Digital Twin for Additive Manufacturing and Smart Manufacturing Education

Abstract The use of additive manufacturing (AM) or 3D printing has greatly increased in various industry sectors in the last five years. As such, it is the responsibility of educational institutions to support efforts to embed this technology in their curriculum to prepare students for the future. One crucial task is to teach students how to use modern technology to evaluate the quality of AM parts because AM has not reached the point of competing with traditional manufacturing in terms of surface finish and repeatability. Moreover, the printed parts are often treated as black boxes with invisible defects, such as pores and cracks. Such non-transparency significantly challenges the qualification and certification of additively manufactured parts. In this paper, we present a term-long project designed for a new AM course offered at University A to demonstrate the challenges and benefits of AM. At the beginning of the project, we teach students how to develop a digital twin of the 3D printing process. This is later used to provide in-situ monitoring of the process and visualize the internal defects towards predictable quality control of the printed parts. Students are exposed to multi-model sensors to monitor the printing processes, including a microphone for acoustic emission, XYZ stage encoders for the print head position, and nozzle temperature, a close-up camera mounted on the printhead, and a second camera to image every layer's surface. When the digital twin of the printing process is built, computer vision algorithms will be used to detect the defects based on the real-time camera. Through changing process parameters, students will learn how to optimize part quality and how defects in parts can be eliminated using the digital twin. This paper presents a reference implementation with technical and pedagogical details for the education community.