Institute of Technology, All Rights Reserved