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:Inverse Problems in the Prediction of Reservoir Pe
troleum Properties using Multiple Kernel Learning
- *Backhouse\, L\, Demyanov\, V\, Christie\, M (H
eriot Watt University)
DTSTART;TZID=Europe/London:20111216T100000
DTEND;TZID=Europe/London:20111216T103000
UID:TALK35000AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/35000
DESCRIPTION:In Reservoir engineering a common inverse problem
is that of estimating the reservoir properties suc
h as Porosity and Permeability by matching the sim
ulation model to the dynamic Production data. Usin
g this model\, future predictions can then be made
and the uncertainty of these predictions quantifi
ed using Bayes Rules. \n\nMultiple Kernel Learning
(MKL) is an inverse problem that maps input data
into a feature space with the use of kernel functi
ons. MKL is a predictive tool that has been applie
d in the Petroleum Industry to estimate the spatia
l distribution of Porosity and Permeability. The p
arameters of the kernels and the choice of the ker
nels are determined by matching to hard data for P
orosity and Permeability found at the wells thus p
roducing a static model that is used as input into
the dynamic model. \n\nIn this paper we show how
we combine the above mentioned inverse problems. W
e estimate the Porosity and Permeability into a st
atic model then match to the dynamic production da
ta to tune the parameters in the Multiple Kernel L
earning Framework. Specifically we integrate the M
LE estimation from the MKL objective Function into
the History Matching Function. \n\n\n
LOCATION:Seminar Room 1\, Newton Institute
CONTACT:Mustapha Amrani
END:VEVENT
END:VCALENDAR