BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Strength prediction of polymer composite laminates under uncertain
 ties using theory-guided machine learning - Prof Pedro Camanho\, Mechanica
 l Engineering Department\, University of Porto
DTSTART:20230224T140000Z
DTEND:20230224T150000Z
UID:TALK193579@talks.cam.ac.uk
CONTACT:Hilde Hambro
DESCRIPTION:This work represents a first study towards the application of 
 theory-guided machine learning techniques in the prediction of design allo
 wables of notched polymer composite laminates accounting for material and 
 geometric uncertainties. Building on data generated analytically\, using e
 ither phase-field methods or finite fracture mechanics\, and reduced repre
 sentations of composite lay-ups\, four machine learning algorithms are use
 d to predict the strength of composite laminates with notches of several g
 eometries and the corresponding statistical distribution\, associated to m
 aterial and geometrical variability.\n \nExcellent representations of the 
 design space (relative errors of around ±10%) and very accurate represent
 ations of the distributions of notched strengths and of the corresponding 
 B-basis allowables used in aircraft structural design are obtained. Gaussi
 an-based models proved to be the most reliable approach as a result of its
  continuous nature\, accuracy\, and fast training process. This work serve
 s as basis for the prediction of first-ply failure\, ultimate strength and
  failure mode of composite laminates based on non-linear finite element si
 mulations across different length scales\, providing significant reduction
 s of the computational time required to virtually certify composite aircra
 ft structures.
LOCATION:Oatley Seminar Room\, Department of Engineering
END:VEVENT
END:VCALENDAR