Projets novateurs réalisés

Power Monitor Load Disaggregation for the Electric Grid

For decades the electric grid has remained a passive system that has delivered electricity to many
households and businesses. As utility companies look at converting a passive electric grid to a smart
grid, a number of sensors and smart meters must be deployed throughout the grid system to achieve
this objective. Deploying vast amounts of sensors and smart meters becomes a costly and timely affair.
This project looks at adapting nonintrusive load monitoring (NILM) algorithms for the smart grid.
NILM algorithms have been used for monitoring appliances being used within a horne using one
power meter. Using NILM algorithms has the potential to minimize the amount of sensors deployed
over the smart grid (saving cost and time). This project investigates extending, optimizing, and
evaluating NILM algorithms for the smart grid.

Toward Net Zero Transportation: Environmental benefits of Hydrogen-fuelled Trucks

Hydra Energy Corporation Commercial Demonstration Project based in Prince George, British Columbia 12.4 MT/day of waste hydrogen will be captured, purified and transported to an onsite hydrogen refueling station from which Class 8 tractor-trailer truck fleets will refuel daily. Hydra’s demonstration project can provide a significant reduction in fleet greenhouse gas(GHG) emissions, Particulate Matter and other air emissions.

The R&D project results will advance the scientific knowledge of hydrogen in internal combustion engines, support the granting of a third party verification/certification of the environmental benefits of Hydra’s technology, and will help to expand the availability of clean energy transportation technologies in and beyond B.C.

Using a mobile application to support at-risk student re-entry into post-secondary education in the era of COVID-19

The aim of this 2-year project is to do research to inform the development of, and fully test and develop a mobile application designed to improve the experience of (particularly at-risk) post-secondary (PSE) students in addressing COVID-19-related issues. Our key concern is that COVID-19 has not only disrupted important and significant developmental experiences that improve student experience and success, but it has also caused challenges in students’ lives away from university that will spill over into their experience of being post-secondary students. We seek to address these disruptive factors via assessment of need and development of a mobile application which can link students with supports necessary to address those needs. A mobile application approach is particularly valuable in an era where social distancing will reduce the opportunity for face-to-face and in-person communication for at least the next 18-24 months.

Rural Asset Management in a Changing Climate

The Rural Asset Management in a Changing Climate (RMACC) project will develop asset management capacity and guide the implementation of asset management plans and policies that account for future challenges posed by climate change with a cohort of five small municipalities. The objective of the project is to perform asset management planning in a comprehensive manner for climate resilient and sustainable infrastructure.

À la croisée de la recherche et de l’intervention : une synthèse des connaissances théoriques, empiriques et pratiques sur les options d’apprentissage soutenant la capacité à travailler ensemble à l’ère de l’industrie 4.0

L’avènement de la quatrième révolution industrielle incite les organisations à revoir leurs pratiques afin d’outiller leurs membres quant à leur façon de travailler ensemble (McKinsey Global Institute, 2018). Cette nouvelle réalité fait ressortir le besoin de disposer de moyens de développement adaptés à la complexité et à la diversité des contextes des travailleurs. Face au volume de connaissances éparses, disparates, voire contradictoires, les décideurs organisationnels se heurtent à un véritable casse-tête quant au choix des options d’apprentissage à mettre en oeuvre (Cumberland, Herd, Alagaraja, & Kerrick, 2016). Devant cet enjeu organisationnel, la présente étude vise à dresser un portrait des connaissances portant sur les options d’apprentissage susceptibles de soutenir le développement de la capacité à collaborer avec autrui. Pour ce faire, le projet s’appuie sur une méthode systématique de recherche de données documentaires qui place la consultation des utilisateurs potentiels des connaissances au coeur de la démarche scientifique.

Developing a facility-level energy load shape simulation method for utility planning

Utility planners require an understanding of how and when energy is used among their customers. Hourly “load shapes”, that represent facility electricity of natural gas use for each hour of the year, often disaggregated into individual “end-uses” such as space heating, lighting or water heating are the most detailed information of this type typically available to planners.
These load shapes allow planners to make bottom-up estimates of aggregate loads over their electricity or gas distribution system over various geographies, allowing them to ensure transmission/distribution infrastructure and generation/commodity supply will meet demand, integrate intermittent renewable resources, and estimate the effects of efficiency, demand-response and distributed generation measures.
This research will build on previous work by the partner to estimate annual hourly energy use by facility type and energy end-use in utility service territories. The methodology is expected to be based on energy load research for similar facilities (i.e. office buildings), equipment (i.e. chillers), or actions (i.e. application of energy conservation or demand response measures), and adjusted based on climate data, known annual energy use at the system level, and qualitative/quantitative survey data.

Helping Servus Members Reach Financial Goals via Transfer Learning

In this self-contained project we will investigate how machine learning can be applied to help provide personalized financial advice. Machine learning is a term that designates types of artificial intelligence that rely on learning behaviors from data or experience. Specifically, the goal of this work is to apply machine learning to Servus Credit Union’s Noble Purpose “Shaping Member Financial Fitness” to provide personalized recommendations to individual members who have set specific financial goals.

An exploratory study of Asian International Students‘ experiences in Ontario Schools

A rapid increase in the number of Asian international students attending K-12 schools has led to the investigation of the complex, multi-layered aspects of the life experience and needs of Asian international students in Ontario schools. Working closely with View-Wide International Education Group and using multiple case studies, this study will explore the factor of facilitating their transition and articulate the nature and challenges of Asian international students’ experiences in adjusting to new school settings in Ontario. It will also investigate their perception of the preparedness for studying in Canada and ensure their needs of academic and social success in educational endeavour. The study is important in that it can provide teachers, counselors, school administrators, early study abroad agencies, and after-school institutes with concrete data based upon the experiences of Asian international students

Development of a Framework for Risk Assessment for Horizontal Directional Drilling Applications

Horizontal directional drilling (HDD) is a steerable trenchless construction method adapted in the
1970’s from the horizontal oil well drilling technology. HDD has become a very successful worldwide
technology for trenchless utility and pipeline installations becoming an annual multi-billion dollar
industry across five continents. The HDD process involves considerable risks based on many hazards
and uncertainties including risk resulting from unknown ground conditions, construction risks, and
environmental concerns. In response to the critical demand from industry to address these HDD
issues, the University of Alberta and The Crossing Company (TCC) established a collaborative
research partnership. The proposed research will focus on developing a risk assessment framework
addressing the geotechnical and construction risks involved with HDD to develop risk mitigation
strategy guidelines and predictive model development to enable trenchless technology stakeholders to
appropriately and successfully plan and construct HDD installations.

Détection proactive/préventive d’anomalies dans les machines à commande numérique (CNC)

Au cours des dernières années, CAM Solutions a remarqué que plusieurs de ses clients. des usines munies de machines à commande numérique (CNC), éprouvent systématiquement de la difficulté à les faire fonctionner à pleine capacité et selon un horaire établi. En effet, l’apparition constante de pannes et de problèmes imprévus nuit à la performance de ces machines et entraîne des réparations coûteuses avec un coût d’opportunité élevé : ces machines-outils peuvent coûter plus d’un million de dollars, les matières premières et la main-d’oeuvre entraînant représentent un coût horaire variable important. Les temps d’arrêt sont donc perturbateurs et surtout coûteux.
Pour remédier à ce problème, le projet supervisé suivant a été mis en place. Il aura comme objectif trouver et modéliser les causes principales des pannes / anomalies affectant les machines industrielles CNC. Ceci permettra ultimement de développer une solution complète permettant de mitiger proactivement leur occurrence. Pour arriver à modéliser adéquatement ce phénomène, des méthodes d’apprentissage chronologiques, supervisées et/ou non supervisées, seront utilisées sur les données disponibles (liées aux différentes machines, pièces et tâches) dans le but d’effectuer de l’inférence et de la prévision.

Intelligent Orchards: Redefining the Production and Management of Tree Fruits

One of the main challenges in tree fruit orchards is to accurately predict apple yield and identify the health of individual trees (e.g., healthy foliage, fruit development, detecting and identifying diseased trees). Manual performance of these tasks is labour intensive and costly. Therefore, automated processes provide novel solutions with enhanced accuracy, efficiency, and productivity. The proposed solution is to develop an automated system that utilizes machine vision and artificial intelligence strategies to accurately count flower blooms and apples, detect and identify diseased fruit and trees, and improve yield estimates for producers. Automating this process will reduce labour costs and improve apple yields for producers. The system will provide truly meaningful and easy-to-digest information to the farmer about the orchard, including the predicted productivity, suggested trimming, and possible growing issues (such as unhealthy trees or areas that require more attention).

Visual-haptic Representation for Zero-shot Learning

Humans recognise objects in the world leveraging multi-modal sensory inputs beyond visual aspects (images and videos). Touch based information (Haptics) possesses rich information about structure, shape and other objetness properties. In this work, we will study and learn cross-modal representations between vision and touch. To connect vision and touch, we plan to introduce a zero shot classification task of recognising unseen object categories from shapenet dataset using haptics signals. We will train our model to encode the haptics information to a view agnostic embedding space that captures the geometrical aspects of the object. To support our claims, we will use shapenet dataset, a repository consisting of CAD models of various categories of objects that can be rendered from different views and the Johns Hopkins Modular Prosthetic Limb for haptics data. Our hypothesis is that our learnt representation can help transfer representations across modalities, for zero shot classification and object retrieval.