Projets novateurs réalisés

Integration of Machine Learning and AI Based Optimization from IoT Datastreams and Business Information Systems

Internet of things (IoT) includes of multitude of sensors from a wide variety of applications. These sensors produce high volume and high velocity data. Recently there has been much interest in application of such technologies to improve agricultural practices. The sensors that are installed in the field transmit real time data regarding numerous environmental variables of interest. This data is then used to forecast a future state and to make a well informed business/operation decision according to an expected future state. One of the challenges in application of such technology is to improve prediction accuracy of the forecast. This project will design a generalized framework based on machine learning and artificial intelligence methodologies to improve prediction accuracy that will ultimately result in reduced operating costs and higher yields in agriculture sector.

Acceptabilité des données probantes du monde réel lors de soumissions règlementaires de médicaments sur ordonnance

Actuellement, les médicaments sous ordonnance sont testés rigoureusement durant plusieurs années, d’abord sur des cellules animales et humaines, puis sur des animaux et finalement chez des humains avant de pouvoir être approuvés par une agence règlementaire telle que Santé Canada. Pour certains médicaments, les études cliniques chez les humains peuvent être difficiles ou même impossibles à effectuer. Pour tenter de régler ce problème, l’industrie pharmaceutique se penche sur l’utilisation de données provenant du monde réel, ce qui permettrait d’éviter d’avoir à faire des études cliniques complexes pour certaines populations sensibles. Ces données dites « réelles » peuvent provenir de patients utilisant le médicament à l’étude et dont le médecin fait un suivit serré pour s’assurer que le médicament soit efficace et qu’il ne cause pas trop d’effets indésirables. Ce type de données n’est pas primé pour démontrer l’efficacité d’un médicament, mais l’avancement des technologies personnelles et des banques de données rendent ces données accessibles de plus en plus. Le projet proposé ici consiste donc à faire une revue de ce qui est accepté à travers le monde quant à l’utilisation de ces données et d’en faire la critique pour aider l’industrie à les utiliser de manière adéquate.

A reconciliation of the top-down and bottom-up approaches to risk capital allocations

Two overarching approaches to allocate the aggregate risk capital stand out nowadays. These are the top-down approach that entails that the allocation exercise is imposed by the corporate centre, and the bottom-up approach that implies that the allocation of the aggregate risk to business units is informed by these units. Briefly, the top-down allocations start with the aggregate risk capital that is then replenished among business units according to the views of the centre, thus limiting the inputs from the business units. The bottom-up approach does start with the business units, but it is, as a rule, too granular, and so may lead to missing the wood for the trees.
In this research project, we will connect the bottom-up approach to allocate the aggregate risk capital to a general class of Dirichlet distributions defined on the n-dimensional simplex. The new class, which is called herein “mixed-scaled Dirichlet distributions’’ contains the classical Dirichlet distribution as a special case, exhibit a multitude of desirable closure properties, and emerge naturally within the multivariate risk analysis context.

Study of magneto-elastic torque meter for power steering system

Permanent magnets and magnetic field sensors are widely used in the automotive industry for the detection of the rotational speed of a gear or the torque measurement of gearbox shafts, vehicle axles, and electric motors. For instance, the action of a driving wheel can modify a magnetic pattern encoded in a transmission shaft. Magnetic sensors positioned around the shaft can detect the stress applied on the driving wheel. This is what we call a magneto-elastic torque meter. It provides the signal for the operation of the power steering system. In this project, a student will develop a model to study the performance of this system to help fabricate a competitive, highly sensitive and reliable magneto-elastic torque meter. Torque monitoring is very important for real-time control and fault diagnosis of a variety of mechanical equipment, such as the rotation of shafts in many things like pumps, rotational cutting equipment, gearbox shafts, vehicle axles, and electric motors. The research on magnetic field detection could also find potential applications in the detection of magnetic anomalies, geoprospection, security and biomedical imagin

The Development Of A Large-Scale Multidimensional WebApplication To Support Data Visualization, Mining & Analysis

For this project, we will develop the statistical theory to support, as well as a full functioning prototype

for, a large-scale multidimensional data visualization application. This application, which will

be fully scalable, will include an interactive user dashboard with customizable widgets. By

taking a methodological approach to research, along with an agile approach to development,

we will create tangible outputs which will have both commercial & educational value, and will

make a substantive contribution to the Canadian technology space. This work will also help

significantly advance the mining, modeling & visualization ability of Canadian technology

media companies

Découverte d’antigènes spécifiques au cancer du sein

Le risque de développer un cancer du sein chez la femme au cours de sa vie est de 11%. Chaque année dans le monde, environ 1 675 000 femmes reçoivent un diagnostic de cancer du sein. D’ici 2020, le coût total des soins pour leurs patients atteindra plus de 2 milliards de dollars par an au Canada (20,5 milliards par an aux États-Unis). Plusieurs types de cancers bénéficient des avancées en immunothérapie. Presque toutes les immunothérapies anticancéreuses efficaces reposent sur le ciblage d’antigènes spécifiques de tumeurs (AST), c’est-à-dire de peptides associés au CMH (PAC) présents exclusivement sur les cellules cancéreuses. Cependant, le cancer du sein n’a pas bénéficié de la révolution immunothérapeutique. Dans le cadre de ce projet nous profiterons des avancées immunopeptidomiques de protéogénomique des laboratoires Perreault et Thibault afin d’identifier des AST dans des lignées cellulaires et corréler leur présence dans des échantillons primaires du cancer du sein.

Textile-Based Wearable Assistive Devices

Neuromuscular impairments resulting from conditions such as spinal cord injuries, strokes, and multiple sclerosis can greatly affect the quality of life of those affected, preventing them from carrying out typical activities of daily living. A number of active assistive devices have been proposed to provide upper limb assistance for such neuromuscular conditions. These devices primarily rely upon rigid linkages and conventional electromechanical motors to exert forces on the body, and therefore do not conform to the user’s body and can become prohibitively heavy, often requiring the devices to be used in a stationary setting such as in a rehabilitation clinic, typically with additional assistance from therapists. Recent work has sought to develop soft, conformal devices using inflatable textile balloons to provide similar assistance. Textiles are a natural material for such an application given their high strength, minimal weight, and capability to adapt to complex 3D geometries such as those of the human body. The mechanical properties of the textile greatly affect the performance of the assistive device. These properties are difficult to tune, however, and are typically limited by what options are commercially available from textile manufacturers.

CFD Modeling of Combustion System / Slinger Combustion of Gas Turbine Engines

The proposed research program involves the development of a slinger combustion system for small gas

turbine engine application. Development of slinger combustion system allows eliminating individual

nozzles in the combustor, simplifying fuel system, reducing combustor temperature pattern factor,

more uniform combustor flowfield that leads to lowering engine emissions.

A relatively new combustion approach will be intensively studied to develop and proof the Slinger

concept with the collaborating partners, the Indian Institute of Technology Kanpur (IIT Kanpur) and

National Aerospace Laboratories (NAL) in Bangalore, India. Eventually a prototype will be built with

various detailed measurements. The main objective of the work at Université Laval is to study

numerically this new concept and to integrate spray measurements into CFD calculations to improve

their reliability in terms of more accurate results to provide the best predictive tool with the required

technological advantage of any innovative progress.

Modeling of membrane fatigue in fuel cells

Polymer electrolyte membrane fuel cells are a clean source of energy and a promising candidate for replacing internal combustion engines in automotive applications. However, the commercial use of fuel cells in such applications faces significant challenges on cost and durability. This project aims to address the concerns around fuel cell durability with a focus on the degradation of the polymer electrolyte membrane, a key component of the fuel cell. Specifically, the aim is to develop a deep understanding of how the typical fuel cell in-situ conditions, such as temperature, relative humidity, and cell voltage govern the chemical and mechanical degradation of composite fuel cell membranes.

Sillage de systèmes de propulsion en soufflerie

Les véhicules aériens à décollage et atterrissage vertical (VTOL en anglais, pour Vertical Take-Off and Landing), grâce à leur capacité au vol stationnaire, sont particulièrement bien adaptés aux applications d’observation et d’inspection de structures, mais leur efficacité énergétique en vol “de croisière” est généralement médiocre. Ce travail porte sur l’étude expérimentale des caractéristiques du sillage d’un systèmes VTOL en soufflerie. Les essais en soufflerie permettront de mesurer les champs de pression pariétale autour du profil et de mesurer les forces et le moment agissant sur les hélices. Des visualisations de l’écoulement autour du profil faciliteront la compréhension des résultats obtenus et apporteront des éléments nouveaux.

Amélioration des performances d’Automation Studio TM et communication avec des systèmes embarqués

Automation Studio™ offre un environnement convivial pour la simulation. Il peut échanger de l’information avec l’environnement externe via OPC. Toutefois les logiques de contrôle restent internes et ne donnent pas la possibilité de transfert vers un contrôleur industriel cible. Une communication avec l’environnement externe donne un avantage à Famic technologie pour interagir avec les contrôleurs industriels ou avec les PLC. La génération de code est la solution la plus naturelle pour optimiser l’opération de la plateforme cible. Deux objectifs seront visés dans cette demande et se déclinent en 3 différents projets. Le premier objectif est relié à la génération du code pour cibler des plateformes temps-réel (microcontrôleur, PLC, …) et le second vise à développer une nouvelle boite à outils pour la modélisation et simulation robotique. L’objectif du premier aspect est donc de développer un nouveau module de génération de code pour 2 plateformes cibles : un microcontrôleur de type Arduino et un PLC. Le deuxième est celui de développer un nouveau module de robotique.

Impact of garlic supplementation and Residual Feed Intake profile on the occurrence of liver abscesses in feedlots

Feed efficient finishing cattle with none or less severe cases of liver abscesses (LA) will improve the competitiveness and sustainability of the beef industry. The antimicrobial properties of garlic have not been evaluated towards mitigating the severity of LA. Studies evaluating the association between feed efficiency and LA and the integration of emerging feeding strategies such as garlic supplementation are sparse. This project will evaluate the relationship between garlic supplementation strategies and residual feed intake (RFI) profiles on the severity of LA in finishing cattle. The project will recruit approximately 200 feedlot steers for the 2-yr project where each steer will be assigned to 1 of 4 garlic-supplementation groups. Individual feed and supplement intake will be measured while the RFI profile of each steer will be calculated. The finished steers will be harvested at an approved facility where LA scores will be collected by an expert assessor. This project will determine the associations between performance and LA scores.