Excellence in Research: Reliable Additive Manufacturing of Advanced Metal Alloys Through AI-Enabled Real-Time Monitoring and Process Control

Developing advanced metal alloys for aerospace, energy systems, and national defense often requires lengthy cycles of material design, manufacturing, testing, and redesign. This project seeks to accelerate that process by enabling additive manufacturing systems to monitor material formation in real time and use artificial intelligence to improve manufacturing decisions during production. By integrating real-time sensing, artificial intelligence, and advanced manufacturing, the project aims to improve the quality, reliability, and efficiency of manufacturing advanced metal alloys for demanding engineering applications. Tungsten–tantalum alloys will serve as a challenging demonstration material because of their potential for use in extreme environments and the difficulty of producing them with consistent internal structures. The project will strengthen research capacity at Prairie View A&M University through collaboration with Texas A&M University while creating interdisciplinary research and education opportunities spanning computing, materials science, and manufacturing. Undergraduate and graduate students will participate in research, curriculum development, seminars, workshops, and professional training, and outreach activities will introduce high school students to advanced manufacturing, artificial intelligence, and data science through hands-on learning experiences that expand participation in science, technology, engineering, and mathematics. By accelerating materials innovation and advancing intelligent manufacturing technologies, the project will promote the progress of science, strengthen the advanced manufacturing workforce, enhance economic competitiveness, support energy security, and advance technologies important to the national defense. The technical objective is to develop an artificial intelligence-enabled framework for real-time monitoring and process control during electron beam powder bed fusion additive manufacturing of advanced metal alloys. Backscattered electron and x-ray detectors will be integrated into an open-architecture manufacturing platform to monitor changes in material properties and microstructural evolution during fabrication. Tungsten–tantalum alloy samples produced under varied processing conditions will be characterized to establish relationships among manufacturing parameters, sensor measurements, alloy composition, and resulting microstructures. Artificial intelligence models will learn compact representations of complex microstructures, predict microstructural evolution from manufacturing and sensor data, quantify prediction uncertainty, and recommend processing strategies that achieve desired material properties. The initial implementation will employ Bayesian neural networks to provide trustworthy, uncertainty-aware predictions, while the framework is designed to incorporate emerging artificial intelligence models as the research progresses. Sequence-learning models will identify adaptive scanning strategies for closed-loop process control. The resulting framework will be evaluated across multiple alloy compositions and component geometries and integrated with the manufacturing platform for real-time optimization. The project will establish a broadly applicable foundation for artificial intelligence-enabled monitoring and intelligent process control in additive manufacturing, while providing datasets, artificial intelligence models, and sensing methodologies that accelerate the reliable manufacturing of advanced metal alloys. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.