BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Inverse Problems in the Prediction of Reservoir Petroleum Properti
 es using Multiple Kernel  Learning - *Backhouse\, L\, Demyanov\, V\, Chris
 tie\, M (Heriot Watt University)
DTSTART:20111216T100000Z
DTEND:20111216T103000Z
UID:TALK35000@talks.cam.ac.uk
CONTACT:Mustapha Amrani
DESCRIPTION:In Reservoir engineering a common inverse problem is that of e
 stimating the reservoir properties such as Porosity and Permeability by ma
 tching the simulation model to the dynamic Production data. Using this mod
 el\, future predictions can then be made and the uncertainty of these pred
 ictions quantified using Bayes Rules. \n\nMultiple Kernel Learning (MKL) i
 s an inverse problem that maps input data into a feature space with the us
 e of kernel functions. MKL is a predictive tool that has been applied in t
 he Petroleum Industry to estimate the spatial distribution of Porosity and
  Permeability. The parameters of the kernels and the choice of the kernels
  are determined by matching to hard data for Porosity and Permeability fou
 nd at the wells thus producing a static model that is used as input into t
 he dynamic model. \n\nIn this paper we show how we combine the above menti
 oned inverse problems. We estimate the Porosity and Permeability into a st
 atic model then match to the dynamic production data to tune the parameter
 s in the Multiple Kernel Learning Framework. Specifically we integrate the
  MLE estimation from the MKL objective Function into the History Matching 
 Function. \n\n\n
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR