Projets novateurs réalisés

Modélisation de l’évolution de la température au cours du mélange de pâtes de boulangerie

Lors de la réalisation de pâtes de boulangerie, les ingrédients de base sont mélangés dans de grandes cuves afin de créer un pâton qui sera ensuite mis en forme et cuit. Lors du mélange, les ingrédients se lient pour passer de poudres et liquides à une pâte épaisse et visqueuse. La température de la cuve doit être soigneusement contrôlée afin de maximiser l’action des levures. Les systèmes de refroidissement actuels ont été conçus à force d’essais/erreur, et dans les faits la température peut localement être trop élevée, rendant la pâte impropre à la cuisson. De grandes quantités de pâtes sont ainsi jetées par manque de compréhension fine de la répartition de la température au cours du mélange. Le présent projet propose de palier ce manque.

La nature très changeante de la pâte au cours du mélange ainsi que l’impossibilité par l’instrumentation de connaître la température à l’intérieur de la pâte ne permettent pas de considérer une étude uniquement expérimentale. Dans ce projet, il est proposé d’utiliser des calculs par ordinateur, couplés à des vérifications sur certains points de température précis en surface de la cuve, afin de répondre à cette problématique industrielle et ainsi éviter les pertes alimentaires.

Distributed and Decentralized Optimization for Electricity Consumption Flexibilities and Network Constraints

In the context of the “Smart Grid”, the exploitation of electricity consumption flexibilities will be a key lever in the optimization of the electric system. It can help to limit power demand peaks, increase the share of renewable energy and limit network infrastructure investments. Consumption flexibilities, which are offered by many residential consumers, are distributed resources: from an optimization perspective, this leads to the development of new decentralized methods, often referred to as Demand Response.
A decentralized view point is also necessary to consider the various actors interacting in the system : smart consumers, energy aggregators, distribution and transmission system operators and electricity markets.
In this project, the objective is to study this complex system considering the different actors in interaction, to propose optimized and decentralized coordination mechanisms, and to evaluate the potential benefits of Demand Response techniques related to network investments. The project will call for different techniques, from operations research, continuous optimization, game theory and data analysis.

Étude de la composition phytochimique de 15 cultivars d’argousier cultivés au Québec dans un objectif de développement de marché

L’argousier est un petit fruit nordique bien adapté au climat boréal. Son fruit, l’argouse, se distingue des autres petits fruits par la diversité de ses composés bioactifs. Il est notamment riche en caroténoïdes et en polyphénols. Notre projet vise à comparer la composition chimique des fruits de 15 cultivars d’argousier actuellement cultivés au Québec. Il se divise en trois objectifs : 1- caractériser le profil physicochimique des argouses (taux d’humidité, teneur en huile, teneur en sucres, teneur en vitamine C, contenu en caroténoïdes et polyphénols totaux, potentiel antioxydant); 2- établir des fiches techniques regroupant ces données de caractérisation ainsi que leurs caractéristiques morphologiques et agronomiques, dans le but de fournir aux producteurs une orientation du marché à privilégier pour chaque cultivar (alimentaire, nutraceutique et/ou cosmétique); 3- Réaliser une analyse métabolomique non-ciblée afin de différentier les cultivars selon leur profil phytochimique.

Optimiz AIOps data normalization and visualization

Approximately 70% of data collected by ITOps is not actionable, either due to lack of timeliness or context, and real-time normalized data stored in a data lake is needed to produce actionable data. Also, AIOps tools are often poorly configured and lead to alert fatigue. The benefit to the partner organization is that they will be able to capture actionable AIOps data that can be normalized for Artificial Intelligence (AI) and Machine Learning (ML) analysis. Based on the analysis done by AI and ML interventions can be automatically applied to an enterprise IT system.

Developing a Submaximal EMG-Assisted Subject-Specific Musculoskeletal Model of the Spine: Combined Computational and Experimental Study – Year two

There is no direct way to measure spinal loads, and existing methods are indirect and invasive; alternatively, musculoskeletal models can predict spinal loads accurately and economically. Existing musculoskeletal models of spine substantially simplify the spine and/or use optimization algorithms to estimate muscle forces. Thus, we aim to develop a subject-specific musculoskeletal model of the spine which has realistic representation of the spine (by using a detailed finite element model) and is driven by measured biological inputs (kinematics and electromyography). Kinematics and electromyography of 48 individuals with/without back pain are collected and used as model inputs. The proposed model addresses some of the fundamental limitations and shortcomings of existing models, provide us with new outputs (e.g., stress/strain field in intervertebral disc, novel failure analysis), and can be used to analyze individuals with back pain; therefore, it can shed light on likely biomechanical roots of back pain. This project is in collaboration with IRSST (which aims to improve occupational safety of workers, workers rehabilitation, as well as workers recovery), and IRSST will use the outcome of this project to potentially decrease biomechanical occurrence of back pain in workers as well as to develop new evaluation and treatment protocols.

Current Waveform Parameters of CN Tower lightning Return Strokes andM-Cornponents

Although several attempts have been carried out to derive statistics about eN Tower lightning
current waveform parameters, which are essential for the design of appropriate protection
systems, the results have been limited use because of the nature of the collected data. The
absence of both the continuous and continuing currents from the collected data, made it
impossible to distinguish between return strokes, which are the most destructive components
of the lightning flash, and the M-components. The aim of this proposal is to solve this problem
with the aid of the available extensive high-speed camera records. The results of this project
will not only provide proper data to engineers dealing with lightning protection problems, but
also answer many fundamental questions about the difference between return-stroke current
impulses and those that correspond to M-components.

Developing a Submaximal EMG-Assisted Subject-Specific Musculoskeletal Model of the Spine: Combined Computational and Experimental Study

There is no direct way to measure spinal loads, and existing methods are indirect and invasive; alternatively, musculoskeletal models can predict spinal loads accurately and economically. Existing musculoskeletal models of spine substantially simplify the spine and/or use optimization algorithms to estimate muscle forces. Thus, we aim to develop a subject-specific musculoskeletal model of the spine which has realistic representation of the spine (by using a detailed finite element model) and is driven by measured biological inputs (kinematics and electromyography). Kinematics and electromyography of 48 individuals with/without back pain are collected and used as model inputs. The proposed model addresses some of the fundamental limitations and shortcomings of existing models, provide us with new outputs (e.g., stress/strain field in intervertebral disc, novel failure analysis), and can be used to analyze individuals with back pain; therefore, it can shed light on likely biomechanical roots of back pain. This project is in collaboration with IRSST (which aims to improve occupational safety of workers, workers rehabilitation, as well as workers recovery), and IRSST will use the outcome of this project to potentially decrease biomechanical occurrence of back pain in workers as well as to develop new evaluation and treatment protocols.

Culture media optimization for oocytes from peri-pubertal heifers

The use of oocytes from peri-pubertal heifers represents an important reproductive management asset, assisting with the rapid multiplication of high merit genetic material and the reduction of generation interval. However, oocytes from peri-pubertal heifers have reduced developmental capability leading to embryos of lower quality. Furthermore, the in vitro culture conditions for the embryo production were not originally developed for oocytes collected from such young animals. Therefore, problems such as inadequate oocyte maturation (cytoplasmic, nuclear and molecular) have been observed. The hypothesis of this study is that optimization of In vitro maturation (IVM) conditions will improve oocytes quality from peri-pubertal heifers enhancing overall embryo production and embryo quality. IVM media will be supplemented using various molecules known to have an effect on energy and lipid metabolism, molecules targeting different cellular components such as the mitochondria or the endoplasmic reticulum. The morphological and molecular phenotype of the embryos produced from peri-pubertal heifers’ oocytes will be analysed. The epigenetic profile of the embryos produced under those optimized culture conditions will also be assayed in order to minimize epigenetic perturbation. The overall goal of the project is to optimize IVM conditions for oocytes collected from peri-pubertal heifers to produce high quality embryo.

Modeling of combined heat and power generation in Canada

Canada ambitions to achieve a deep decarbonization of its economy by 2050, with a reduction of its greenhouse gas (GHG) emissions up to 90% relative to their 2005 levels. Such a transition from the current situation will have important impacts in particular on its energy sector. In this context, SIEMENS Canada, a manufacturer of energy systems, wishes to study the potential deployment in Canada of flexible combined heat and power (CHP) plants.

Such a potential deployment must be analyzed in the context of competitive energy markets, where different technological solutions may be used in different sectors to abate GHG emissions. The TIMES energy model, developed within the ETSAP program of the International Energy Agency, offers a systemic way to analyze the entire energy sector from the extraction of primary energy until the consumption of final energy. This model enables in particular to analyze possible evolution of the energy sector over the long run (2050 or beyond).

A TIMES model for Canada has already been developed and successfully used. Several modeling improvements can, however, be envisioned. Beyond this, the objective of this project is to assess the potential deployment of flexible CHP plants under different policiy and technological scenarios in Canada.

Validation of a novel digitized trail making test – Year two

Age-based diseases of cognitive dysfunction such as dementia and Alzheimer’s are a growing problem in Canada that already cost the healthcare system over $10 billion annually. The ability to provide the general public with ubiquitous access to iterative cognitive function testing would be extremely valuable to the individual, society, and science. LRDG has commenced development of a digital cognitive test to achieve this goal. However, before the test can be commercialized, its function must be verified, and test outcomes must be validated. In the context of this fellowship, these activities will comprise a year of user experience (UX) testing in close communication with the test developers at LRDG to refine the test to beta level. The second year will involve iterative within-subjects and cross-validation with a non-digital version of the test. This fellowship is the initial step in what is envisioned as a broader multi-year research collaboration involving neurophysiology and multiple subject populations.

Validation of a novel digitized trail making test

Age-based diseases of cognitive dysfunction such as dementia and Alzheimer’s are a growing problem in Canada that already cost the healthcare system over $10 billion annually. The ability to provide the general public with ubiquitous access to iterative cognitive function testing would be extremely valuable to the individual, society, and science. LRDG has commenced development of a digital cognitive test to achieve this goal. However, before the test can be commercialized, its function must be verified, and test outcomes must be validated. In the context of this fellowship, these activities will comprise a year of user experience (UX) testing in close communication with the test developers at LRDG to refine the test to beta level. The second year will involve iterative within-subjects and cross-validation with a non-digital version of the test. This fellowship is the initial step in what is envisioned as a broader multi-year research collaboration involving neurophysiology and multiple subject populations.

Fault Ride-Through Method for Cascaded CSC-HVDC BasedOffshore Wind Farm

The fault ride-through (FRT) capability of a wind farm is major challenge because the disconnection of
a large wind farms caused by network disturbances may lead to system instability and cascaded
failures. Recent studies mostly focused on FRT capabilities for the voltage-source converter (VSC)
based wind farms. Although, the current-source converter (CSC) based wind energy conversion system
(WECS) is considered potential candidates for the medium-voltage high-power WECS, the FRT
methods for such system has been rarely studied in the literature. Moreover, the low dc-link inductance
of the CSC based system exhibit significant challenge for the FRT capability compared to its VSC
based counterpart, where dc-link capacitance usually 4 p.u. or higher [1]. In order to overcome the
problem, this research proposes to formulate a novel FRT method for Cascaded CSC-HVDC based
offshore wind farm.