Grant: Alliance Grants

Shot peening through the introduction of compressive residual stresses is essential to enhance metal component fatigue performance in aerospace applications. However, this process concurrently increases surface roughness and compromises the fatigue resistance. Subsequent vibratory finishing mitigates this roughness, though at added complexity and cost. Integrating both shot peening and vibratory finishing into vibratory peening offers efficiency and cost-saving benefits, yet process optimization remains challenging due to numerous operational parameters and a lack of suitable numerical tools. This NSERC/Safran/Vibra Finish research proposal aims to develop numerical frameworks to optimize vibratory peening processes on complex aircraft parts, focusing on the Ti-6Al-4V titanium alloy. We will integrate conventional Finite Element Method (FEM) techniques with machine learning approaches to design frameworks capable of optimizing machinery design and processing parameters. Our primary objective is to create a tailored Discrete Element Method (DEM) - FEM model to accurately simulate the vibratory peening processes. We aim to optimize these processes by leveraging advanced numerical techniques and machine learning, particularly graph neural networks. Through this work, we aim to advance the practical application of vibratory peening technology in aerospace manufacturing. The project aims to enhance the durability and reduce the weight of aerospace parts, thereby minimizing the environmental impact of air travel and boosting industrial competitiveness. Safran's subsidiaries in Canada, including Safran Landing Systems and Safran Helicopter Engines, are poised to benefit significantly from the research outcomes. The development of new numerical tools for process optimization is anticipated to accelerate the integration of new aircraft parts, fostering innovation and adoption of advanced manufacturing practices.