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:Seehorn
DTSTART;TZID=Europe/Stockholm:20230626T120000
DTEND;TZID=Europe/Stockholm:20230626T123000
UID:submissions.pasc-conference.org_PASC23_sess179_pap113@linklings.com
SUMMARY:StyleGAN as a Deconvolutional Operator for Large Eddy Simulations
DESCRIPTION:Paper\n\nJony Castagna and Francesca Schiavello (Science and T
 echnology Facilities Council)\n\nWe present a novel deconvolution operator
  for Large Eddy Simulation (LES) of turbulent flows based on the latest St
 yleGAN deep learning networks. We exploit the flexibility of this architec
 ture in separating the different layers of the GAN generator, which can be
  seen as instantaneous fields of the LES. These can be moved in time via i
 ntegrating the corresponding filtered Navier-Stokes (NS) equations. The su
 bgrid-scale (SGS) stress tensor is obtained from the reconstructed field, 
 rather than ad-hoc turbulence models. We trained a StyleGAN-based network 
 (MSG-StyleGAN) with 5000 images of a decaying 2D-Homogeneous Isotropic Tur
 bulence (2D-HIT) starting at <em>Re Pi</em> = 60 using a 256x256 grid mesh
  size. We then reconstructed a DNS simulation, point by point, using a 32x
 32 resolution via research into the latent space of the GAN until the diff
 erence between internal fields and LES fields are within a given tolerance
 . Results show convergence towards the ground truth DNS solution as the to
 lerance approaches zero.\n\nDomain: Engineering\n\nSession Chair: Charles 
 Moulinec (Science and Technology Facilities Council)
END:VEVENT
END:VCALENDAR