Theses and Dissertations

ORCID

https://orcid.org/0009-0008-1578-5556

Advisor

Chen, Zhiqian

Committee Member

Chen, Jingdao

Committee Member

Fang, Xin

Committee Member

Mu, Lin

Committee Member

Wang, Haifeng

Date of Degree

12-12-2025

Original embargo terms

Immediate Worldwide Access

Document Type

Dissertation - Open Access

Major

Computer Science

Degree Name

Doctor of Philosophy (Ph.D.)

College

James Worth Bagley College of Engineering

Department

Department of Computer Science and Engineering

Abstract

Network flows govern a wide range of critical systems, from tangible infrastructures like transportation and power grids to replicable processes such as information spread and epidemics. While tangible flows obey conservation laws and physical constraints, replicable flows, like rumors or viruses, can grow, decay, or vanish unpredictably. Despite their increasing interaction in real-world settings, these flow types are typically modeled in isolation, using disconnected mathematical frameworks. This dissertation presents a unified modeling approach that bridges the gap between conserved and replicable flows by embedding principles from fluid dynamics into graph-based propagation models. I introduce physically informed extensions to classical models, such as a Navier-Stokes-inspired SIR model and a fluid-dynamic reinterpretation of the Independent Cascade model. These formulations incorporate external forces, momentum, and dissipation to bring conservation awareness to probabilistic spreading processes. To support interpretability and efficient control of network flows, I also develop scalable tools: a Sobol-based feature attribution method for influence maximization, a Bayesian optimization framework for source localization and influence blocking maximization, and a directional tensor embedding system for multilayer propagation alignment. Several of these contributions are validated on real and simulated systems, including livestock epidemic data (VSV), social influence networks, and a co-simulation platform integrating EV traffic with power grid dynamics. Together, these efforts lay the foundation for a graph-based fluid dynamics framework. It unifies stochastic, physical, and multilayer propagation into a cohesive, extensible theory. This work opens new avenues for modeling hybrid flows, improving intervention strategies, and discovering governing equations from data using tools like symbolic regression and Koopman theory.

Download

Share

COinS