In-process Monitoring and Optimization of Additive Manufacturing With Sensors And Artificial Intelligence/Machine Learning for Enhancing Future Systems and Enabling Remote Printing

Additive manufacturing (AM), particularly Fused Filament Fabrication (FFF), offers a versatile approach to producing components with intricate geometries. However, maintaining print quality and mechanical performance remains challenging due to factors like layer bonding and extrusion consistency. This project focuses on developing advanced issue detection and monitoring software for FFF processes, leveraging real-time data from various sensors. Current Addiguru software utilizes optical imaging to identify common issues such as nozzle clogging and extrusion errors. Building on this, the project incorporates infrared sensing to monitor temperature profiles, enhancing defect detection capabilities. Machine learning algorithms will integrate data from optical, infrared, and other machine sensors (e.g., filament diameter, feed rate) to optimize printer settings dynamically, improving print quality and process efficiency. The proposed solution aims to benefit the Army, Department of Defense (DoD), and commercial manufacturers by reducing downtime, enhancing efficiency, and ensuring higher product quality. Real-time monitoring aligns with industry trends towards automation and smart manufacturing, potentially offering significant cost savings. A recent survey indicates potential savings of 20-40% through real-time AM monitoring. This Phase I effort will prototype a multi-sensor, AI-driven system to demonstrate real-time parameter optimization and defect detection, with plans for further enhancement using acoustic sensors in Phase II. The technology's commercialization could open a market opportunity exceeding $30 million annually, distinguishing itself from competitors by its comprehensive sensor fusion and AI-based corrective algorithms.