The Royal Institution for the Advancement of Learning/McGill University

Published purpose

The abrasive nature of ores in mining operations causes significant wear to ground engaging tools during extractive processing which results in high maintenance costs and unscheduled process disruptions. Advances in integrating various high-performance materials to low-cost parts has faced manufacturing barriers for mining requirements due to key gaps and challenges in using laser powder direct energy deposition (DED) to volume build-up and repair complex geometries. This Project will use Artificial Intelligence (AI) to enhance DED so as to decrease process development iterations while maintaining high-accuracy deposition of fine features at fast process speeds. To feed AI analysis, existing sensors and innovative technologies, such as laser scanners and laser ultrasonics, will be integrated as a system to monitor process parameters, part geometry and quality. A Machine Learning (ML) computational engine will be used to develop AI models/algorithms to enable quality prediction and parametric optimization through an AI-enhanced adaptive toolpath planning and evaluated on 3 relevant demonstrators (rock crusher tooth, ground engaging bucket tooth, and slurry pump impeller).