Doctoral Theses and Dissertations

Permanent URI for this collectionhttps://hdl.handle.net/10294/2900

Browse

Now showing 1 - 20 of 29

Open Access
Bayesian probabilistic projections of future climate over Canada based on the RCM ensemble
(Faculty of Graduate Studies and Research, University of Regina, 2024-03) Song, Tangnyu; Huang, Guohe (Gordon); Veawab, Amy; Zhu, Hua; Deng, Dianliang; Li, Jonathan
In this research, a series of approaches are proposed to address the challenges in generating robust probabilistic projections of climatic variables (e.g., temperature, precipitation, and Intensity-Duration-Frequency curves) and analyzing the associated uncertainties based on the Regional Climate Model (RCM) ensemble. The proposed approaches have been applied to Canada for demonstrating their effectiveness. Specifically, a new discriminant-Bayesian Model Averaging (BMA) ensemble climate modeling (DBEC) approach is developed to help address the neglected spatiotemporal variations of model biases. Through the proposed method, the BMA weights are used as prior distributions to drive the Bayesian discriminant analysis in order to generate refined weights for individual ensemble models according to their spatiallyand temporally-clustered performance. The results suggest that the DBEC approach can improve both the accuracy and reliability of ensemble projections to some extent, especially in winter and Arctic regions. The probabilistic projections of temperature for three future periods under two emission scenarios are then obtained through the proposed DBEC model. The results indicate that comparatively larger temperature increases can be observed in Arctic regions. In addition, the magnitude of uncertainties is found to be negatively correlated to the elevation. Then a new multi-dimensional discriminant-BMA ensemble approach (MDBE) is developed to quantitively characterize the relationships between the modeling performances and climatic conditions. Through the comparative assessments of the proposed approach against three other ensemble methods, its effectiveness in generating the probabilistic projections of annual and seasonal precipitation over Canada has been illustrated. In detail, the R2 and percentage coverage will increase up to 0.15 (from 0.52 to 0.67) and 20% (from 60% ~ 80%), respectively. The generated projections suggest that significant precipitation increases are observed in future periods, especially in the Arctic regions. The warming climate could be the primary reason for such increases. Moreover, the intensified atmospheric radiative cooling is also a possible explanation for the winter precipitation increase. Finally, a new CDF-distance-based method is proposed to generate ensemble projections of IDF curves over Canada. Compared with the traditional ensemble methods, the proposed CDF-distance-based ensemble approach depends less on the simulated accuracy of annual maximum precipitation time series. Consequently, it can improve both the accuracy and reliability of the probabilistic projections in IDF curves. The proposed method has been applied to Canada for assessing the future changes of the IDF curves. The results suggest that the upward shifts of the IDF curves under all return periods are observed under changing climate conditions. Moreover, the percentage changes of precipitation intensities increase with return periods.
Open Access
Investigating the role of water and nanoparticles in the performance of CO2 and ethane-based cyclic solvent injection process for heavy oil recovery
(Faculty of Graduate Studies and Research, University of Regina, 2024-04) Li, Yishu; Zeng, Fanhua (Bill); Gu, Yongan (Peter); Jia, Na (Jenna); Du, Zhongwei (David); Qing, Hairuo; Zhang, Yin
Cyclic solvent injection (CSI) process is a promising method for enhancing heavy oil recovery in thin or deep heavy oil reservoirs. Foamy oil flow is a major phenomenon of the CSI process, but the impact of foam stabilizers like nanoparticle and water on its performance has not been well understood. Therefore, an in-depth understanding of waterflooding and nanoparticle affecting CSI performance is a crucial step toward further enhancing oil recovery. In addition, most studies only focus on one method. In fact, in order to develop an oil reservoir cost-effectively and efficiently, a variety of methods have to be applied in sequence. In this study, a well-designed experimental investigation was conducted to determine the role of water and nanoparticle in CSI process. First, several foundemental live oil depletion tests were applied to investigate the feasibility of using foamy oil additives in the CSI process, determine the optimum concentration of the additives, and evaluate the best operation conditions on the performance of the CO2 based CSI process. Second, three CO2 based CSI tests were performed using a cylindrical sandpack at different injection pressure. Test 1 conducted a normal CO2 based CSI process for comparison. Test 2 applied a waterflooding process and followed by a CO2 based CSI process. Test 3 was a hybrid process consisting of the sequence: CSI-waterflooding-CSI-Nanoparticle solution flooding- CSI. Third, two ethane based CSI tests (A blank test and a similar combination test) were applied in order to verify and compare CO2 based CSI process. Experimental parameters and results were monitored and recorded. Comprehensive data analytics were performed to examine the effect of water and nanoparticle and identify enhancing mechanisms for the CSI process. Experiment results indicate that using nanoparticles as foam stabilizers in CO2 live oil pressure depletion tests can enhance oil recovery while reducing gas recovery, with higher nanoparticle concentrations resulting in even greater oil recovery. An optimal depletion rate of 6 kPa/min was identified, and the waterflooding process had a positive impact on the performance of both CO2 and ethane based CSI methods, improving mass transfer by expanding gas/oil contact areas. The nanoparticle solution flooding applied before CSI process were able to effectively stabilize foamy oil even at high water saturation levels. The integration of CSI, water flooding, and nanoparticle solution flooding yields an overall oil recovery factor of 69.5% for the combined CO2 based hybrid EOR process and 76.1% for the ethane based hybrid EOR process. Simulation results show that the dispersed gas model exhibits strong alignment between oil and gas production data and pressure distribution in CO2 based live oil pressure depletion tests. For CO2 based CSI process, by incorporating a modified foamy oil model with separate sets of relative permeability curves for injection and production stages, and employing the dispersed gas model during production stage, a robust history-matching of oil, gas, and water production data is achieved simultaneously. In the Lloydminster area, thousands of wells have a water cut exceeding 90% after the waterflooding process and are subsequently abandoned due to the absence of an effective way to continue development. This study meticulously examines potential enhancements for the CSI process, which involve utilizing foam stabilizers in combination with synergistic water flooding methods. The findings of this study provide practical solutions to address the technical challenges encountered during the subsequent development of heavy oil wells with high initial water saturation. Additionally, the study highlights the promising prospects of combination approaches in enhancing heavy oil recovery.
Open Access
Nanocellulose-based materials for sustainable soil remediation and water purification
(Faculty of Graduate Studies and Research, University of Regina, 2024-01) Yin, Jianan; Huang, Guohe (Gordon); Ng, Tsun Wai Kelvin; Zhu, Hua; Deng, Dianliang; Zeng, Hongbo
Soil and water pollution are intricately linked environmental issues that have gained significant global attention due to their adverse effects on ecosystems, public health, and overall sustainability. In this dissertation, the pressing need for sustainable pollutant treatment using ecofriendly and biodegradable nanocellulose (CNC) biopolymers is addressed. This research focuses on the development of CNC-based materials, characterization of their adsorption behaviors, evaluation of CNC-mediated algal toxicity, and exploration of the application of the these materials in sustainable soil remediation and water purification. In the first part, the utilization of CNC nanofluid as an eco-friendly agent for the remediation of phenanthrene (PHE) contaminated soil is proposed. This marks the first exploration of CNC nanofluid’s effectiveness in mobilizing PHE in soil, with a focus on the influence of environmental factors. The findings demonstrate the critical role of temperature and ionic strength in PHE removal. This study also reveals the interactions between CNC and soil components, elucidating the primary PHE removal mechanism. Additionally, our research highlights the detoxification effect of CNC nanofluid on PHE-contaminated soil, providing a promising alternative for site remediation. In the second part, inspired by the hierarchical fibrous structure and antibacterial properties of natural silkworm cocoons, a guanidine-functionalized sericin/nanocellulose aerogel (GSNA) is designed for application in the rapid removal of both bacteria and heavy metals from water. The grafted polyhexamethylene biguanide (PHMB) endows the biomimetic aerogel with exceptional bactericidal activity. The incorporated sericin protein brings abundant surface functional groups for heavy metal complexation. Moreover, this study provides in-depth insights into the bonding mechanism between metal ions and GSNA through density functional theory (DFT)-assisted X- ray absorption near edge structure (XANES) analysis, representing a pioneering effort in elucidating the adsorption mechanism of heavy metals within nanocellulose-based aerogels. In the third part, a recyclable sericin/nanocellulose composite aerogel (SNCA) is introduced for efficient tetrabromobisphenol A (TBBPA) removal from water. The developed SNCA exhibits exceptional compressibility, hydrophilicity, and adsorption capacity. In addition, the SNCA can be easily recycled through a simple compression method, demonstrating remarkable reusability even after 10 regeneration cycles. Furthermore, toxicity evaluations reveal that SNCA effectively mitigates the adverse effects of TBBPA on freshwater algae, emphasizing its environmental friendliness. DFT calculations provide insights into the TBBPA adsorption mechanism, indicating the involvement of hydrogen bonding and electron donor-acceptor interactions. In the fourth part, the investigation reveals that the presence of CNC significantly reduce ZnO NP aggregation, enhancing bioavailability and toxicity to freshwater algae. The interaction of ZnO NPs with CNCs leads to envelopment of algal cells and induces oxidative stress, affecting membrane lipids and antioxidant enzyme activity. The introduction of CNCs enhances intracellular transportation of Zn ions, influencing substance flow between algae cells and the environment. This study advances our understanding of the combined effects of multiple nanomaterials on aquatic organisms, allowing for the identification of composite risks. In summary, this research explores novel and sustainable approaches for pollutant treatment and environmental remediation, utilizing biodegradable nanocellulose materials. These efforts contribute to reducing environmental impact and promoting eco-friendly solutions for soil and water purification.
Open Access
Experimental Evaluation of Innovative Catalytic Heat Exchangers for Energy-efficient Amine-based Post-combustion CO₂ Capture Processes
(Faculty of Graduate Studies and Research, University of Regina, 2025-04) Yang, Congning; Tontiwachwuthikul, Paitoon; Chan, Christine; Idem, Raphael; Choi, Phillip; Sema, Teerawat; Raina-Fulton, Renata; Ricardez-Sandoval, Luis
This PhD research focused on enhancing CO2 desorption performance in post-combustion carbon capture processes by developing and optimizing catalytic heat exchangers. The guiding principle was to address challenges, including operational complexity and high energy consumption, while minimizing costly modifications to existing piping and infrastructure in pilot plants. The feasibility of novel aqueous piperazine-based biphasic solvents was initially investigated to reduce solvent flow rates and optimize heat exchanger size. Although these solvents demonstrated promising absorption and desorption performance, challenges such as high viscosities, suboptimal phase split ratios and low amine concentrations in the rich phase limited their applicability in the catalytic heat exchanger system. An innovative agitated jacket vessel with a coil heat exchanger (JVC-EX) was developed and experimentally validated. Compared to the conventional fixed catalyst bed desorber, the JVC-EX using benchmark MEA solvent and solid acid catalyst HZSM-5 achieved approximately 30% catalytic enhancement, a 50% reduction in catalyst demand, and a 22% decrease in energy consumption while maintaining excellent operational stability and flexibility. However, concerns emerged regarding catalyst durability due to mechanical stirrer-induced attrition, highlighting the need for further mechanical optimization. To address this limitation, a spouted bed and jet-flow catalytic heat exchanger (SBJ-EX) was introduced as a non-agitated alternative. The SBJ-EX demonstrated a 70% improvement in heat transfer efficiency compared to traditional plate heat exchangers and delivered excellent CO2 desorption performance at lower operating temperatures. Its spouted design effectively minimized catalyst attrition, ensured system stability, and enabled faster catalyst replacement, significantly reducing maintenance downtime. Both catalytic heat exchangers showed strong adaptability for integrating existing and new industrial-scale carbon capture systems. Overall, this thesis provided valuable insights into the design, operation, and optimization of novel catalytic heat exchangers, emphasizing their potential to drive the adoption of catalysts in commercial-scale carbon capture applications.
Open Access
A security risk assessment framework for IoT systems
(Faculty of Graduate Studies and Research, University of Regina, 2024-12) Waqdan, Mofareh Abdullah S; Mouhoub, Malek; Louafi, Habib; Shahriar, Nashid; Hepting, Daryl; Uddin, Md. Sami; Al-Anbagi, Irfan; Amamra, Abdelfattah
The emergence and growth of the Internet of Things (IoT) have changed how we live and interact with technology. The seamless integration of connected devices, from household to industrial equipment, has brought about a new era of interconnectedness. However, this rapid expansion of the IoT also introduces new security concerns that need to be assessed. Assessing the security risks associated with deploying and using this technology is crucial. Consequently, organizations need a risk assessment framework that helps identify, evaluate, and manage the risks of IoT, including data privacy and confidentiality, system integrity, availability, and performance. The stateof- the-art has been given significant attention to security risk assessment in traditional cybersecurity with powerful computer systems, but the challenges of deploying IoT devices and their associated vulnerabilities have been overlooked. In this thesis, we first present a novel IoT security risk assessment framework for the healthcare environment, in which we have improved upon existing methodologies. The proposed framework dynamically calculates the risk score for different device profiles, considering their population and other parameters, such as network protocols, device heterogeneity, device security updates, device physical security status, device history status, layer history status, and device criticality. Second, we present a customizable framework for assessing the security risk of deploying and utilizing IoT devices in various environments. We dynamically calculate risk scores for different devices, considering their importance to the system and their vulnerabilities, among other parameters. The customizable framework considers the important parameters of the devices, their vulnerabilities, and how they impact the overall risk assessment. The importance of these devices and the severity of vulnerabilities are incorporated in the framework using the well-known Multi-Attribute Decision Making (MADM) methods, namely, Simple Additive Weighting (SAW) and Weighting Product (WP). Finally, the risk is assessed on a setup comprised of IoT devices widely deployed in healthcare systems, such as emergency rooms.
Open Access
Determinants of food choice: The role of nudging, affective forecasting, habitual behaviour, and values
(Faculty of Graduate Studies and Research, University of Regina, 2025-01) Wallace, Jamie Charles Terrence; Arbuthnott, Katherine; Carleton, Nicholas; Pennycook, Gordon; Mahani, Akram; Forestell, Catherine A.
Modern eating habits have created a tremendous burden of disease in Canada and other western countries. There is a large research literature that has investigated interventions for improving dietary health in the population, including work that focuses on rational or conscious factors (e.g., the provision of health information) and heuristic factors (e.g., social norms and priming healthier food choices). More broadly, researchers have worked to identify determinants of food choice to better understand why people select particular foods, resulting in an array of known factors that in some way predict food choice. The current research was designed to examine and compare empirically supported predictors to inform future interventions and to test a possible priming intervention in an online meal selection context that presents both healthy and unhealthy meal options to participants. Participants were exposed to one of two primes, or a control condition, and then made either a whole or processed food meal selection. Participants also rated their affect towards the foods and completed questionnaires that measured food choice values and meanings, habitual behaviour, typical meal sources, and demographics. Results indicated that affect ratings and familiarity were the strongest predictors of food choice, although other meanings and values, such as health values and habitual behaviours were also important. Priming did not influence meal selection in the current study. The current results suggest that interventions focusing on developing positive feelings, familiarity, and habitual behaviour towards healthy whole foods are likely to be more successful than interventions focusing on rational factors; however, values and meanings were also statistically significant predictors of whole food meal selections and may also be useful for improving food choice.
Open Access
‘My spirit is fed’: Exploring land-based, culturally appropriate active living strategies to facilitate holistic well-being among on-reserved youth: A Smart Platform study
(Faculty of Graduate Studies and Research, University of Regina, 2025-04) Walker, Susannah Lynn; Zarzeczny, Amy; Katapally, Tarun; Dupeyron, Bruno; Coates, Ken; Klest, Bridget; Green, Jacquie
A constructivist grounded theory approach with an Indigenous lens was used for a qualitative analysis of questions on culture, mental health, physical activity, and land-based activities. This analysis was undertaken as a part of the Smart Platform: Smart Indigenous Youth (SIY) project. The goals of the SIY were to increase physical activity using a cultural land-based active living intervention along with a technological component through an app. This thesis focused on the impact of the intervention on the mental health of on-reserve Indigenous youth in southern Saskatchewan; eleven on-reserve Indigenous youth shared their perceptions in focus groups before and after participating in the land-based intervention and thematic data analysis was performed. Themes included keeping culture going, community and cultural aspects of physical activity, and the importance of schools as key locations for reconnecting with culture. A theory of Indigenous identity was developed to provide insight into the complicated aspects in reconnecting with culture, especially the pressure, obligation, and responsibility that Indigenous youth feel towards passing on Indigenous culture. A notable finding was the improvement in mental health reported by participants after participating in the land-based intervention. Policy recommendations include the importance of early childhood access to cultural programs, the necessity of a mental health component as a part of land-based programs, and the need for inclusivity in land-based program teachings.
Open Access
Dielectric characterization of materials and lossy filter design using reflected group delay
(Faculty of Graduate Studies and Research, University of Regina, 2024-12) Walia, Gaurav; Laforge, Paul; Wang, Zhanle (Gerald); Azam, Shahid; Teymurazyan, Aram; Zarifi, Mohammad Hossein
Filters play a significant role in different types of communication systems such as radars, cellular mobile and satellite communication. They are helpful in improving the performance of a communication system by restricting the transmission to the intended frequency band and rejecting the interfering signals from outside. Filters are expected to provide distortion free transmission to the signals in the passband and thus require flat in-band response and an adequate amount of out-of-band rejection. Filter technologies employing coaxial, waveguide and dielectric resonators can meet these requirements but at the cost of large size. Low loss in these filter technologies can be identified by their high quality factor (Q-factor). Higher insertion loss (lower Q-factor) in the filter response is acceptable based on the communication system using that filter and its position in that system. Filter design must deal with the electrical and physical specifications based on the required filter response. The concept of lossy filter is based on the fact that loss in a filter is distributed among the various resonators in a way that helps to achieve flat in-band response. This is achieved at the cost of degraded insertion loss. In this way, lossy filters provide an effective solution to the applications that can afford low Q-factor and higher insertion loss. Resonator is a building block in a filter network and a good understanding of resonators is quite useful in developing an insight of a filter network. Resonators also have a good deal of applications in methods used to perform dielectric characterization of materials. Keeping this connection of resonators and filter in view, the initial research presented in this thesis is focused on lossy resonators. This was helpful in developing some methods for dielectric characterization of materials. In addition to that, the research carried out using lossy resonators has also been helpful in understanding the behavior of resonators in a filter network with the change in amount of loss. The thesis provides a detailed discussion on methods developed to determine the dielectric properties of materials using the reflected group delay of lossy resonators. Methods of dielectric characterization proposed in the thesis can be categorized as methods using an overcoupled coaxial resonator for testing the materials filled inside it and methods using a capacitively coupled microstrip circuit for testing the microstrip substrates. Mathematical models for these methods are based on the reflected group delay of an overcoupled lossy resonator and can be applied in a procedural manner to extract the dielectric constant and loss tangent of the material under test. The errors in extracting dielectric constant and loss tangent are 2%-5% and 30%-100% respectively and depend on the resonator type, material under test, and test setup parameters. The methods are validated through the characterization of various materials using both the coaxial and microstrip resonators. A method of lossy filter design using reflected group delay is also proposed in this thesis. The research presented in this part of thesis describes the effect of decrease in Q-factor on the in-band response in terms of insertion loss and return loss. It also explains the effect of loss on the group delay of a filter network and introduces the use of negative group delay in lossy filter design. The proposed method tends to recover the loss of in-band flatness due to decrease in Q-factor through resistive cross coupling. A circuit model for microstrip coupled line bandpass filter with different specifications are derived using the proposed method to study the improvement in the in-band flatness. The method of group delay is then integrated with Implicit Space Mapping (ISM) technique to derive EM models for the corresponding circuit models. The quantitative analysis of the simulated scenarios using the proposed method shows a significant improvement of 0.5-2.0 dB in the in-band flatness of the filter response. Finally, a microstrip filter is fabricated and tested to validate the method proposed for lossy filter design.
Open Access
A mixed methods study on barriers and facilitators to exercise for suicidal ideation management
(Faculty of Graduate Studies and Research, University of Regina, 2024-10) Vig, Kelsey Danielle; Asmundson, Gordon; Hadjistavropoulos, Heather; Hadjistavropoulos, Thomas; Totosy, Julia; Gibb, Brandon E.
Suicide is a leading cause of premature death. Innovative and effective interventions are needed to prevent suicide deaths. Randomized controlled trials (RCTs) have demonstrated that a variety of structured exercise programs (e.g., aerobic exercise, resistance training exercise) improve mental health, including reducing anxiety and depressive symptoms. Moreover, failure to meet established guidelines for physical activity is associated with increased odds of experiencing suicidal behaviours. Exercise may, therefore, be one intervention option to reduce the suicidal behaviours (i.e., suicidal ideation [SI] and plans for suicide) that often precede suicide. In order to benefit from the effects of exercise, individuals with suicidal ideation must perceive exercise as an accessible, acceptable, and effective treatment option, otherwise they are unlikely to initiate and sustain an exercise program. This mixed-methods dissertation includes two studies that explored how individuals with SI perceive and experience exercise, with an emphasis on identifying facilitators and barriers to exercise. In Study 1, grounded theory methods were used to analyze data from semi-structured interviews with 17 adult Canadian participants with past-month SI. The overall theory derived from Study 1 suggests that exercise for individuals with SI is complex and should be tailored to each individual. This theory is made up of a core category of individualization, as well as four key concepts that relate to three major categories. The four key concepts of the theory are that facilitators and barriers to exercise (a) have individualized weights/impacts on exercise decisions, (b) are cumulative, interactive, and opposing forces, (c) are dynamic, and (d) exist on a spectrum from internal to external. The three major categories included in the study theory are (a) the cognitive-behavioural cycle, (b) priorities, values, and identity, and (c) interpersonal factors. In Study 2, 261 Canadian adult participants with past-month SI completed an online survey. The survey included measures of suicidal behaviour, facilitators and barriers to exercise (open-ended and closed-ended questions), past-week physical activity, and demographic and health questions. Due to the exploratory nature of the study, quantitative analyses were restricted to descriptive statistics. The qualitative and quantitative results of Study 2 supported and added to the theory developed in Study 1, including offering additional evidence of the core category, the four key concepts, and the three major categories. Most participants thought exercise can reduce SI. Improved health, both mental and physical, was a commonly reported motivator to exercise, and poor mental health was also a commonly reported barrier to exercise. Overall, the results of both studies demonstrated the importance of individualization when it comes to exercise for individuals with SI. Exercise may or may not be an accessible, acceptable, and/or effective intervention for any given individual with SI. The results may be used by clinicians, researchers, policy makers, and advocacy groups considering whether exercise might be an intervention option for individuals with SI. The results may assist future researchers who endeavor to develop exercise-based interventions for individuals with SI by providing a theoretical framework to guide intervention development and study planning (e.g., by highlighting the need to anticipate and address individual and fluctuating facilitators and barriers). Keywords: suicidal ideation, exercise, physical activity, facilitators and barriers to exercise, exercise adherence, grounded theory
Open Access
Real-time Evaluation of an Automated Computer Vision System to Monitor Pain Behaviour in Older Adults
(Faculty of Graduate Studies and Research, University of Regina, 2024-09) Stopyn, Rhonda Jennifer Nicole; Hadjistavropoulos, Thomas; Asmundson, Gordon; Gallant, Natasha; Taati, Babak; Paranjape, Raman; Jutai, Jeffrey W.
A large body of literature supports the systematic observation of facial expressions as a tool for assessing pain in both younger and older adults. Such observation is especially critical for older adults who have limited ability to communicate their pain experience due to dementia. While frequent monitoring of pain behaviours in dementia is constrained by resource limitations, computer vision technology has the potential to mitigate these challenges, especially in long-term care environments where many people with severe dementia reside. A computerized algorithm designed to assess pain behaviour in older adults with and without dementia was recently developed and validated against video recorded images. The algorithm was incorporated within an automated system that provided alerts when facial pain expressions were detected. This study conducted the first live, real-time evaluation of the automated pain behaviour detection system with community-dwelling older adults in a laboratory setting. Testing involved a total of 65 participants completing three safely-administered experimentally-induced pain tasks using a thermal pain stimulator. A video camera was used to facilitate recording and automatic processing of facial activity. Pain behaviour detection occurred when systemgenerated pain intensity scores of the facial expressions displayed by participants exceeded a predetermined threshold score. When the incidence of facial pain expression occurred, an electronic notification (e.g., email and a signal light) was generated as notifications of pain behaviour detection. Participants completed continuous self-report pain intensity ratings during the thermal pain tasks. Receiver Operating Characteristic curve analyses were used to determine the sensitivity and specificity of the system in identifying pain- and non-pain facial expressions using gold standard manual coding completed by trained coders. Gender differences were also explored in relation to system performance. Correlational procedures were used to evaluate the relationship between pain intensity scores generated by the system, continuous self-report pain ratings, observational pain coding, and stimulus temperatures. This study supported the potential viability of the automated pain behaviour detection system in correctly identifying live, real-time instances of facial pain expressions in older adults. System-generated pain behaviour scoring achieved a maximal greater correlation with gold standard manual coding compared to prior testing using video-recordings. Specifically, system performance improved when more intense facial pain expressiveness was considered compared to more subtle facial expressions at lower pain intensities. In comparing system scoring to manual coding, there was not a one-to-one correspondence in coding but a range of comparative values that varied from participant to participant. Correlational analyses showed that continuous self-report pain ratings were weakly correlated with system scoring and manual coding. While average pain scores remained homogenous across genders, results suggested that the system performed better at identifying pain expressions for women compared to men. As expected, the pain-related facial movements of brow lowering and levator contraction were unique predictors of system-generated scores. Future evaluations of the system involving field trials in long-term care settings with older clinical populations would further elucidate the performance of the system. This technology is expected to aid in the assessment of pain in people living with dementia while addressing resource constraints in long-term care environments and reduce burden for caregivers. Keywords: Pain, aging, technology, older adults, computer vision, dementia
Open Access
Machine learning-based models for failure prediction and propagation in smart grid systems
(Faculty of Graduate Studies and Research, University of Regina, 2024-09) Salehpour, Ali; Al-Anbagi, Irfan; Bais, Abdul; Wang, Zhanle (Gerald); Yow, Kin-Choong; Louafi, Habib; Ameli, Amir
The smart grid connects components of power systems and communication networks in an interdependent two-way system that supplies or receives electricity to or from prosumers and collects data that enables it to react to usage levels and interference from threats, such as cyber-attacks. Cascading failures resulting from cyberattacks are one of the main concerns in smart grid systems. The use of artificial intelligence (AI)-based algorithms has become more relevant in identifying and forecasting such cascading failures. However, existing models that study the propagation of cascading failures either omit the impact of the communication network or power characteristics on the propagation process. To address this gap, in this thesis, we propose a set of novel cyber-attack failure propagation models in smart grids. First, our realistic failure propagation (RFProp) model addresses the system’s heterogeneity by assigning different roles to its components. We define rules and interdependencies for failure propagation and propose a new model for studying cascading failures. In addition, the RFProp graph-based model identifies the most vulnerable nodes and implements power flow analysis to guarantee that all transmission lines work below capacity and remove lines exceeding capacity. Our results establish that by considering both power and communication characteristics and interdependencies, cascading failures are modeled more accurately. In the second step, we propose a novel earlystage failure prediction (ESFP) model based on supervised machine learning (ML) algorithms. We use the RFProp model to generate a dataset for training these algorithms and predicting the state of a system’s components after a failure propagates in that system. Using the ESFP model, we predict failures of all of a system’s elements in the early stages of failure propagation. We use the XGBoost algorithm and consider the features of both the power and communication networks that provide high accuracy in the prediction process for failures. We also identify the location of the initial failures, as this allows for further protection plans and decisions. In the third step, we use the real-time digital simulator (RTDS) to develop a real-time early-stage failure prediction (RESP) model that simulates the power system in real time and makes it more realistic. We evaluate the RESP model’s effectiveness using the IEEE 14-bus system, which results in the XGBoost algorithm achieving a high accuracy in predicting attacks and with a lower testing time. Finally, we introduce a real-time attack prediction (RTAP) model based on a real-time testbed designed to examine the impact of cyber-attacks on smart grid systems. We utilize real-time simulators, including RTDS and network simulator 3 (NS3) to emulate the behavior of power and communication networks. Using this model, we employ various ML algorithms to detect cyber-attacks. We evaluate the effectiveness of the proposed model using an IEEE 14-bus test case, demonstrating high accuracy and efficient testing time.
Open Access
Adaptive systems for DDoS attacks detection and mitigation in IoT networks
(Faculty of Graduate Studies and Research, University of Regina, 2025-01) Saiyed, Makhdumabanu Farukali; Al-Anbagi, Irfan; Bais, Abdul; Laforge, Paul; Louafi, Habib; Karimipour, Hadis
The rapid growth of IoT devices has revolutionized industries while exposing IoT networks to cybersecurity threats, particularly DDoS attacks, which compromise network stability. Traditional detection methods struggle to address the constraints of resource-limited environments, scalability, and the need for lightweight, optimized, and reliable systems. This thesis addresses these challenges through five objectives aimed at adaptive DDoS detection and mitigation systems for IoT networks, balancing accuracy, resource efficiency, and adaptability. The first objective focuses on developing a Flow and Unified Information-based DDoS detection system (FLUID) for small-scale IoT networks, enabling DDoS detection with minimal computational overhead. The FLUID system uses flow metrics and unified information measures, to detects both high and low-volume attacks while optimizing resource use. The second objective introduces a system with novel hybrid feature selection to enhance detection accuracy in medium-scale IoT networks. By combining Genetic Algorithm and t-test for DDoS Attack Detection (GADAD), this system improves feature selection efficiency and supporting binary and multiclass classification. For large-scale networks, the third objective is the design of a Deep Ensemble Learning with Pruning (DEEPShield) system that integrates CNN and LSTM architectures, optimized through post-training pruning and a novel preprocessing method. This system achieves high detection accuracy with low resource demand, suitable for resource-constrained IoT environments. The fourth objective focuses on optimizing deep learning-based detection systems to enhance resource efficiency and explainability using the OMEGA, ADEPT, and SHIELD systems. The Optimized Ensemble Learning with Pruning (OMEGA) and Interactive and Explainable Optimized Learning (ADEPT) systems apply techniques like genetic algorithms and differential evolution for resource efficiency. The SHAP-Based Explanation and Lightweight DDoS Attack Detection (SHIELD) system uses SHapley Additive exPlanations (SHAP) for interpretability of individual predictions. The final objective addresses adaptive mitigation through a Game-Theoretic DDoS Defense Strategy Model (GT-DDSM) that dynamically adjusts defense strategies based on attack intensity. These systems are evaluated on metrics such as accuracy, precision, recall, F1-score, and scalability, while optimization efficiency is assessed by preprocessing time, inference speed, memory usage, and model size. Explainability is assessed through SHAP and priority assessment values, while mitigation effectiveness is measured by gradients, cumulative payoff, mitigation time, resource utilization, and network QoS parameters.
Open Access
Influence of pore structure and fluid properties on the dynamics of foamy oil production: an experimental and numerical analysis
(Faculty of Graduate Studies and Research, University of Regina, 2024-12) Sabeti, Morteza; Torabi, Farshid; Zeng, Fanhua (Bill); Muthu, SD Jacob; Mobed, Nader; Leung, Juliana
Cyclic Solvent Injection (CSI) stands out as one of the leading solvent-based post-CHOPS Enhanced Heavy Oil Recovery (EHOR) methods, celebrated for its energy efficiency, improved oil quality, and environmental benefits. Given the escalating concern over greenhouse gas emissions, exploring the use of CO2 in EHOR methods is crucial for mitigating the greenhouse effect. Studies have demonstrated that mixing CO2 with other solvents can enhance CSI performance by leveraging foamy oil flow as the primary driving force. Our state-of-the-art microfluidic systems, developed in-house, offer precise visualizations of the process, and enable controlled simulation of reservoir properties throughout experimental series. In this study, experiments were conducted on both porous and non-porous media to examine the influence of additives, solvent type, and pressure reduction rate on foamy oil production. Utilizing heavy oil from Canada, we conducted fundamental tests, including composition analysis, Constant Composition Expansion (CCE), and Differential Liberation (DL) tests, to characterize the oil and its gas-saturated live oil state. The solvents employed were CO2 and CH4, with CH4 chosen for its cost-effectiveness despite its lower performance compared to C3H8. However, a mix of CO2 and CH4, with added surfactants, yielded improved bubble generation and stability. The project focused on three primary input parameters: solvent type, surfactant concentration, and pressure reduction rate, with their ranges informed by prior research. Minitab software guided the experimental design, suggesting 15 tests on a bulk microfluidic model to observe the dynamics of live and foamy oil under varying pressures. The results of 15 tests highlighted two individual tests ii with superior stability and the lowest energy usage. In addition, optimal input parameters for reducing the oil in place and increasing production rate were identified using Minitab for further application. These optimal conditions were then tested on three different porous microfluidic models to assess the impact of porosity on foamy oil expansion, with porosities set at 31%, 35%, and 40%. The 31% porosity model exhibited the highest stability and expansion, indicating that lower porosity restricts bubble movement, hindering their coalescence and growth. Micro-analyses on bubble dynamics and movement were performed, followed by non-equilibrium reaction studies using CMOST to determine the impact of each parameter on oil production. Initial tuning for solvent type, surfactant concentration, and pressure reduction rate was conducted in CMOST, with subsequent adjustments for porosity and relative permeability. Numerical analysis of bulk phase expansion and three-porosity model results from experimental tests were utilized to derive tuning coefficients. A key novelty of this study is the derivation of non-equilibrium equations that incorporate variables such as pressure reduction rate, porosity, solvent type, and surfactant concentration in the foamy oil process. Finally, the optimized parameters for maximizing heavy oil expansion were implemented in a cylindrical sand pack model. Both simulation and experimental results from the sand pack test indicated that oil production became negligible at lower pressures during the pressure depletion test. The optimal oil production, achieving the highest benefit, was found to reach 37% of the initial oil in place using the pressure depletion technique.
Open Access
Develop innovative methodology to optimally fill in missing values and predict progression on multiple sclerosis
(Faculty of Graduate Studies and Research, University of Regina, 2024-11) Pilehvari, Shima; Peng, Wei; Shirif, Ezeddin; Khan, Sharfuddin; Fan, Lisa; Bui, Francis
Applying Machine Learning (ML) to predict and track Multiple Sclerosis (MS) progression is a significant advancement in medical research, with the potential to enhance patient outcomes. Accurate MS prediction enables personalized treatment, timely interventions, and improved quality of life by slowing disease progression and preventing complications. This research aims to deepen our understanding of MS by developing ML models and comprehensive risk assessments to support early prognosis, guide treatment strategies, and reduce disease impact. A major challenge in medical research, especially in predicting MS progression, is effectively managing missing data in MS datasets. This study introduces an innovative sequential Multi-Imputation (MI) bootstrapping method to address the challenge of missing data in MS datasets. Initially, several ML algorithms, including k-Nearest Neighbors (kNN), Random Forest (RF), and Multilayer Perceptron (MLP), are evaluated for imputation efficiency. RF and MLP perform best, achieving overall accuracies of 92% and 91.5%, respectively, in handling missing data more accurately than other models. Given the effectiveness of RF and MLP in capturing complex patterns in data, these models are selected for further development. The next step applies Multi-Imputation (MI) bootstrapping in a sequential manner, prioritizing features based on the strength of their relationships, as determined by Pearson correlation analysis. This statistical technique identifies features with the highest correlations, ensuring that attributes with stronger relationships with other attributes, are imputed first. These imputed features then inform the next imputation in the sequence, cooperating with the subsequent ranked feature in the order. Bootstrapping, a resampling technique that involves replacement, creates multiple training datasets by repeatedly sampling from the original data, enhancing the robustness of the imputation process. The proposed sequential imputation method integrates bootstrapping with RF, achieving an accuracy up to 97 % for MS datasets. This iterative approach effectively imputes missing data attributes while accounting for feature significance and relationships. The results also show that prioritizing normalization improves scaling impact, and that the significant features in the original dataset are crucial to the accuracy of MS missing data estimations. These findings provide valuable insights into effective imputation techniques for MS prediction, offering a foundation for future improvements in handling missing data in specific datasets. In addition, this study solves the common overfitting problem caused by data imbalance through a comprehensive method combining feature extraction, undersampling, Synthetic Minority Oversampling Technique (SMOTE) and optimal threshold method. Support Vector Machine (SVM), Logistic Regression (LogR), Decision Tree (DT), RF, KNN, MLP and Naive Bayes (NB) are used for prognostic modeling while examining risk factor associations. The results showed that the proposed method prevented overfitting during model training and developed a robust MS progression prognosis model, achieving a prediction accuracy of 98%, particularly for SVM and MLP The methods proposed in this dissertation can help develop more concise guidelines for the medical research communities and improve their evaluation processes. These innovations not only advance prognostic analysis in MS, but also pave the way for future research focused on optimizing patient outcomes and treatment strategies.
Open Access
Secure and scalable blockchain mechanisms for IoT applications
(Faculty of Graduate Studies and Research, University of Regina, 2025-01) Pathak, Aditya Kalpesh; Al-Anbagi, Irfan; Laforge, Paul; Paranjape, Raman; Hamilton, Howard; Stakhanova, Natalia
Integrating blockchain with IoT ensures secure, transparent data exchange through immutability and consensus mechanisms, preventing data tampering. However, the increasing number of IoT devices raises risks like unauthorized access and network attacks. Blockchain scalability issues also affect throughput and latency, challenging real-time IoT applications. This thesis addresses these challenges through four contributions that aim to improve the security, scalability, and efficiency of blockchainbased IoT networks, balancing security with performance needs. Our first contribution is to develop an end-to-end security mechanism for IoT networks, called the trust-based ABAC mechanism for IoT networks (TABI). TABI integrates edge computing and blockchain technology to mitigate risks from malicious devices and offload computational tasks to edge layers. It operates on Hyperledger Fabric (HLF), a permissioned blockchain that enhances throughput and latency through its executeorder- validate architecture. Our second objective is to provide scalability within blockchain-based IoT networks using a sidechain-based trust and access control system, named sidechain-based trust and access control mechanism for IoT networks (SATI). By distributing trust evaluation and access control operations across a separate blockchain or sidechain, SATI improves the scalability of IoT networks. We implement a cross-chain transfer mechanism to ensure communication between the sidechain and the mainchain, thus overcoming a fundamental limitation of traditional blockchain architectures. Our third contribution is to improve the security of the IoT network by introducing a Zero-Knowledge Proof-based Mutual Authentication (ZPMA) mechanism, a privacy-preserving mutual authentication mechanism. Utilizing Zero-Knowledge Proofs (ZKP) based on the quadratic residue technique, Z-PMA ensures secure and private mutual authentication between edge devices and IoT devices. We also implement an incentive mechanism to select additional authenticators from the base station layer to reduce authentication latency and support the demands of low-latency IoT networks. Our fourth contribution is to detect and resolve conflicting transactions in HLF-based IoT networks at an early stage, known as the early-stage conflict transaction resolution (ECR) mechanism. ECR identifies and resolves conflicting transactions at an early stage using a local cache at the endorsement phase of the HLF transaction processing. Additionally, ECR uses dependency model and an efficient reordering process to distribute transactions in a way that minimizes conflicts. This mechanism enhances the performance of HLF-based IoT networks by reducing the impact of conflicting transactions, ultimately improving throughput and latency.
Open Access
New probabilistic approaches for detecting and evaluating concept drift in data streams
(Faculty of Graduate Studies and Research, University of Regina, 2025-01) Parasteh, Sirvan; Sadaoui, Samira; Butz, Cory; Uddin, Sami; Yow, Kin-Choong; Shafiq, Omair
In modern applications like online shopping, financial forecasting, and real-time fraud detection, data distributions frequently shift, causing predictive models trained on historical data to underperform. This phenomenon, known as Concept Drift (CD), presents a major challenge in adaptive learning environments, necessitating ongoing monitoring and adjustment to accommodate evolving data streams. Active drift detection methods, which track changes in data distribution or model performance, offer a targeted solution by prompting adaptations only when significant shifts are detected. However, existing active methods face challenges: distribution-based approaches may miss subtle drifts or respond to non-critical changes (virtual drift), while performance-based methods, which detect shifts impacting model accuracy (real drift), can overreact to transient noise, leading to unnecessary adaptations. These challenges underscore the need to balance sensitivity and stability in CD detection. To address these issues, we propose a hybrid approach that combines insights derived from data distribution through probabilistic measures, such as marginal probability distribution of input data or classifier confidence, with error-based detection, offering a more robust and precise solution for managing CD. The first key contribution is the development of SPNCD, a probabilistic method leveraging Sum-Product Networks (SPNs) to detect real and virtual drifts by analyzing shifts in the joint probability distribution of features and class labels. Inspired by the Bayesian CD definition, SPNCD integrates prediction error, which assesses model performance, and the marginal distribution, which captures changes in data distribution. Building on this approach, we then develop the PRDD algorithm, which uses the classifier’s confidence as an indirect estimate of data distribution similarity, alongside error rates, to detect real drift with precision and timely response in dynamic data streams. Based on these foundations, we develop NPRDD, an enhanced method specifically designed for noisy data environments, which combines cross-entropy-based surprise measures with predicted class probabilities to distinguish genuine drifts from noise. We further enhance PRDD with two detection strategies: 1) PRDDW that uses a sliding fixed-sized window approach to determine the proportion of real-drift candidates, and 2) PRDDS that adopts a reward-aging mechanism to compute a drift score based on recent drift events. To ensure the usability of these two methods, we present a parameter optimization procedure using Bayesian optimization to find robust default parameter values that generalize well in various scenarios. To validate each of the proposed methods, we conduct an exhaustive experimental study involving different synthetic data streams, simulating abrupt and gradual drifts. These studies also compared our methods to several benchmark drift detectors. Moreover, we devise a theoretical framework to understand the impact of critical components of our methods. Also, specifically for PRDDWand PRDDS, we design an empirical framework to generate 4,000 unique synthetic data streams, define the drift regions, and present several metrics to assess the performance of the base learners and detectors; the latter needs to be improved in the literature. The empirical results show that our methods outperform existing CD detection methods in most cases in classification and detection-based metrics and rank among the top performers, underscoring their robustness and practical applicability. Moreover, our experimental framework provides benchmarks for reproducible evaluations, setting a new standard for future research in CD detection.
Open Access
Enriched model categories and the Dold-Kan correspondence
(Faculty of Graduate Studies and Research, University of Regina, 2024-10) Ngopnang Ngompe, Arnaud; Frankland, Martin; Stanley, Donald; Fallat, Shaun; Herman, Allen; Zilles, Sandra; Ponto, Kate
The work we present in this thesis is an application of the monoidal properties of the Dold–Kan correspondence and is constituted of two main parts. In the first one, we observe that by a theorem of Christensen and Hovey, the category of nonnegatively graded chain complexes of left R-modules has a model structure, called the Hurewicz model structure, where the weak equivalences are the chain homotopy equivalences. Hence, the Dold–Kan correspondence induces a model structure on the category of simplicial left R-modules and some properties, notably it is monoidal. In the second part, we observe that changing the enrichment of an enriched, tensored and cotensored category along the Dold–Kan correspondence does not preserve the tensoring nor the cotensoring. Thus, we generalize this observation to any weak monoidal Quillen adjunction and we give an insight of which properties are preserved and which are weakened after changing the enrichment of an enriched model category along a right weak monoidal Quillen adjoint.
Open Access
Beam asymmetry in the reaction channel γp→ηΔ + at Glue X
(Faculty of Graduate Studies and Research, University of Regina, 2024-08) Neelamana, Varun; Papandreou, Zisis; Lolos, George; Huber, Garth; Stevens, Justin R.; McBeth, Joyce; Watts, Dan
Photoproduction mechanisms studied in the GlueX experiment allows the mapping of light mesons in unprecedented detail with particular interest in exotic meson candidates. This is achieved by impinging an 8.2-8.8 GeV linearly polarized photon beam on a liquid hydrogen target. The measurement of beam asymmetry Σ will help constrain quasi-particle t-channel exchange processes using Regge theory. Understanding the photoproduction exchange mechanisms is a crucial ingredient in establishing hybrid and exotic photoproduced light meson states. Σ is extracted from the azimuthal angular distribution between the meson production plane and the polarized photon beam. In particular, we will report results on the beam asymmetry measurements for η in the reaction p →η Δ+. This reaction with a recoiling Δ+ will allow for comparison and validation of theoretical calculations and provide additional validation of the η asymmetry with a recoiling proton. The different isospin of the Δ+ imposes additional restrictions that further constrain allowed Regge exchanges. The results were similar to η-proton i.e Σ ≈ 1 but showed a deviation from theoretical models of the η - Δ+ especially towards higher t values. This may help guide modifications to these models for production and exchange processes involving η meson.
Open Access
Energy literacy in the Canadian elementary classroom
(Faculty of Graduate Studies and Research, University of Regina, 2025-02) Mosscrop, Larkin Elizabeth; Hurlbert, Margot; Coates, Ken; Longo, Justin; Bazzul, Jesse; Novog, Dave
This dissertation presents a body of research that addresses the policy issue of the effectiveness and role of curriculum in building energy literacy. Energy literacy, which encompasses broad content knowledge as well as affective and behavioural characteristics, will empower students to make responsible and appropriate energy-related choices, and embrace changes in the way they use and produce energy. Students who are energy literate will be more capable of engaging in thoughtful energy-related decisions as they become adults, informing policies and energy projects moving forward. A framework to assess if the elementary science curricula across Canada would meet the required elements to establish energy literacy in elementary school students was developed. This framework suggests that many of the curricula do not support energy literacy but rather focus on content knowledge. The science curriculum assessment identified all aspects of energy expected to be understood by the end of grade 8. This assessment was used to formulate the student survey, and the in-class focus group. Students showed general energy literacy during the focus group discussions in addition to significant learning and interest in energy after the discussion. Grade 6 students showed surprisingly high degrees of understanding and application when compared to the middle-school students. Teachers were interviewed to evaluate their use of curriculum in the classroom, teaching practices, and understanding of science literacy. Teaching science was raised as a way to increase engagement and accessibility for those who struggle in other areas in school. Science through self-directed project-based learning does not require the same level of literacy or numeracy for engagement. However, this focus on inquiry-based learning was bound by the teacher having enough engagement to answer questions and involve students emphasizing the need for professional development for new or complex topics. The ever-increasing demands on teachers to integrate social issues (e.g., environmental and climate justice) to science education is particularly relevant when considering that energy literacy has three core pillars including affect or the values and beliefs one has. This is particularly important with an increasingly polarized world, where students are bombarded with polarized media and face teachers who have their own implicit and explicit biases. Education must focus on facts while still providing a strong foundation of scientific skills that enable students to develop their own beliefs about science. The curriculum is not structured in such a way as it too has many implicit biases, such as defining energy as either renewable or nonrenewable. These findings support the complex relationship between knowledge, affect, and behaviour, underscoring the importance of using educational strategies that focus not only on cognitive development but engage the whole student in the learning process while still maintaining balance and focus on scientific outcomes. The results also provide evidence for using educational pedagogies that incorporate projects and inquiry-based models to connect the content to student lives outside of school. The results of this study do provide some insight as to the utility of curriculum as a policy tool, that is that curriculum is only one part of a complex system and curriculum reform alone is unlikely to equate to changes in the classroom. Overall, the results of this study support the need for wider implementation of science professional development, including project-based energy education and the creation of resources that can be easily and freely accessed by teachers. Key words: education, energy literacy, science literacy, pedagogy
Open Access
Exploring Chinese EFL teachers’ use of translanguaging in the classroom: an ethnographic case study
(Faculty of Graduate Studies and Research, University of Regina, 2025-03) Li, Meihan; Hébert, Cristyne; Boutouchent, Fadila; Schroeter, Sara; Bardutz, Holly; Riches, Caroline
Translanguaging as a sociolinguistic theory and a pedagogy has captured the attention of language instructors in recent years, especially those working with culturally and linguistically diverse students. Most research on using translanguaging in teaching English has primarily been conducted in English as an Additional Language (EAL) context, particularly in the United States (see studies by Canagarajah, 2012; Carbonara & Scibetta, 2022; Creese & Blackledge, 2010; García & Kano, 2014). However, as a theory and pedagogy highly valued by language educators, its application should be extended to a broader context, reaching more diverse language learners. The literature suggests that translanguaging remains insufficiently investigated in English as a Foreign Language (EFL) context, particularly at the tertiary education level. To address this gap, this study aimed to investigate the perceptions and classroom practices of translanguaging pedagogy among Chinese EFL instructors at the university level in China, while also examining the complex relationship between their perceptions, beliefs, and classroom practices. This ethnographic case study employed qualitative research approach within the constructivist/ interpretivist paradigm (Lincoln, et at., 2011). Five EFL instructors that came from two different teaching units at a tertiary university were selected as the participants. Data were collected mainly through classroom observations and semi-structured interviews. Additional data sources, including demographic questionnaire, field-notes, and foreign language education documents at both the institutional and national levels, were used to support the dataset and triangulate the results. The study’s findings revealed that although most participants knew little about translanguaging theory and pedagogy, they had already frequently and naturally employed similar strategies in their classrooms. Five types of pedagogical translanguaging: explanatory (instructional) translanguaging, evaluative translanguaging, cognitive translanguaging, affective translanguaging, and directive translanguaging, all of which were commonly directed by instructors in various classroom settings. Two student-directed translanguaging strategies, namely interactive translanguaging and interrogative translanguaging, were spontaneously initiated by students. Most of the time, teacher participants were able to create a safe translanguaging space for students to interact in. The findings also revealed that, despite all participants using translanguaging strategies in their daily teaching, they held diverse perceptions about its use in the classroom. Participants with more knowledge of translanguaging theory had more positive perceptions of guiding students to use their first language and other multimodal resources in learning a new language. Overall, the frequency, and manner in which instructors engaged in translanguaging in class were substantially in line with their individual perceptions of the theory. In some cases, inconsistencies between their teaching practices and verbal assertions were attributed to potential influences on teachers’ perceptions and practices, which manifest across four levels: the individual level, the classroom level, the institutional level, and the socialpolitical/ national level. Key words: translanguaging; teachers’ translanguaging perceptions; teachers’ translanguaging practices; EFL in Chinese tertiary education.