TY - GEN
T1 - Data-driven modeling of large-eddy simulations for fuel injector design
AU - Milan, P. J.
AU - Mondal, S.
AU - Torelli, R.
AU - Lusch, B.
AU - Maulik, R.
AU - Magnotti, G. M.
N1 - Funding Information:
The submitted manuscript has been created by UChicago Argonne, LLC, Operator of Argonne National Laboratory (Argonne). Argonne, a U.S. Department of Energy Office (DOE) of Science laboratory, is operated under Contract No. DE-AC02-06CH11357. The U.S. Government retains for itself, and others acting on its behalf, a paid-up nonexclusive, irrevocable worldwide license in said article to reproduce, prepare derivative works, distribute copies to the public, and perform publicly and display publicly, by or on behalf of the Government. This research has been supported by the U.S. DOE under contract LDRD Prime 2020-0094. We gratefully acknowledge the computing resources provided on Bebop, a high-performance computing cluster operated by the Laboratory Computing Resource Center at Argonne National Laboratory. This research also used resources of the Argonne Leadership Computing Facility, which is a DOE Office of Science User Facility supported under Contract DE-AC02-06CH11357. We also would like to thank Convergent Science for providing CONVERGE licenses and technical support for this work.
Publisher Copyright:
© 2021, American Institute of Aeronautics and Astronautics Inc, AIAA. All rights reserved.
PY - 2021
Y1 - 2021
N2 - The objective of this work is to develop an efficient emulator framework using modern machine learning techniques to predict spatiotemporally evolving flow dynamics for design exploration. This emulation framework encompasses autoencoders for nonlinear dimensionality reduction, deep neural networks for regression and predictive modeling, and a reconstruction algorithm for recovery of the flowfields. As a demonstration case, the turbulent multiphase flow development in a side-oriented single-hole diesel injector is considered for a wide range of design parameters, fuel properties, and operating conditions. The prediction accuracy of the flow features for new design settings is assessed, and the efficiency of the proposed emulator is evaluated.
AB - The objective of this work is to develop an efficient emulator framework using modern machine learning techniques to predict spatiotemporally evolving flow dynamics for design exploration. This emulation framework encompasses autoencoders for nonlinear dimensionality reduction, deep neural networks for regression and predictive modeling, and a reconstruction algorithm for recovery of the flowfields. As a demonstration case, the turbulent multiphase flow development in a side-oriented single-hole diesel injector is considered for a wide range of design parameters, fuel properties, and operating conditions. The prediction accuracy of the flow features for new design settings is assessed, and the efficiency of the proposed emulator is evaluated.
UR - http://www.scopus.com/inward/record.url?scp=85100310593&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85100310593&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:85100310593
SN - 9781624106095
T3 - AIAA Scitech 2021 Forum
SP - 1
EP - 15
BT - AIAA Scitech 2021 Forum
PB - American Institute of Aeronautics and Astronautics Inc, AIAA
T2 - AIAA Science and Technology Forum and Exposition, AIAA SciTech Forum 2021
Y2 - 11 January 2021 through 15 January 2021
ER -