TY - GEN
T1 - Feedback control of scanning laser epitaxy (SLE) - Estimation of melt pool temperature and size
AU - Bansal, Rohan
AU - Basak, Amrita
AU - Das, Suman
PY - 2019/1/1
Y1 - 2019/1/1
N2 - The properties of additively manufactured parts significantly depend on the instantaneous temperature distribution and cooling rates in the melt pool and surrounding regions, respectively. The open-loop control of additive manufacturing (AM) cannot compensate for the stochastic variations in the melt pool temperature and size during the process. Closed-loop control can play a significant role in maintaining a desired melt pool temperature by controlling the laser power in situ. However, in order to effectively implement closed-loop control, two critical parameters need to be calculated such as the melt pool temperature and melt pool size. To calculate the melt pool temperature, in this work, a thermal imaging camera calibrated via the emissivity value of the material and validated through a Pyrometer was used. To calculate the melt pool size, two machine vision algorithms are implemented. It was found that the connected component labeling algorithm performed better than the Canny edge detection algorithm.
AB - The properties of additively manufactured parts significantly depend on the instantaneous temperature distribution and cooling rates in the melt pool and surrounding regions, respectively. The open-loop control of additive manufacturing (AM) cannot compensate for the stochastic variations in the melt pool temperature and size during the process. Closed-loop control can play a significant role in maintaining a desired melt pool temperature by controlling the laser power in situ. However, in order to effectively implement closed-loop control, two critical parameters need to be calculated such as the melt pool temperature and melt pool size. To calculate the melt pool temperature, in this work, a thermal imaging camera calibrated via the emissivity value of the material and validated through a Pyrometer was used. To calculate the melt pool size, two machine vision algorithms are implemented. It was found that the connected component labeling algorithm performed better than the Canny edge detection algorithm.
UR - http://www.scopus.com/inward/record.url?scp=85075330089&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85075330089&partnerID=8YFLogxK
U2 - 10.7449/2019/MST_2019_52_59
DO - 10.7449/2019/MST_2019_52_59
M3 - Conference contribution
T3 - MS and T 2019 - Materials Science and Technology 2019
SP - 52
EP - 59
BT - MS and T 2019 - Materials Science and Technology 2019
PB - Materials Science and Technology
T2 - Materials Science and Technology 2019, MS and T 2019
Y2 - 29 September 2019 through 3 October 2019
ER -