Machine learning based prediction of melt pool morphology in a laser-based powder bed fusion additive manufacturing process

AbstractLaser-based powder bed fusion (L-PBF) has become the de facto choice for metal additive manufacturing (AM) processes. Even after considerable research investments, components manufactured using L-PBF lack consistency in their quality. Realizing the crucial role of the melt pool in controlling the final build quality, we predict the morphology of the melt pool directly from the build commands in an L-PBF process. We leverage machine learning techniques to predict quantitative attributes like the size as well as qualitative attributes like the shape of the melt pool. The area of the melt pool is predicted using an LSTM network. The outlined LSTM-based approach estimates the area with 90.7% accuracy. The shape is inferred by synthesising the images of the melt pool by using a Melt Pool Generative Adversarial Network (MP-GAN). The synthetic images attain a structural similarity score of 0.91. The precision and accuracy of the results showcase the efficacy of the outlined approach and pave the way for real-time monitoring and control of the melt pool to build products with consistently better quality.Keywords: Additive manufacturingmachine learningin-situ monitoringL-PBFLSTMGAN AcknowledgementsThe views and conclusions contained in this document are those of the authors and should not be interpreted as representing the official policies, either expressed or implied, of the Army Research Laboratory, or the U.S. Government. The U.S. Government is authorised to reproduce and distribute reprints for Government purposes notwithstanding any copyright notation herein. We also thank Brandon Lane (Mechanical Engineer, NIST) for generating and providing the melt pool data sets for our experiment.Data availability statementThe data that support the findings of this study are available from the corresponding author, RR, upon reasonable request.Disclosure statementNo potential conflict of interest was reported by the author(s).Additional informationFundingThis research was sponsored by the Army Research Laboratory and was accomplished under Cooperative Agreement Number W911NF2020237.Notes on contributorsZhibo ZhangZhibo Zhang, currently a research scientist at KLA, holds a Ph.D. degree in Mechanical Engineering from the University at Buffalo, SUNY, which he received in 2021. He also obtained an M.S. degree in Mechanical Engineering from the same institution in 2017. Before moving to the USA, he worked as a mechanical engineer in China. Zhang's research interests cover a broad range of areas, including machine learning, hybrid modeling, intelligent manufacturing, and image processing. Presently, he is focused on investigating the intersection of physics and machine learning.Chandan Kumar SahuChandan Kumar Sahu is currently pursuing his Ph.D. degree in automotive engineering at Clemson University. He received his bachelors from National Institute of Technology Rourkela, India, in 2014, and his masters from the State University of New York at Buffalo, USA, in 2020, both in mechanical engineering. He led the homologation of the development of new models at Honda Motorcycle and Scooter India from 2014 to 2017. He is interested in machine learning in general, with a focus on its geometric attributes and mechanical applications. His research investigated additive manufacturing, augmented reality, cyber-physical systems, and engineering design applications using computational tools like machine learning, graph theory, and natural language processing. His current research aspires to develop mathematical models of requirements to capture their interrelationships to investigate the role of requirements during the evolution of a system.Shubhendu Kumar SinghShubhendu Kumar Singh received a B.E. degree, in Mechanical Engineering, from the Birla Institute of Technology, Mesra, Ranchi, India, in 2014 and an M.S. degree in Mechanical Engineering from the State University of New York at Buffalo, USA, in 2019. He is pursuing his Doctoral Degree in Automotive Engineering from Clemson University, USA. He has worked as an Engineer at Tractors and Farm Equipment Limited (TAFE Ltd.), India. Part of his job responsibilities at TAFE Ltd. involved solving technical and operational issues related to the tractors. Specifically, the problems related to the Powertrain (Engine, Turbocharger) and Transmission system. He has also worked as a Machine Learning intern at the Los Alamos National Lab, USA. He is a Research Assistant at Geometric Reasoning and Artificial Intelligence Laboratory (GRAIL) at Clemson University. In recent years, he has worked on various projects at the intersection of Artificial Intelligence and Engineering. A few of these projects have been supported by DARPA, ONR, and CESMII. His current research endeavours are focused on Artificial Intelligence, Computer Vision, Physics-Informed Machine Learning and their application in Design-Manufacturing, Fault Diagnostics and Prognostics, and Cyber-Physical Systems.Rahul RaiRahul Rai joined the Department of Automotive Engineering in 2020 as Dean's Distinguished Professor in the Clemson University International Center for Automotive Research (CU-ICAR). He also holds appointments in the Computer Science and Mechanical Engineering departments at Clemson University. He is Associate Director of Artificial Intelligence Research Institute in Science and Engineering (AIRISE) at Clemson University. He directs the Geometric Reasoning and Artificial Intelligence Lab (GRAIL, which is located at both CU-ICAR and Center for Manufacturing Innovation (CMI). Previously, he served on the Mechanical and Aerospace Engineering faculty at the University at Buffalo-SUNY (2012–2020). Dr. Rai also has industrial research centre experiences at United Technology Research Center (UTRC) and Palo Alto Research Center (PARC). Dr. Rai received his B.Tech. degree in 2000 and M.S. degree in 2002 in Manufacturing Engineering from the National Institute of Foundry and Forge Technology (NIFFT), Ranchi, India, and Missouri University of Science and Technology (Missouri S&T) USA, respectively. He earned his doctoral degree in Mechanical Engineering from The University of Texas at Austin USA in 2006. Dr. Rai's research is focused on developing computational tools for Manufacturing, Cyber-Physical System (CPS) Design, Autonomy, Collaborative Human-Technology Systems, Diagnostics and Prognostics, and Extended Reality (XR) domains. By combining engineering innovations with methods from machine learning, AI, statistics and optimisation, and geometric reasoning, his research strives to solve important problems in the above-mentioned domains. His research has been supported by NSF, DARPA, ONR, ARL, NSWC, DMDII, CESMII, HP, NYSERDA, and NYSPII (funding totalling more than $20M as PI/Co-PI). He has authored over 100 papers to date in peer-reviewed conferences and journals covering a wide array of problems. Dr. Rai is the recipient of numerous awards, including the 2009 HP Design Innovation, 2017 ASME IDETC/CIE Young Engineer Award, and 2019 PHM society conference best paper award. Additionally, Dr. Rai is Associate Editor of the International Journal of Production Research and ASME Journal of Computing and Information Science in Engineering (JCISE) journals and has taken significant leadership roles within the ASME Computers and Information in Engineering professional society.Zhuo YangZhuo Yang is a research associate at National Institute of Standards and Technology (NIST). He received a Ph.D. degree in Mechanical Engineering from University of Massachusetts Amherst in 2018. His bachelor degree was in Mechanical Engineering from Beijing University of Technology in 2009. He was a mechanical engineer in Beijing Wandong Medical from 2009–2012, mainly focusing on developing radiology equipment such as CT and MRI. His current research interests include Additive Manufacturing, design optimisation, machine learning, and real-time control. His current project is about data fusion and registration in powder bed fusion additive manufacturing.Yan LuYan Lu Dr. is a member of the System Integration Division at the Engineering Lab. Her research interests at NIST include additive manufacturing data registration, data integration and fusion, and smart manufacturing system information modeling and integration architecture. Before joining NIST, Dr. Lu was the head of Grid Automation and Production Operation and Optimization Research Group at Siemens Corporation, Corporate Technology. With Siemens, she has led and successfully delivered tens of million dollars of corporate funded and government funded research projects in the areas of survivable control systems, energy automation and building energy management systems. She has published more than 100 peer reviewed journal and conference papers and was granted more than 15 patents in industry and building automation technology. Dr. Lu also worked for Seagate Research Center for two years on developing hard disk drive servo control.