Machine Learning for Practical and Scalable Regression Test Selection and Prioritization

In the context of systems with a large codebase, Continuous Integration (CI) significantly reduces integration problems, speed up development time, and shorten release time. While regression testing is widely practiced in the context of CI, it can be time-consuming and resource intensive for large codebases where the execution of test cases is time and resource intensive.

Unit Commitment Problem Integrated with Plug-in Electric Vehicles and Renewable Energy Sources

Electric Power is crucial challenge for every nation. Canada is a developing country and the development of any country is nudged by the amount of electricity used by that country. The size of the power system is growing exponentially due to heavy demand of electricity in all the sectors viz. agricultural, industrial, residential, commercial and charging of Plug-in EVs/HEVs etc. The ever-rising demand for energy has led professionals to look out for renewable sources of energy and with its growing influence.

Development of a novel methodology for predicting reliability of battery energy storage in marine applications

Battery energy storage systems (BESS) play a critical role in the electrification of the transportation industry. Despite many recent advancements in lithium ion chemistry at the cell-level, prediction of battery system lifetime remains a challenge at the system-level. Corvus Energy is the world’s leading supplier of safe, innovative, and reliable energy storage solutions for all segments in the maritime industry. This project will study the reliability of Corvus Energy's market-leading Orca Energy BESS to better predict reliability of future BESS.

Weakly supervised representation learning for sequential and composite actions.

Camera enabled AI-based personal assistants will need to recognize human actions in order to be safe and effective. Current machine learning approaches for action recognition require extensive datasets of annotated videos that depicting the actions to be recognized. Such datasets are expensive to acquire. The goal of this project is to decrease the annotation required to train viable action recognition systems.

Discover anomaly signatures from time series data of telecommunication networks

Failures in a telecommunication network harm the communication quality. Once happened, if the system cannot solve it by self-healing, such anomaly may even result in serious problem and result in massive economic loss. In this project, we will design and develop a system to predict these failures in advance using the status values of the hardware facilities.

Design of real-time a localization algorithm for high-flow environments on embedded processor

The deployment of tidal turbine equipment in the Bay of Fundy can potentially impact marine life in the environment. There is a large population of harbour porpoises that can be affected by this equipment, and it is important to detect their presence and locate them. In this work, a tool to visualize harbour porpoises in real-time will be developed so that it can be readily available to visitors at the Fundy Ocean Research Center for Energy Visitor Center in Parsborro. For this purpose, a signal processing tool will be developed on a high-speed processor.

Integration of Hybrid Distributed Resources in Three Different Systems With and Without Global Adjustment

The use of hybrid distributed energy resources (HDERs) has grown rapidly during the last decade as a way to reduce the stress placed on the utility grid by society’s ever-expanding power consumption. They consist of fossil fuel generators, battery energy storage systems, and renewable energy systems and can be designed to interact with the grid in various system architectures to supply end-user loads. It is important to size and integrate HDERs appropriately so that they can meet load demands, while still being cost-effective.

Self-powered nasal wearable sensor for diagnosis of Coronavirus 2019

Corona Virus 19 creates adverse effects on the global health system leading to economic crisis. Although trials to develop vaccines are underway, it would take a while for the general public. Wearable sensors have been receiving increased attention toward monitoring the wearer’s well-being and playing a significant role in the healthcare industry. The upper respiratory tract is the first site for hosting and transmission of COVID infection. Therefore, developing a wearable nasal sensor will achieve in-home continuous monitoring of virus in the nasal cavity.

Developing low-intensity pulsed ultrasound circuits to enhance cellulosic bioethanol yield for renewable energy

As the world finite supply of fossil fuels diminishes, viable alternatives are needed to empower the transportation, heating, and industrial needs of modern civilization. Lignocellulosic ethanol promises a means to transform abundant plant wastes into an energy-dense, carbon-neutral liquid combustible, compatible with current infrastructure. Unlike the first-generation biofuels derived from corn or sugarcane, lignocellulosic ethanol does not infringe on food production, although extra steps are needed to convert bulk cellulose into simple sugars suitable for fermentation.

Portable Device for COVID-19 Nucleic Acid Tests

We have seen the recent surge of COVID-19 because we cannot afford another massive shut-down. A portable and timely detection device for the COVID-19 nucleic test is more desirable. In this proposal, we propose to develop an impedance-based nucleic acid testing device. Our LOC devices are low cost, user friendly alternatives to centralized lab tests. By partnering with Hidaca Ltd., our ultimate goal is to have this made-in-Canada technology available on the market as soon as possible to benefit Canadians and the Canadian economy.