Projets novateurs réalisés

Optimal actuator location for semilinear systems

Consider the temperature distribution in a classroom; to change it, we need to install a heater somewhere in the classroom. With a heater, we can raise the temperature, as well as change its distribution such that less heat flows to doors and windows, and more to seats. The location of the heater in the room significantly affects performance. There will be a location that minimizes spending on heat. In this project, we try to answer such similar questions for wide class of physical systems not just heating processes, for example fluid flow, coupled flow/diffusion, structural vibrations. The focus is on nonlinear models as the linear cases have already been studied. Example of nonlinear models are the railway track model, and also many diffusion and flow processes. The linear theory cannot be used to investigate such models.

Spherical harmonic representation of sound field using a velocity microphone for source localization and source separation

The rapid emergence of voice interfaces in our everyday devices has driven the need to develop audio systems that are robust to noisy environments. Traditionally, arrays of pressure based microphones are used to listen into a specific sound and block out the surrounding noise. However, this comes at a tradeoff of increased size and decreased audio quality. Recently, Soundskrit has developed a new type of microphone that can directly measure the particle velocity of a sound field. This microphone inherently measures the vector components of the incoming sound, providing directional information of the acoustic environment. In this project, we will investigate how information on the sound’s particle velocity can be leveraged to perform accurate sound source localization and separation while maintaining a relatively small footprint. Such a system could enable new features in a variety of consumer devices.

TRICHUM: Translating Research into Innovation for Cannabis Health at Université de Moncton

Supplying cannabis, and cannabis products, to a legalized retail market represents a major economic opportunity and has been identified as a growth opportunity under New Brunswick’s Economic Growth Plan (Province of New Brunswick, 2016). Indeed, the retail market value for recreational cannabis in Canada is expected to reach $8.7 billion annually (Deloitte, 2016); however, current production capacity will not meet the anticipated demand. Because the commercial production of cannabis has been illegal, and production has been largely artisanal and clandestine, cannabis has not benefitted from the application of biotechnologies (particularly genomics) that have led to massive increases in yields and sustainability of other agricultural production systems. In this innovative project, and working in close collaboration with OrganiGram Inc., a concerted research-driven approach will be developed to support the development and the growth of the cannabis industry in Atlantic Canada through the application and adaptation of novel biotechnologies. TO BE CONT’D

Les effets de l’intelligence compétitive et de la capacité d’absorption sur la performance de l’innovation des PME canadiennes (extension)

L’incertitude et la complexité de l’environnement externe incitent les entreprises, notamment les PME à performer en innovation afin de développer un avantage concurrentiel et de rester compétitives. L’innovation requiert des idées créatrices. Toutefois, la majorité des PME ne disposent pas de ressources suffisantes pour la recherche et développement ce qui rend sa survie difficile. L’intelligence compétitive peut combler ce manque de ressources par la collecte d’information pertinente au processus d’innovation. Cependant, les PME doivent être dotées des capacités et des compétences permettant d’aller au-delà de l’acquisition de l’information, rendant possible la conversion de l’information externe en connaissances utiles et applicables au processus d’innovation. Cette recherche vise donc à étudier les effets des processus d’intelligence compétitive et de la capacité d’absorption sur la performance de l’innovation au sein des PME québécoises. TO BE CONT’D

A Unified Framework for Remote Monitoring the State and Performance of Photovoltaic Power Plants

Energy produced using the solar radiation as the source is one of the most prominent parts of the clean energy mix. More than 300 GW of photovoltaic systems (PV systems) of different size supply the daily needs of millions of families and industries around the world. Photovoltaic panels are installed on the roof of houses and buildings, or they constitute large-scale photovoltaic power stations. Monitoring the energy production of the PV systems has a crucial role in both predictability and maintenance. To effectively monitoring the systems, the measures of the energy produced need to be remotely read and compared with the estimate production values that are calculated considering the solar radiation. With these data, many views and indexes related to the operating status of the plants can be given. Also, suggestions on how better use the self-produced energy can be supplied. TO BE CONT’D

Effects of Water Saturation and Foamy Oil Stability Agents on Foamy Oil Flow and the Cyclic Solvent Injection Process

This project is to perform systematic studies to better understand effects of water saturation and foamy oil stability agents on foamy oil flow and the cyclic solvent injection process and provide fundamental parameters for field-scaled prediction. Equilibrium and non-equilibrium PVT tests will be conducted to monitor and analyze the foamy oil performance under effects of the water saturation through a live-oil brine system with water emulsification. Also, pressure depletion tests will be performed to investigate the foamy oil performance under effects of the water saturation in porous medium. Well designed constant concentration expansion (CCE) tests will be carried out to examine effects of the foamy oil stabilizer and select an optimal one. Numerical simulation models will also be built to perform history matching and predicting study. Then scaling criteria from laboratory tests to field applications will be established based on experimental results, numerical simulation models and field data.

Accelerate Transaction Latency of Pool Mining in Cryptocurrency Networks

In this project, using such mainstream cryptocurrencies as BitCoin and Ethereum as representatives, the intern will analyze the transaction collection strategies of their mining pools, and then collect transactions and the corresponding blocks data to build a large dataset, from which the computing power of different mining pools and their proportions will be analyzed, together with the transaction latencies of pool mining. We will also identify potential enhancement through the analysis and measurement, particularly on energy and delay optimization. Coinchain is a BC-based startup company focusing on advanced cryptocurrency and blockchain technologies, and their application in industrial and commercial scenarios. It delivers global enterprise-level blockchain solutions to leading companies worldwide, and provides one-stop customized services such as product and information platforms, as well as smart contracts and trading platforms. TO BE CONT’D

Learning PDF Document Structures using Recursive Neural Networks

Portable Document Format or PDF is the de facto standard for presenting textual-visual content. In this project, we aim to develop a machine learning framework for PDF document understanding. Despite the recent proliferation of deep learning-based methods for the analysis and processing of natural images, there have been considerably less efforts on designing similar approaches for highly structured data such as documents. Our project will explore two novel ideas. First, we will develop a structured and organizational representation of PDF documents which is built on labeled content blocks (e.g., heading, figure, list, caption, etc.). Second, we will investigate how recursive neural networks (RvNN), one type of deep neural networks that have been utilized to language parsing, can be adopted and formulated for learning PDF document structures.

Energy Simulation and Lifecycle Costing of Advanced Glazing Systems

Highly efficient glazing, such as those offered by ECO-Insulating Glass, provide a significant

opportunity to reduce the energy consumption of buildings. This project quantifies the energy

savings, cost savings and reduction in heating and cooling demand resulting from using ECO

glazing in five major Canadian cities for a typical medium sized office building, a home and a

school. Four window options are evaluated and compared to windows, which just meet the

building code requirements. The effect of using higher insulation levels in conjunction with

ECO glazing will also be evaluated. The optimal scenarios will be selected from a total of 225

possible scenarios, using an advance building simulation software program from the U.S.

Department of Energy. The resulting values and trends will be useful for ECO glazing sales,

market analysis and for the shift towards higher performance buildings in Canada.

Structural Inequality, Resistance, & the Future of a Continent: Examining Colonialism’s Toxic Legacies

Last year, the world’s leading health researchers found toxic pollution to be the greatest cause of disease and premature death in the world today. However, while the findings acknowledge that deaths caused by toxins to be most prevalent among marginalized groups, there is no mention of socio-historical structures in producing these toxic patterns. Through this research I examine the relationship between structural violence and toxic geographies in relation to legacies of colonialism across North America and will conduct a comparative case study that looks at the Mount Polley Mine Disaster (MPMD) and the Flint Water Crisis (FWC) to think about solutions for addressing toxicity, anew. TO BE CONT’D

Quantitative micro-computed tomography for cartilage and joint mechanobiological measurement

This research aims to measure complex cartilage and joint mechanobiology, the dynamic interaction between cells and the physical factors, such as force, in their environment. This will be done by combining quantitative micro-computed tomography (microCT), image guided mechanical evaluation, and quantitative morphometric analysis. The aim is to combine these approaches into a single platform to provide longitudinal, quantitative, and in vivo measurement. Combining bone and cartilage imaging into a single modality is important since the two are tightly inter-regulated and communicate and respond to their loading environment among each other. To accomplish this, initial steps are to establish longitudinal contrast-enhanced CT protocols for use in time-lapse microCT imaging and integrate image-guided mechanical evaluation. This work will provide comprehensive all-in-one measurement of joint and cartilage mechanobiology. TO BE CONT’D

A Deep Learning Approach to Soft Sensor Design and Process Optimization for an Industrial Nickel Extraction Process

The objective of this project is to use artificial intelligence (AI) approaches to solve complex industrial problems. The two biggest advantages of AI-based approaches are the ability to continuously learn and also learn adequately from historical data. Traditionally, many process information are unmeasurable during live operations because of instrumentation limitations. Also, plants are not sufficiently optimized to maximize production quality, while minimizing waste. Using AI-based approaches, we can develop complex non-linear models from historical to predict the unmeasurable process information. The models are also continuously learning from the new data coming into the plant. To optimize the process operations, another family of AI algorithms called reinforcement learning will be used. These algorithms will learn the whole process, including what happens when each process variable is changed. With this knowledge, reinforcement learning can then provide the optimal sets of inputs to maximize the plant productivity, while minimizing its waste.