BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//talks.cam.ac.uk//v3//EN
BEGIN:VTIMEZONE
TZID:Europe/London
BEGIN:DAYLIGHT
TZOFFSETFROM:+0000
TZOFFSETTO:+0100
TZNAME:BST
DTSTART:19700329T010000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0100
TZOFFSETTO:+0000
TZNAME:GMT
DTSTART:19701025T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
CATEGORIES:Isaac Newton Institute Seminar Series
SUMMARY:Applications of convolutional neural networks to t
urbulence - Koji Fukagata (Keio University)
DTSTART;TZID=Europe/London:20220331T120000
DTEND;TZID=Europe/London:20220331T123000
UID:TALK171206AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/171206
DESCRIPTION:Application of machine learning is currently one o
f the hottest topics in the fluid mechanics field.
While machine learning seems to have a great poss
ibility\, its limitations should also be clarified
. In our group\, we have started a research projec
t to construct a nonlinear feature extraction meth
od by applying machine learning technology to &ldq
uo\;turbulence big data\,&rdquo\; extracting the n
onlinear modes essential to the regeneration mecha
nism of turbulence\, and deriving the time evoluti
on equation of those nonlinear modes. In this pres
entation\, we will introduce some examples on lear
ning and regeneration of temporal evolution of cro
ss-sectional velocity field in a turbulent channel
flow using convolutional neural network (CNN). We
will also introduce the application of CNN for su
per-resolution analysis and reduced order modeling
of turbulence. We also introduce our attempts to
interpret the nonlinear modes extracted by CNN aut
oencoder and to use them for an advanced design of
flow control\, as well as an attempt for uncertai
nty quantification and applications to experimenta
l data.
LOCATION:Seminar Room 1\, Newton Institute
CONTACT:
END:VEVENT
END:VCALENDAR