BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Supervised learning for soil identification and deformation predic
 tion in excavation based on Bayesian inference. - Yingyan Yin\, CUED
DTSTART:20171116T160000Z
DTEND:20171116T164500Z
UID:TALK81921@talks.cam.ac.uk
CONTACT:Magdalena Charytoniuk
DESCRIPTION:In the past few decades\, the demand of construction in underg
 round spaces has increased dramatically in urban areas with high populatio
 n densities. However\, the impact of the construction of underground struc
 tures in surrounding infrastructures raises a lot of concerns since the mo
 vements caused by deep excavations might damage adjacent structures. Unfor
 tunately\, the prediction of the geotechnical behaviour is difficult due t
 o uncertainties and lack of information of the underground environment.The
 refore\, to ensure safety\, engineers tend to choose unfavourable conditio
 ns for the design of excavation supporting systems\, which usually leads t
 o a conservative design that requires unnecessary material and constructio
 n time.\n\nAdaptive design provides a way to avoid such redundancy by usin
 g the most probable conditions and incorporating knowledge learned during 
 the construction progress. The monitoring data obtained during the constru
 ction is used to update the model and therefore the accuracy of the predic
 tion of the ground response will be improved. This process is recognised a
 s the back analysis\,a core procedure in the Observational Method.\n\nThis
  process can be realised by using supervised learning. In this research\, 
 a probabilistic model coupled with Bayesian inference is developed which i
 s not only able to learn the relations between the input soil parameters a
 nd the response\, but also identify the underlying uncertainties from all 
 sources. Moreover\, it integrates subjective information and objective eng
 ineering experience information in a rational and quantitative way. Furthe
 rmore\, under this probabilistic setting\, the uncertainty information is 
 also contained in the prediction\, which is crucial to the confidence base
 d decision making.
LOCATION: Cambridge University Engineering Department\, Lecture Room 6
END:VEVENT
END:VCALENDAR