A Digital Twin Approach to Study Additive Manufacturing Processing Using Embedded Optical Fiber Sensors and Numerical Modeling

• Published in: Journal of lightwave technology Vol. 38; no. 22; pp. 6402 - 6411
• Main Authors: Zou, Ran, Liang, Xuan, Chen, Qian, Wang, Mohan, Zaghloul, Mohamed A. S., Lan, Hui, Buric, Michael P., Ohodnicki, Paul R., Chorpening, Benjamin, To, Albert C., Chen, Kevin P.
• Format: Journal Article
• Language: English
• Published: New York IEEE 15-11-2020; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Additive manufacturing; Design optimization; digital twin; Digital twins; distributed sensing; Electroplating; Fiber optics; Finite element method; Lenses; MATERIALS SCIENCE; Mathematical models; Metals; Nickel base alloys; Numerical models; Optical fiber sensors; Optical fibers; Optical frequency; Reflectometers; Sensors; Spatial resolution; Strain; Strain analysis; strain measurement; Substrates; Superalloys; Temperature measurement; Temperature sensors
• ISSN: 0733-8724; 1558-2213
• DOI: 10.1109/JLT.2020.3010722

One of the major challenges for metal-powder-based additive manufacturing is measuring and mitigating residual strain induced during the manufacturing processes. This article reports distributed fiber optic sensors embedded in Inconel alloy components as experimental means to validate numerical models of additive manufacturing process. Electroplating was used to deposit a metal protective jacket onto standard telecom single-mode fibers for strain measurements, Fiber sensors were embedded in an Inconel alloy substrate using the laser engineered net shaping (LENS) process. Using a Rayleigh-scattering optical frequency domain reflectometer (OFDR), temperature changes, and residual strain in the metal substrate were monitored with 5 mm spatial resolution during the LENS process. Using finite element analysis, temperature and strain profiles induced by the LENS deposition processes were also numerically studied. Discrepancies between the simulated temperature and strain profiles and those measured directly were less than 10%. Results presented in this article demonstrates a digital twin approach to fuse modeling results with distributed fiber sensor measurement data to study additive manufacturing process toward design and fabrication process optimization.