Student Publications
Engineering Graduate Student Publication Showcase
Welcome to a collection of published papers by our Engineering graduate students. Here, you can explore the research and findings from our students as they contribute to various engineering fields.
Each paper reflects their hard work and dedication, showcasing a range of topics. We invite you to browse through their publications and see the valuable insights they’re bringing to the Engineering community.
Join us in recognizing their hard work and the impact they are making in academia and industry!
Winter 2026 Publications
ESE
HybridRDG: Zero-Shot ASD Biomarker Detection from Multi-Paradigm EEG via Hybrid Deep-Riemannian Domain Generalization
Domain Generalization (DG) research has progressed rapidly on image benchmarks, yet remains largely unexplored in biosignals, where domain shift is geometric rather than stylistic. In an electroencephalogram (EEG), each subject constitutes a distinct domain characterized by a unique covariance structure and event-related potential timing, with no semantic anchors and typically fewer than 50 labelled subjects. We introduce Hybrid Deep-Riemannian Domain Generalization (HybridRDG), a zero-shot DG framework for clinical biomarker detection that identifies carriers of an autism-linked genetic variant from multi-paradigm EEG without accessing any target-subject data during training. HybridRDG fuses two complementary branches: an adversarially regularized EEGNet that captures temporal waveform morphology, and a Riemannian geometry branch that captures spatial covariance structure via paradigm-specific multi-window decomposition, Ledoit-Wolf shrinkage, and Log-Euclidean alignment, all fitted exclusively on source subjects. Evaluated against ten baselines spanning Riemannian, deep learning, and DG methods across five paradigms under strict subject-held-out protocols, HybridRDG achieves the highest subject-level balanced accuracy in four of five settings. These results indicate that geometric covariance structure, not merely Euclidean feature alignment, is central to zero-shot EEG generalization.
Ayesha Anzer, Abdul Bais; Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 2026, pp. 3151-3160
https://openaccess.thecvf.com/content/CVPR2026W/DG-EBF/html/Anzer_HybridRDG_Zero-Shot_ASD_Biomarker_Detection_from_Multi-Paradigm_EEG_via_Hybrid_CVPRW_2026_paper.htmlEVSE
Hybrid AI-hydrologic flood modeling in prairie agricultural watersheds
Flood simulation in prairie agricultural regions is challenging due to low-relief topography, extensive surface storage, and spatially heterogeneous rainfall-runoff response. This study presents an integrated high-resolution flood modeling framework that combines GIS-based spatial analysis, flood frequency analysis of the observed annual peak flow record, the physically based HEC-HMS hydrological model, and a data-driven Adaptive Neuro-Fuzzy Inference System (ANFIS) optimized by the Imperialist Competitive Algorithm (ICA). The ANFIS-ICA model was developed as a computationally efficient surrogate to emulate HEC-HMS-simulated hydrographs at 3-minute temporal resolution under 10-, 50-, and 100-year return periods. The framework was applied to 16 sub-basins in Maple Creek, a representative prairie region in Saskatchewan, Canada. Based on spatial datasets processed in QGIS, the HEC-HMS model was first calibrated against observed peak discharges from selected historical flood events, and then used to generate training data for the ANFIS-ICA model incorporating rainfall and lagged flow inputs. Model evaluation showed excellent agreement between simulated and reference hydrographs, with NSE and R2 values exceeding 0.95 and RMSE below 0.02 m3/s across training, validation, and testing datasets, indicating strong generalization and high surrogate fidelity. Hydrograph analysis revealed spatially variable runoff behavior, characterized by relatively rapid flow propagation in sub-basins with limited surface storage and attenuated responses in sub-basins exhibiting stronger surface storage effects, reflecting the heterogeneous hydrological nature of prairie environments. Peak flows were classified into flood risk categories, enabling sub-basin-level vulnerability mapping to inform targeted flood mitigation, early warning, and adaptive agricultural management. The proposed framework provides a validated and potentially transferable tool for efficient, localized flood risk assessment in data-sparse prairie agroecosystems. However, its performance remains dependent on the accuracy of HEC-HMS simulations and the availability of observed peak flow data, and full hydrograph validation is constrained by the lack of high-temporal-resolution discharge records.
Hassanjabbar, A., Zhou, X., Han, T., McCullum, K., Wu, P. (2026). Hybrid AI-hydrologic flood modeling in prairie agricultural watersheds. Earth Systems and Environment.
https://link.springer.com/article/10.1007/s41748-026-01175-7ISE
Enhancing Railway Resilience Across Canada Through IoT Applications
The railway industry is growing fast with the emergence of technologies, and the Internet of Things (IoT) is one of those renaissance technologies. Many countries across the world are exploiting opportunities of IoTs for increasing operational efficiency and making railway a safer mode of transport. IoTs have immersive applications for the railway network in Canada. The resilience of Canadian railway could significantly be impacted by exploiting the opportunities of IoTs in condition monitoring of infrastructure, rolling stocks, maintenance practices, signalling system etc. This paper aims to find IoTs’ significant applications in condition monitoring of railway assets, IoT’s implications in security and surveillance systems, rolling stock maintenance, Passenger Information System (PIS), Freight Information System (FIS), Automatic Train Control Systems with a view to make a resilient railway network across Canada. To extract the opportunities of IoTs and sensors in railway industry, this paper makes a systematic review of journals articles, e-books, book chapters, conference papers and different registers. The study project reviewed several peer reviewed articles and scientific databases and found significant applications of IoTs which can contribute to enhancing the resilience of many sections of Canadian Railways. This paper may encourage future researchers to know more about IoTs’ applications in railway industry of Canada.
Islam, R., & Kabir, G. (2026). Enhancing Railway Resilience Across Canada Through IoT Applications. In CSCE Winnipeg 2025 (pp. 532-541). University of Toronto Press.
https://utppublishing.com/doi/abs/10.3138/csce-2025-0278PSE
Comparative Performance Evaluation of Dendritic and Regional Wormhole Networks During a Solvent-Steam Huff-n-Puff Process in a Post-CHOPS Reservoir
In an unconsolidated heavy oil reservoir, cold heavy oil production with sand (CHOPS) has been the primary production method with good performance; however, the in-situ induced wormholes (i.e., dendritic and regional patterns) due to sand particle migration introduce large constraints for the potential enhanced oil recovery (EOR) processes. In this work, an integrated framework has been proposed to optimize performance of solvent-steam injection in a post- CHOPS reservoir where both dendritic and regional wormhole networks are generated and compared. More specifically, a dendritic wormhole network, obtained from combining rate transient analysis (RTA) and a newly developed pressure-gradient-based (PGB) sand failure criterion, was transformed and embedded into a reservoir geologic model and calibrated with the measured pressure and production profiles. By selecting thermal energy, injection rate, and bottomhole pressure as control variables, a genetic algorithm (GA) is integrated with orthogonal array (OA) and Tabu search to maximize the net present value (NPV) and thus obtain optimal oil recovery. The results indicate that the dendritic wormhole network demonstrates superior performance with respect to oil recovery and NPV compared to the regional wormhole network. For the optimal huff-n-puff (HnP) strategy for each of the six cycles, the dendritic wormhole network achieved an oil recover factor of 6.853% and a NPV of 31.739 × 106 C$, which are respectively 3.5 times and 5.4 times of the regional wormhole network. Such differences mainly arise from the different morphologies of the dendritic and regional wormhole networks with the former being more physically realistic. Foamy oil flow has shown its positive impact on both oil recovery factor and NPV, which is particularly evident in the dendritic wormhole model due to the presence of high permeability channels. In addition to the computational costs, key challenges include uncertainties in characterizing wormhole geometry and assigning its petrophysical properties in a given reservoir. Future work will focus on dynamically updating wormhole characteristics conditioned to field monitoring and production data, quantifying uncertainties associated with wormhole geometries and the formation petrophysical properties, and extending the framework to multi-well patterns and other well configurations.
Hou, S., Yang, D., and Jiang, L. (2026). Comparative Performance Evaluation of Dendritic and Regional Wormhole Networks During a Solvent-Steam Huff-n-Puff Process in a Post-CHOPS Reservoir. Geoenergy Science and Engineering, 214531.
https://doi.org/10.1016/j.geoen.2026.214531PSEN
From Biomass Waste to Green Fuel: Biochar-Based Catalysts for Hydrogen Production
With the increasing demand for energy given by the effects of extreme weathers in the last years, the need to expand the access to renewable energy such as green hydrogen has become a priority in current research. However, one of the main challenges for hydrogen production is the elevated cost of catalysts due to the consumption and lack of availability of rare metals. To reinforce sustainability, biochar, a carbon-rich material has emerged with huge potential. Its properties such as a high surface area and abundant functional groups facilitate catalyst adsorption and the dispersion of active sites, besides the mineral content and surface chemistry tunability, allow the activation and metal impregnation to improve hydrogen production. Considering these characteristics, this paper will highlight all the potential of biochar as a catalyst and catalyst support, the current advances identifying biochar as catalyst in hydrogen production, and the key characteristics that make it adequate in these applications. Finally, the remaining challenges and limitations are described, providing a perspective on future opportunities and research directions.
Rubiano, K. M., Jilani, A., & Ibrahim, H. (2026). From Biomass Waste to Green Fuel: Biochar-Based Catalysts for Hydrogen Production. Energies, 19(4), 1087.
https://www.mdpi.com/1996-1073/19/4/1087