eScholar OntarioTech University
Not a member yet
1901 research outputs found
Sort by
Pacing strategy optimization for time trial cyclists with physiological constraints
Get PDFThis thesis investigates the optimization of pacing strategies for time-trial cyclists by integrating physiological constraints with course-specific environmental factors. Two complementary methods are developed: one directly optimizes velocity profiles using total energy expenditure as a proxy for current energy, while the other optimizes power output at discrete course points. Both outperform constant power pacing, producing personalized strategies that adapt to terrain and wind while respecting individual cyclists’ power profiles, physiology, and energy reserves. The first optimization scheme produces slightly better completion times, but impractical power curves that could create dangerous forces, so the second scheme is ideal. Distinct pacing patterns emerge across cyclist phenotypes, underscoring the need for athlete-specific strategies. Potential applications include a prototype training app offering real-time coaching cues. This modular framework supports future integration with advanced physiological models, real-time feedback control, and empirical validation, bridging mathematical optimization and sports science to enhance cycling performance
Joint species distribution modelling of multiple taxonomic groups shows that water quality explains most of the variation in aquatic community composition along a lake trophic gradient
Get PDFSpringer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.A key component of conservation biology is understanding how community composition responds to environmental conditions, including stressors (e.g., eutrophication). However, this is complicated by the accompanying shifts in taxa co-occurrence due to taxa interactions (e.g., competition, predation). Thus, we need to consider both taxa co-occurrence and abiotic conditions simultaneously to understand the relative influence of these factors on community composition. In this study, we aimed to elucidate the relative role of water quality and taxa co-occurrence in determining phytoplankton, zooplankton, and macroinvertebrate communities in a set of lakes varying in trophic status from oligotrophic to meso-eutrophic. We built joint species distribution models (JSDMs) using a generalized linear latent variable modelling approach to isolate the influence of abiotic variables and taxa co-occurrence for each level of the food web. We found that water quality explained most of the variation in community composition for all groups, especially for zooplankton (78%) and macroinvertebrate (62%) communities. Additionally, total phosphorus was a significant driver of many taxa in these groups. These findings indicate a strong “bottom-up” control of the lower aquatic food web in these lakes and signify an important role of phosphorus across a distinct trophic gradient. Overall, we demonstrate the effectiveness of JSDMs for determining the relative impacts of abiotic variables on community composition which can provide key insights for lake management.NSERC Discovery Grant and Great Lakes Local Action Fun
Automated goal model generation from user stories using Large Language Models
Get PDFIn agile software development, user stories capture stakeholder needs but often fail to represent complex requirement relationships. Goal modeling addresses this by linking high-level goals to specific requirements, but manually transforming user stories into goal models is challenging. This research explores using Large Language Models (LLMs) to automate goal model generation through multi-step prompt engineering. LLMs extract intentional elements—goals, tasks, actors, and resources—and generate Goal-oriented Requirements Language (GRL) models compatible with tools like jUCMNav.
The study evaluates GPT-4, Llama, and Cohere, focusing on syntactic completeness and correctness. GPT-4 outperforms others, particularly in extracting implicit goals and soft goals, but struggles with intricate relationships like means-end and contribution links. Despite limitations, LLMs show promise in automating labor-intensive aspects of goal modeling, making the process more efficient. This research highlights their potential to support requirements engineers and integrate goal modeling into agile workflows
Experimental investigation of a novel sonic hydrogen generator
Get PDFThis thesis presents an investigation of hydrogen generation using ultrasound technology. The experimental setup uses a newly developed integrated system employing ultrasound technology. The system operates at 40 kHz with a power output of 100 W. Various parameters such as temperature and isopropyl alcohol concentration are used to evaluate the hydrogen production process. Presented results indicate a variability in hydrogen production rates under different experimental conditions. For instance, distilled water at 25°C yields a hydrogen production rate of 0.059 μmol/min, which increases to 0.109 μmol/min at 70°C, demonstrating the positive impact of temperature on hydrogen yield. Other water resources such as tap water, lake water and wastewater also show improved hydrogen production rates at higher temperatures. The experimental datasets indicate that distilled water consistently produces the highest hydrogen yield compared to other water types. In terms of alcohol concentration, a 5% concentration produces the highest hydrogen production rate of 0.356 μmol/min. Additionally, CO₂ injection into the reactor enhances hydrogen production by up to 30%, whereas air injection leads to a hydrogen concentration of 83 ppm after 60 minutes. Compared to conventional electrolysis, ultrasound-assisted electrolysis increases hydrogen production from 0.0486 μmol/min to 0.092 μmol/min. The conceptual part of the research demonstrates the integration of this hydrogen production technique into renewable energy-based multigeneration systems, denoted as System 1,2 and 3. System 1 produces 12,839.5 kW of electrical power and 32.92 kg/h of hydrogen, with an energy efficiency increase from 37.68% to 55.32% by varying the solar tower's output temperature from 800 K to 1300 K. System 2 results show an energy efficiency of 83.28% and an exergy efficiency of 58.71%. Finally, system 3 demonstrates power generation using heliostats for solar concentration with energy and exergy efficiencies of 58.28% and 76.75%, respectively
Implicit pen annotation assisted by Large Language Models
No full textIn our modern society, integrating humans and computer systems has transformed everyday tasks, including reading, annotating, and reviewing documents. Annotating documents is an age-old practice that involves adding markings and notes to engage with the material. Although this process is crucial for understanding texts and facilitating collaborative communication, it has not evolved significantly over the years. Tedious and repetitive workflows in current digital annotation tools suggest missed opportunities for more intelligent, adaptive design. This work introduces AnnotateGPT, a document annotation tool with a digital pen. It leverages a Large Language Model (LLM) (1) to infer the underlying purposes of the user’s annotations and (2) automatically generates annotations with the same purpose throughout the document. AnnotateGPT aims to alleviate the burdens of manual annotation, enabling users to focus on tasks that require critical expertise
Real-time process monitoring of Spark-Assisted Chemical Engraving (SACE) machine with data-driven techniques
Get PDFSpark-Assisted Chemical Engraving (SACE) is a promising technique for glass microfabrication, offering good surface quality and machining speed. However, inherent uncertainties arising from gas film instability and the stochastic nature of the process lead to inconsistencies and reduced repeatability in machining quality. Addressing these challenges requires precise online control of the process to improve machining efficiency—an aspect not thoroughly explored in previous studies. This dissertation achieves the groundwork for process control by developing a real-time SACE process monitoring methodology for anomaly detection in glass microchannel fabrication.
The first step in the methodology introduces a characterization algorithm that segments the machining current into formation, discharging, and silent phases to extract distinct process signatures. A deep learning-based time series classification model was trained for this segmentation using two architectures—Temporal Convolutional Network (TCN) and Long Short-Term Memory (LSTM)—achieving classification accuracies of 97.18% and 96.44%, respectively. The second step establishes a correlation between extracted process signatures and SACE machining quality through a decision tree-based supervised learning approach, demonstrating an 88.88% sensitivity in classifying machining quality into three categories: uniform, non-uniform, and fractured. This approach provides an alternative to the challenges of direct quality monitoring.
The third step integrates the anomaly detection algorithm into a real-time environment, leveraging a preemptive multitasking environment to prevent overrun errors while operating at a high frequency of 62.5 kHz for spark detection.
This work establishes a framework for SACE real-time data-driven monitoring, enabling automated defect detection. By demonstrating its effectiveness in real-time process control, this study lays the groundwork for future advancements in glass micromanufacturing, allowing for the control of process parameters to compensate for uncertainties in real-time
Developing and evaluating “paraben-free & me”: an educational toolkit that targets parabens exposure reduction from personal care products
Get PDFWomen use approximately 13 personal care products (PCPs) daily, exposing them to 114 harmful chemicals, such as parabens, every day. Found in cosmetics and hair products, parabens are linked to increased risks of breast cancer and endocrine disorders. Despite these risks, awareness and risk perceptions about parabens remain low among women. Educational toolkits have the potential to effectively raise awareness and promote positive behaviours regarding environmental health risks. This study aims to create an evidence-informed digital educational toolkit to reduce women's paraben exposure. The research includes three phases: (1) a systematic review identifying the characteristics of existing toolkits and the data collection methods to evaluate them; (2) developing the toolkit using the IDEAS framework and the Health Belief Model; and (3) evaluating the toolkit's effectiveness through a four-week randomized controlled trial among female students in Ontario. Outcomes assessed will include participants’ knowledge, risk perception, beliefs, and paraben-free behaviours
Method optimization and monitoring of Campylobacter jejuni and Helicobacter pylori in municipal wastewater
Get PDFCampylobacter jejuni and Helicobacter pylori are two of the most infectious bacteria worldwide. Despite this, reporting of these bacteria severely underrepresent their true burden within populations. Therefore, the goal of this research was to (1) optimize a molecular-based method to detect H. pylori and C. jejuni in wastewater, and (2) examine population-level trends of these bacteria in each wastewater catchment. I created a method that can reliably extract DNA for the target bacteria and qPCR assays to monitor their load in wastewater. I validated the method using wastewater samples from four Ontario municipalities over a year-long monitoring period. The trends showed H. pylori was constantly present in the wastewater with no seasonal changes in abundance, while C. jejuni was more seasonal, correlating with clinically reported cases for most of the municipalities examined. These results confirm the efficacy of using qPCR-based methods for community-level monitoring of under-reported bacterial pathogens in wastewater
Exploring crime, deviance and community among Sri Lankan Tamil youth in Canada
Get PDFThe Sri Lankan Tamil community remains significantly understudied within Western literature on youth crime and deviance, despite being one of the largest diaspora populations in Canada. Much of the existing research generalizes offending patterns under the broad label of "South Asian youth," often neglecting the distinct cultural, historical backgrounds and experiences of specific groups within this category. This thesis seeks to address this gap by examining community, crime and deviance among second- generation Sri Lankan Tamil youth in Canada. Through semi-structured interviews with members of the Tamil diaspora, this study uncovers key challenges faced by Tamil youth, including strained family dynamics and difficulties navigating community ties. The findings offer a novel contribution by applying an intersectional lens that bridges diaspora studies and criminology, offering insights into the specific experiences shaping crime and deviance among diasporic youth
Chemical Valley: media representations of an inconvenient truth and a comforting lie
Get PDFThis thesis examines media representations of pollution in Sarnia’s Chemical Valley, focusing on how these portrayals impact marginalized communities, particularly the Aamjiwnaang First Nation. Through an ultra-realist lens, this research explores how dominant narratives shape public understanding of environmental harm, maintaining neoliberal ideologies that prioritize industrial interests. Using content analysis of articles from the Sarnia Observer and the Sarnia Journal, key concepts such as fetishistic disavowal, special liberty, and enlightened catastrophism reveal underlying ideologies contributing to environmental injustice. This study also highlights connections between media ownership and the framing of pollution, underscoring how these narratives reinforce existing power structures and limit calls for policy change. Findings suggest that media portrayals in Chemical Valley serve neoliberal capitalist interests, perpetuating environmental inequity. By critically engaging with media narratives, this research aims to inform strategies for environmental justice and advocate for more equitable representation in media discourse