BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:The Statistical Finite Element Method - Mark Girolami (University 
 of Cambridge)
DTSTART:20210602T100000Z
DTEND:20210602T113000Z
UID:TALK158842@talks.cam.ac.uk
CONTACT:Elre Oldewage
DESCRIPTION:The finite element method (FEM) is one of the great triumphs o
 f modern day applied mathematics\, numerical analysis and software and com
 puting technology. Every area of the sciences and engineering has been pos
 itively impacted by the ability to model and study complex physical and na
 tural systems described by systems of partial differential equations (PDE)
  via the FEM .\n\nIn parallel the recent developments in sensor\, measurem
 ent\, and signalling technologies enables the phenomenological study of sy
 stems as diverse as protein signalling in the cell\, to turbulent combusti
 on in jet engines\, to plastic deformation in bridges.\n\nThe connection b
 etween sensor data and FEM is currently restricted to data assimilation fo
 r solving inverse problems or the calibration of PDE based models. This ho
 wever places unwarranted faith in the fidelity of the underlying mathemati
 cal description of the actual system under study. If one concedes that the
 re is ‘missing physics’ or mis-specification between generative realit
 y and the mathematical abstraction defining the FEM then a framework to sy
 stematically characterise and propagate this uncertainty in FEM is require
 d.\n\nThis talk will present a formal statistical construction of the FEM 
 which systematically blends both mathematical description with observation
 al data and provides both small and large scale examples from 3D printed s
 tructures to working rail bridges currently operated by the United Kingdom
  Network Rail.
LOCATION:https://eng-cam.zoom.us/j/82019956685?pwd=WUNSVVcrdC9IZGxQOHFhSTh
 jUjd2dz09
END:VEVENT
END:VCALENDAR