Innovative Projects Realized

Investigation of Microorganisms’ Co-culture System for Biomass/Biofuel Production

The research focuses on the crucial needs for the optimization of biofuel production process. It supports the Canadian energy and environment sectors which are seriously searching for more efficient process, targeting the increasing concern of the society with respect to the fossil fuel energy resources depletion and environmental footprints.

Chemical and biological characterization of the non-tetrahydrocannabinoid profiles of medicinal plants

We propose to perform detailed studies that characterize chemically and biologically and the unaltered ancestral plant species of the Cannabis genus. Chemically, these studies will provide the chemotype profiles for each parental species for a spectrum of non-tetrahydrocannabinoid compounds. Biologically, these studies will provide the pharmacological profiles for each parental species. These data will enable the development of medical cannabis and cannabinoids (the right drug and the right dose) that can then be used for clinical trials (the right person/disease) to ultimately identify the therapeutic role of cannabinoid based pharmaceutics in disease management (the right time).

Backtesting de strategies en finance

Le projet consiste a evaluer I’excedent dee rendements relies a des strategies

d’investissements precises. Ces strategies d’investissement seront basees sur

I’identification d’anomalies de marche non exploitees par les investisseurs. Ces

anomalies seront pour la plupart de nature comptable. Apres I’indentification des

variables com ptables. je devrai choisir les titres pertinents selon nos criteres avec Ie

logiciel de recherche Factset. Ensuite. je devrai creer un portefeuille fictif avec les

titres selectionnes afin de suivre son evolution dans Ie temps. Enfin je conclurai par

les apprentissages et Ie bien-fonde des ces strategies.

Autonomous structure detection and inspection using unmanned aerial systems

In this project, a new method is developed to optimize the performance of an Unmanned Aerial Vehicle (UAV) for autonomous detection and on-the-job view-planning of infrastructure elements with the purpose of their accurate three-dimensional (3D) modeling. The existing view-planning approaches in the literature have mostly modeled non-complex or small-scale objects and have rarely been adapted to flying robots. In addition, the target object is often identified by human operators. This research addresses these problems by training a drone to find the desired object of interest in an unknown environment during an inspection task without human interventions. To this end, first, a technique for object detection will be developed to recognize and locate the target object while the drone is exploring the environment. Second, based on the available information about the desired object, the drone will start next-best-view and motion planning to acquire an adequate photogrammetric network of images in order to reconstruct the inspection target in 3D both accurately and completely. This research will have important impacts on the evolution of infrastructure monitoring and assessment approaches using UAV systems.

Advancing Human Performance in the Canadian Football League

North American professional football players struggle with mental health challenges such as addiction and depression during and following their athletics careers. Despite the fact that these athletes value positive emotional, psychological, and social mental health, little is known about ways organisations can protect and promote these parts of athletes’ mental health. Therefore, the purpose of this research is to explore challenges and opportunities CFL players face to experiencing positive mental health during a professional season. This research will help inform a CFL Human Performance Program designed to improve player well-being and maximise on-field performance.

Identifying Questions for Game-Based Learning through Deep Learning

Game-based learning tools often make use of questions to measure and encourage learning, but generating questions can be challenging, especially at the scale that companies like Axonify are required to do. In this project, the intern will design, implement, and evaluate a system that can apply machine-learning on a corpus of text (e.g., a textbook) to automatically generate questions that can be used in game-based learning tools. This system will allow Axonify to scale their products to larger corpora of source material, larger sets of questions, and ultimately have a much larger market as a result.

Sensitivity Analysis of Gas and Particulate Matter Emissions from Future Power Generation in the Province of Alberta

The Province of Alberta (AB) has decided to phase out coal power generation by 2030 and increase renewable electricity production to 30% of total power generation, also by 2030 with the remaining 70% of the power generation being dependent on natural gas. It has been conjectured that part of generation portfolio could be diversified to include nuclear power generation. The current proposal aims at studying available power generation (seasonally) in Alberta and create a model to predict their gas (CO2, CH4, and NOx: mainly N2O, but also NO and NO2) and PM1 (particulate matter) emissions in time using different generation portfolios. Once this model is verified against gas emission data obtained from the literature, future seasonal emissions will be predicted after varying the generation portfolio to include a certain amount of nuclear power generation (from 0 to 25% of the total output). An uncertainty analysis of the prediction will also be performed.

Investigating polysaccharide-protein conjugation: characterization and effects of glycation conditions

Pneumonia remains the single leading cause of childhood death under age 5 worldwide. The price per dose of current vaccines is high and supply is limited due to a complex manufacturing process and low yield, significantly reducing its distribution in developing nations.1 A newly patented vacuo dry-glycation process promises much higher efficacy than the conjugation chemistry used currently, paving the way towards a much lower dosage cost. and its vaccine is a kind of polysaccharide-protein conjugate system. However, the process conditions required for activation of the polysaccharide by vacuo dry-glycation have not been optimized, and the characteristics of distinct serotype polysaccharides and the corresponding activated polysaccharides and conjugate products are not fully reported and studied. This research addresses these deficiencies, enabling PnuVax Inc. to further the development of a more affordable vaccine that can be used in Canada and around the world to reduce childhood death due to pneumonia.

Analyse structurale des formations de fer au nord-est du réservoir Manicouagan

Champion Iron Mines est une société de développement et d’exploration québécoise effectuant de l’exploration régionale dans la plus importante région de production de minerai de fer du Canada. Le minerai se trouve dans des formations de fer plissées en une géométrie complexe lors de la formation de la chaîne de montagne « Grenville » il y a un milliard d’années. Le projet proposé vise à faire une analyse structurale de deux projets minier situés près entre le réservoir Manicouagan et Fermont. Les résultats permettront de tester des modèles conceptuels proposés récemment pour expliquer des structures d’orientations perpendiculaire à celle attendue pour une chaîne de montagne. Déterminer la géométrie 3D des gisements améliorera l’estimation des réserves et la localisation d’autres gîtes de fer potentiels. Enfin, ce projet contribuera à la formation d’un étudiant spécialisé en géologie de terrain, une expertise rare, mais cruciale pour le milieu d’exploration minière du Canada

Pattern Recognition: The Exploration of Machine Learning Algorithms in Archaeological Site Prediction, Fraser River Valley, British Columbia

Golder Associates Ltd., teaming with the Seyem’ Qwantlen Business Group (Kwantlen First Nation), was retained by the Township of Langley to develop a model to predict the location of unrecorded archaeological sites on a 10,000 year-old landscape located in the Fraser River Valley, British Columbia. Conventional predictive modelling techniques are common practice however with the increased availability of more powerful computers and software there is a growing potential for using machine learning algorithms to predict a wider variety of archaeological site types with greater accuracy. For this Pattern Recognition Project (PRP) the intern will complete a machine learning literature review and an examination of the local archaeology to identify potential machine learning methodologies and algorithms to predict site locations as well as the best environmental, physiographic and cultural variables to input into the model. This research will be used to create a report which describes the PRP, its results, and recommendations. The results generated by the PRP will be used to pilot a new approach for archaeological predictive modelling using machine learning algorithms and will ultimately be used to assist the Township in responsibly managing local archaeological sites.

Ultra-thin graphene oxide membranes for efficient humidity harvesting

The main goal of the proposed work is to develop an ultra-thin and selective GO membrane capable of separating water vapor or steam from air. For this purpose, a suitable membrane supports required to hold the GO sheets. Therefore, the GO sheets will be deposited on various membrane supports and their performance in terms of selectivity, permeability, and mechanical strength will be evaluated. Then, the effect of GO layer number on the selectivity and permeation rate will be investigated. Once the best performing membranes have been determined, the impact of feed humidity and temperature on the permeation rate and selectivity of water vapor transport through the membranes will be evaluated. Evercloak will supervise the intern and support out of lab research activities on their premise. Evercloak will benefit as the results of this project will inform key decisions within their technology development roadmap in addition to talent acquisition/training.

Using Machine Learning for audio analysis and synthesize

Voices.com, the largest online marketplace for voice talent, have identified Machine Learning as an enabler for
future growth. In particular, incorporating Natural Language Processing (NLP) into structured queries and
automatic classification of sample recordings. The first phase of this research involving NLP is in the process of
being integrated into production. The second phase will be to automatically classify sound samples. This has been
historically difficult resulting in low levels of accuracy, but we will take advantage of new ML techniques, and one
of the world’s largest databased of tagged audio. This classification will cover areas of current research, such as
gender and age detection, but extend to new areas including style and emotion. Having completed this
classification, we will be able to incorporate emotion into voice synthesis, increasing the acceptance and usability
of Voice AI.