UWSpace
UWSpace is the University of Waterloo’s institutional repository for the free, secure, and long-term home of research produced by faculty, students, and staff.
Depositing Theses/Dissertations or Research to UWSpace
Are you a Graduate Student depositing your thesis to UWSpace? See our Thesis Deposit Help and UWSpace Thesis FAQ pages to learn more.
Are you a Faculty or Staff member depositing research to UWSpace? See our Waterloo Research Deposit Help and Self-Archiving pages to learn more.
Communities in UWSpace
Select a community to browse its collections.
Now showing 1 - 13 of 13
- The University of Waterloo institution-wide UWSpace community.
Recent Submissions
A Graph Neural Network Based Approach for Predicting Wildfire Burned Areas
(University of Waterloo, 2025-02-10) Das, Ursula
Wildfires annually cause substantial economic and environmental losses and has a detrimental impact on human lives and health due to the release of their harmful byproducts. Moreover, wildfire incidents have exhibited an alarming surge in frequency as well as severity in recent years due to increased urbanization near forested areas coupled with climate change, highlighting the need for advanced technologies to predict wildfire behavior in advance and mitigate its impact. In recent years, the enormous strides in machine learning research coupled with the increased availability of wildfire data through various sources such as remote sensing and the increased availability of computational resources have fueled the rise of data-driven approaches across all stages of wildfire management. Despite the growing adoption of machine learning-driven approaches in wildfire mitigation, the primary focus has been on analyzing historical patterns and identifying the causes leading to wildfire patterns rather than predicting wildfire behavior. The prediction of wildfire behavior over time, such as the burned area has been largely underexplored. This study aims to address this gap by advancing data-driven methods for predicting wildfire behavior during the active fire stage and aiding in resource allocation efforts.
This study adopts a Graph Neural Network based framework for predicting the burned area resulting from a wildfire ignition. While CNN-based architectures have been widely employed to model wildfire behavior, including spread prediction, as a semantic segmentation task, these architectures impose specific limitations on geospatial data due to their reliance on fixed-size inputs and local receptive fields. Graph Neural Network (GNNs), have shown success in capturing the long-range dependencies and irregular-sized inputs inherent in geospatial data, such as wildfires, making them a viable alternative to CNNs. To this end, a GNN-based approach is adopted to model wildfire burned area prediction. A framework is developed to represent spatial wildfire data and its influencing factors as homogeneous graphs followed by the development of three distinct GNN models based on different message-passing mechanisms to process the graph-structured data. The results obtained through various experiments illustrate the efficacy of Graph Neural Networks in modeling wildfire behavior. In terms of Precision, most GNN models outperform the segmentation models, with the highest achieving a score of 0.4536. For AUROC, all GNN models demonstrate superior performance, reaching a maximum of 0.9377. Based on AUPRC, the Graph Convolutional Network (GCN) model surpasses all others, including segmentation models, with a top score of 0.4787. These findings underscore the potential of Graph Neural Networks (GNNs) as a powerful tool for wildfire behavior modeling and supporting resource allocation initiatives.
Data-Driven Predictive Control: Equivalence to Model Predictive Control Beyond Deterministic Linear Time-Invariant Systems
(University of Waterloo, 2025-02-07) Li, Ruiqi
In recent years, data-driven predictive control (DDPC) has emerged as an active research area, with well-known methods such as Data-enabled Predictive Control (DeePC) and Subspace Predictive Control (SPC) being validated through reliable experimental results. On the theoretical side, it has been established that both DeePC and SPC methods can generate equivalent control actions as one can obtain from Model Predictive Control (MPC), for deterministic linear time-invariant (LTI) systems.
However, similar results do not yet exist for the application of DDPC beyond deterministic LTI systems. Therefore, the objective of our research is to generalize this theoretical equivalence between model-based and data-driven methods for more general classes of control systems.
In this thesis, we present our contributions to DDPC for linear time-varying (LTV) systems and stochastic LTI systems. In our first piece of work, we developed Periodic DeePC (P-DeePC) and Periodic SPC (P-SPC) methods, which generalize DeePC and SPC from LTI systems to linear time-periodic (LTP) systems, as a special case of LTV systems. Theoretically, we demonstrate that our P-DeePC and P-SPC methods have equivalence control actions as produced from MPC for deterministic LTP systems, under appropriate tuning conditions. As an intermediate step in our theoretical development, we extended certain aspects of behavioral systems theory from LTI systems to LTP/LTV systems. This includes extending Willems’ fundamental lemma to LTP systems and the defining the concepts of order and lag for LTV systems.
In our second piece of work, we proposed a control framework for stochastic LTI systems, namely Stochastic Data-Driven Predictive Control (SDDPC). Our SDDPC method theoretically achieves equivalent control performance to model-based Stochastic MPC, under idealized conditions of appropriate tuning and noise-free offline data. This method, which applies to general linear stochastic state-space systems, serves as an alternative to the data-driven method previously proposed by Pan et al., which also achieved theoretical equivalence to Stochastic MPC but was limited to a narrower class of systems. Beyond the theoretical assumption of noise-free offline data, we performed our SDDPC method in simulations with practical noisy offline data. The simulation results demonstrated that our SDDPC method outperforms benchmark methods, achieving lower cumulative tracking cost and lower rate and amount of constraint violation.
Transition metal doped ceria catalyst prepared by direct precipitation method for thermocatalytic conversion of carbon dioxide via reverse water gas shift
(University of Waterloo, 2025-02-05) Xia, Wenxuan
Since the beginning of the industrial revolution, mankind has utilized large amounts of fossil fuels to obtain energy, which has led to the emission of large amounts of greenhouse gases such as carbon dioxide. How to reduce CO2 and utilize CO2 to obtain high-value products has become a hot topic in today's research. The thermocatalytic reduction of CO2 by using renewable H2 is expected to be a potential solution to these challenges.
In this experiment, the reverse water gas shift (RWGS) reaction of various loaded transition metal doped cerium (MCeO2) catalysts (M = Fe, Co, Ni and Cu) was investigated. The desired catalysts have been synthesized by utilizing the direct precipitation method. The reverse water gas shift reaction has been extensively studied including reaction tests and some characterizations such as X-ray crystallography (XRD), Brunauer Emmett Teller (BET), Temperature Programmed Desorption (TPD), Inductively coupled plasma - optical emission spectrometry (ICP - OES) etc.
In reaction tests, the performance of M-CeO2 was evaluated in terms of conversion and selectivity by varying the temperature (400°C - 600°C). The resulting reaction products were monitored using an on-line infrared analyzer to identify the formation of carbon monoxide (CO), methane (CH4), and unconverted CO2. T-test results show that transition metal doping has a significant effect in enhancing the surface CO2 adsorption and reduction. effects, including high loading of Fe with higher than 56% CO2 conversion and 100% selectivity to CO at 600 °C, Cu with 100% selectivity to CO but lower CO2 conversion, and Co and Ni with significant methanation ability, especially at high loading.
In addition, the structures of the catalysts before and after the reaction were investigated using XRD. The binding strength of CO2 on the doped CeO2 surface was investigated using the programmed temperature rise desorption (TPD) method. The effect of specific surface on CO2 adsorption was investigated using BET.
This experiment explores the effect of different kinds of transition metal-doped cerium catalysts on the reverse water-gas shift (RWGS) reaction, which reduces excess CO2 emissions and also provides an idea for CO2 conversion and utilization.
The Great Migration, Urban Spatial Structure, and Their Economic and Environmental Impacts in the U.S.
(University of Waterloo, 2025-02-05) Shen, Zixing
This thesis investigates the transformative impact of the Great Migration on urban spatial structure in the United States and its subsequent economic and environmental consequences. The Great Migration (1910-1970), marked by significant African American migration from the South to Northern and Western cities, reshaped the demographic and spatial landscape of receiving cities. Yet, the long-term spatial dynamics and their implications remain underexplored, a gap this study aims to address.
Using a novel dataset constructed through the City Clustering Algorithm (CCA), this research redefines historical urban boundaries, creating “synthetic cities” (Syncities) that more accurately reflect urban development from 1900 to 1970. The analysis reveals how demographic shifts influenced the size and shape of cities, with implications for their economic performance and environmental quality. By employing instrumental variable regression and mediation analysis, this thesis identifies the causal pathways through which these spatial transformations affected income levels and air quality, both immediately following the migration and decades later.
The findings highlight both economic growth and environmental challenges linked to urban expansion. Cities with larger and more dispersed urban forms benefited economically in the short term but faced greater environmental degradation over time. These results underscore the importance of urban spatial structure in shaping sustainable development trajectories.
By bridging historical demographic changes with contemporary urban outcomes, this study offers valuable insights for urban planning and policy, demonstrating how historical conditions continue to shape modern challenges. This work also provides a methodological foundation for future research on the interplay between migration, urban form, and sustainability.
This thesis is organized into seven chapters. Chapter 1 introduces the research goals and questions, explaining the importance of the study. Chapter 2 describes how the new urban spatial dataset was created and provides an overview of the key data. Chapter 3 looks at how urban areas in the U.S. changed between 1900 and 1970, providing background for the main analyses. Chapter 4 studies how the Great Migration affected the size and shape of cities, while Chapter 5 examines the economic and environmental effects of these changes. Chapter 6 explains how changes in urban areas connected the Great Migration to economic and environmental outcomes. Finally, Chapter 7 brings the findings together, discusses what they mean for policy, highlights limitations, and suggests ideas for future research. These chapters work together to show how the Great Migration reshaped American cities and what that means for sustainability today.
Investigating the Dynamics of Meandering River Cutoffs: Relationships with Discharge, Land Cover and Spatial Clustering
(University of Waterloo, 2025-02-04) Sun, Letong
As climate change has become one of the major concerns across the globe, investigating the dynamics of meandering river evolution is substantial for urban river management and flood mitigation plans. In recent years, the study on river cutoff has been given lots of attention, as its occurrences and impacts were unpredictable and catastrophic. This study investigates its relationship with high-flow events, land cover and spatial clustering through flood frequency analysis, cutoff ratio criterion and spatial cluster analysis. 1,186 river cutoffs across the United States are located and identified based on Google Earth Imagery. 12 highly sinuous rivers with high cutoff occurrences are then selected and processed through R Studio and ArcGIS. The results show no strong correlation between high-flow events and cutoff occurrence across the study areas. Discharges with an average of approximately eleven-year return period are associated with cutoff occurrences. With the installation of the cutoff ratio in the dataset, it is found that chute cutoffs with higher CRm_m values are likely to occur on land cover types with lower erosion resistance. Neck cutoffs are usually found in floodplains less susceptible to erosion, particularly in undisturbed vegetated areas. Spatial cluster analysis shows that neck cutoffs are significantly clustered at all scales, whereas chute cutoffs exhibit relatively lower clustering tendencies and tend to be more event-driven. Minimizing the random disturbances in the analysis, this study collectively validates the non-random behaviour of cutoff occurrences, which further calls attention to the importance and viability of assessing and predicting cutoff evolution in urban planning and flood management.