BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Europe/Stockholm
X-LIC-LOCATION:Europe/Stockholm
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20230831T095745Z
LOCATION:Dischma
DTSTART;TZID=Europe/Stockholm:20230626T170000
DTEND;TZID=Europe/Stockholm:20230626T173000
UID:submissions.pasc-conference.org_PASC23_sess163_msa237@linklings.com
SUMMARY:Physics-Informed Machine Learning for Reduced Lagrangian Modeling 
 of Turbulence: Lagrangian LES
DESCRIPTION:Minisymposium\n\nMichael Woodward (Los Alamos National Laborat
 ory, The University of Arizona); Yifeng Tian and Chris Fryer (Los Alamos N
 ational Laboratory); Misha Stepanov (The University of Arizona); Daniel Li
 vescu (Los Alamos National Laboratory); and Misha Chertkov (The University
  of Arizona)\n\nObtaining accurate numerical solutions of turbulent flows 
 with Direct Numerical Simulation (DNS) is intractable for most practical a
 pplications. Thus, building efficient, accurate, and generalizable reduced
 -order models for turbulent flows remains of great interest, however, this
  demands new and creative methods. We approach these challenging problems 
 by incorporating modern advances in Machine Learning, together with Lagran
 gian-based frameworks, such as Smoothed Particle Hydrodynamics (SPH), to m
 odel homogeneous isotropic turbulence at coarse-grained scales. In this ta
 lk, we demonstrate two approaches for learning reduced Lagrangian models: 
 (1) using parameter estimation techniques for exploring a hierarchy of phy
 sics-based and Neural Network (NN) based parameterizations of SPH, and (2)
  generalizing the evolutionary equations of Lagrangian particles with exte
 nded, physics-informed NN structure and functional freedom to construct a 
 Lagrangian Large Eddy Simulation (L-LES) framework. The models are trained
  on Lagrangian particle data obtained from highly resolved weakly compress
 ible DNS data with periodic boundary conditions and a turbulent Mach numbe
 r of 0.08. In each case, varying degrees of physical symmetries are enforc
 ed in the model which we show improves the generalizability to different t
 ime scales (in-distribution test set) and turbulent Mach numbers (0.04, 0.
 16) (out-of-distribution test sets).\n\nDomain: Computer Science, Machine 
 Learning, and Applied Mathematics 
\n\nSession Chairs: Riccardo Balin (Arg
 onne National Laboratory) and Ramesh Balakrishnan (Argonne National Labora
 tory)
END:VEVENT
END:VCALENDAR