Innovative Projects Realized

Research for Innovative Mining Methods

Narrow-vein steeply dipping deposits are challenging to mine economically because they are poorly oriented for surface mining, and underground mining normally requires development of extensive underground infrastructure before mining the vein. Memorial University is currently collaborating with Anaconda Mining for the development of innovative narrow-vein mining (NVM) technology to mine several of these deposits currently held by the company e.g. the Romeo & Juliet Deposit. The research proposed in this MITACS Accelerate Cluster will fund graduate students and post-doctoral fellows to investigate numerous aspects of the proposed narrow-vein-mining method. The proposed research activities and internships will be done by a multi-disciplinary team of graduate students, post-doctoral fellows and faculty supervisors with backgrounds in several engineering disciplines (mining, drilling, mechanical, electronics and civil) and earth sciences (geophysics).

Visual Recognition for Large-Scale and Weakly-Labelled Video Data

The main objective of this project is to investigate, develop and evaluate state-of-the-art computer vision and machine learning techniques, which are suitable for accurate modeling and recognition from large-scale video datasets that are weakly labeled. In particular, we will focus on the learning of visual recognition models for an application area of interest to SPORTLOGiQ Inc. – person re-identification for monitoring and tracking of player, activity recognition and group behavior understanding, and player and team performance evaluation in sports games. Learning recognition models in such cases typically leads to complex and ill-posed optimization problems, where video data sets are weakly-annotated. The recent years have witnessed substantial technical advances in areas such as deep learning (e.g., convolutional and recurrent neural networks), transfer and weakly-supervised learning, information fusion and distributed optimization, which promise to address such complex visual recognition problems, previously thought intractable. TO BE CONT’D

HPGR Model Development for Different Ore Classes

The research program is aimed at developing novel test procedures and models for High Pressure Grinding Rolls (HPGRs) for comminution of metal ores classified by geology and physical properties. The HPGR is an energy efficient technology and there are presently no accepted small scale tests for sizing the technology for large scale operations. The study will improve the accuracy of test methods that were developed at UBC to replace conventional pilot scale tests that require large amount of sample that are expensive and often impractical to obtain. The models will be developed for processing a range of ores. The methods and models represents a new tool for the design, evaluation and optimization of HPGR based comminution circuits. The results of the study will also support the advancement of an energy efficient technology.

Interactive Natural Language Control for Multi-Robot Systems

Grain consumption patterns, their respective nutrient contribution and related health outcomes in Canadians

More than 41 percent of field crops produced in Canada are consumed within this country. However, there is little information available about the common consumption patterns of grain-based foods among Canadians as well as the health outcomes associated with different degrees of grain-based food consumption. Using the most recent Canadian Community Health Survey (CCHS) released in July 2017, this study investigates the data on consumption pattern of grain-based foods and contributions of grains to Canadian diet, health and wellbeing. The data will also be analyzed to understand the contributions of specific grain constituents such as dietary fiber and minor components such as minerals and vitamins to diet, health and wellness of Canadians. The results of this research will benefit the partner organization, consumers and policy makers by providing information about the status of grains consumption in Canada. TO BE CONT’D

Configuration and analysis of engineered biochar as filtration material for effluent wastewater polishing

This research project involves the design of a novel biofiltration system made up of biochar derived from forest wastes and municipal sludge, and its application for sewage treatment prior to discharge into the St. Lawrence River. The aim is to remove pollutants of emerging concerns such as pharmaceuticals, which are not presently eliminated by conventional treatment systems, and further improve the quality of the treated effluent. Outcomes of this project will enable restoration and water quality improvement of the St. Lawrence River as well as broaden research and development in biochar application. The research outcome will provide the partner organization with a new option to propose to its customers involved in designing and operating wastewater treatment infrastructures. and to commercialize the system in the future. The designed biofiltration system may be eventually replicated and applied to other wastewater treatment facilities discharging treated sewage into the St. Lawrence River.

Investigating the effect of a novel regularization technique for neural networks

Despite the fact that neural networks have been widely applied in practice, training such networks can suffer from slow convergence, poor local minima and some other difficulties such as catastrophic forgetting. Such shortcomings severely undermine the applicability and usefulness of neural networks. The objective of this project is to identify the reasons behind such difficulties in training and to investigate the effect of novel regularizations. We will analyze the strengths and shortcomings of existing algorithms and benchmark recently proposed regularizers for neural network training. The performance of these techniques will be evaluated for supervised tasks.

Securitized Tokens as a Service (STaaS)

This research study aims to investigate the potential and performance of a novel Securitized Tokens as a Service platform. Blockchain is the distributed ledger of verified transactions, and smart contract is the programmable part of the blockchain which can automate more complex transactions. We can define the tokens on top of the blockchain platform; and actually, these tokens can stand for anything in real business world which can be transferred. In addition, smart contract is used to define the required rules of the business market through the blockchain. Using our suggested platform, companies can define their own tokens and use them for variety of purposes. These tokens, for example, can be used as a utility tokens, or as a tool for share management, asset management, and fund raising.

A Device for the Effective Use of Composted Chicken Manure in Hydroponics Systems

Innovative agricultural methods may be able to provide affordable food and vegetables in Canada’s North. In support of this, Choice North Farms has partnered with PolarPonics to develop a ‘PoultryPonics’ facility that will reduce production costs by integrating chicken and hydroponic production with an automated composting system. To effectively do this, they will require optimal methodologies for composting chicken manure which will need a high-precision, high-frequency sampling device to measure the nutrient content of the manure solutions that are produced. Peter Tikasz at McGill University has performed preliminary work on composting manure and on the development of device composed of Ion Selective Electrodes (ISEs) to measure the quality of the manure solutions. Peter will continue this work in conjunction with northern agricultural partners which will lead to the development of a northern-tested prototype device that will have applicability for any horticultural operation that utilizes manure-based compost.

Developing Intelligent Approaches for Factories of the Future – An intelligent strategy for robotic part inspection in advanced manufacturing – Year two

This project explores incorporation of artificial intelligence tools for advanced control techniques in order to improve quality on industrial production lines. Particularly, the novelty of this project suggests the use of the recent embedded computer systems and icloud computer assistance. This strategy is currently used for commercial products (i.e., application related to communication). However, this project explores how to use AI and dynamic optimization tools by identifying both dynamic structure and parameters. The purpose of exploring such advance data analysis for industrial automation is meant to predict quality inspection based on high complex identification patterns during production. The complexity of advance product manufacturing impose important limitations for humans to process. Human inspection affected by both temporal processing and quality consistency impacts directly on safety and cost concerns. TO BE CONT’D

Developing Intelligent Approaches for Factories of the Future – An intelligent strategy for robotic part inspection in advanced manufacturing

Human inspection of high quality components in advanced manufacturing, automotive and aerospace applications is challenging, as detecting imperfections that are variable can result in inconsistent part quality decisions. These inspection tasks are very repetitive. Achieving zero incorrect part quality decisions that are required in high tolerance assembly, as well as in critical process operations becomes almost impossible for humans to perform. Currently, robotic systems are being used at various stages of producing parts, as well as in the final stages of manufacturing such as part removal, stacking and other finishing operations. This has resulted in improving overall efficiencies throughout the part production cycle. Robots are being used for quality inspections, and is in its embryonic stage. Robot operators perform time consuming repetitive setups and path planning of its motion to execute these inspections. This inspection procedure has to be repeated for different dimensional parts and quality metrics. TO BE CONT’D

BC Pulp and Paper Bioproducts Alliance: Upgrading black liquor lignin through fractionation and chemical modification

Lignin represents the largest reservoir of natural aromatic compounds available on earth. It is a potential substitute for a range of chemicals currently derived from petroleum, but product development and identification of market opportunities remain challenging. The objective of the research is to utilize industrial softwood kraft lignins from the black liquor of British Columbia mills, process them into uniform lignin streams with known performance specifications by fractionation and chemical modification. The outcome of the research will be to generate and characterize various grades of lignin, with different degrees of reactivity and size, offering tailored lignins to be marketed to different industrial sectors. Finally, a special emphasis will be given to identifying and leveraging any specific lignin properties that are unique to the BC resource and/or applicable to products of interest to BC industries (e.g. wood products, mining, food).