BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Data\, Data Everywhere… How to Shelter from the Digital Tsunami 
 ?
DTSTART:20110701T101500
DTSTAMP:20260511T062200Z
UID:80b40cf140b3118accd2db4ea4c7dccb5df413ffc6e51249d0a6462d
CATEGORIES:Conferences - Seminars
DESCRIPTION:Pr. S. Shah\, Senior Industrial Research in Computer Process C
 ontrol\, University of Alberta\, Canada.\nIt is now common to have archiva
 l history of thousands of sensors sampled every second over long time peri
 ods. Yet we frequently have process engineers complain: “…We are drown
 ing in data but starving for information…”. How can these rich data se
 ts be put to use? \nThis seminar will address the issue of information and
  knowledge extraction from data with emphasis on process and performance m
 onitoring. Most of the major plant\, factory\, process\, equipment and too
 l disruptions are avoidable\, and yet preventable fault detection and diag
 nosis strategies are not the norm in most industries. It is not uncommon t
 o see simple and preventable faults disrupt the operation of an entire int
 egrated manufacturing facility. For example\, faults such as malfunctionin
 g sensors or actuators\, inoperative alarm systems\, poor controller tunin
 g or configuration can render the most sophisticated control systems usele
 ss. Such disruptions can cost in the excess of $1 million per day and on t
 he average they rob the plant of 7% of its annual capacity. Over the last 
 decade the fields of multivariate statistics\, controller performance moni
 toring techniques and Bayesian inference methods have merged to develop po
 werful sensing and condition-based monitoring systems for predictive fault
  detection and diagnosis.  These methods rely on the notion of sensor fusi
 on whereby data from many sensors or units are combined with process infor
 mation\, such as physical connectivity of process units\, to give a holist
 ic picture of health of an integrated plant. Such methods are at a stage w
 here these strategies are being implemented for off-line and on-line deplo
 yment. This presentation will outline the field of sensor fusion - the app
 lication of signal processing methods\, in the temporal as well as spectra
 l domains\, on a multitude and NOT singular sensor signals to detect incip
 ient process abnormality before a catastrophic breakdown is likely to occu
 r. This talk will be complemented with industrial case studies to demonstr
 ate the success of these methods. These same techniques can also be applie
 d in other fields. For example\, the fusion of pixels of information from 
 digital images will be illustrated via application of automated detection 
 and diagnosis of Malaria parasites from microscopic images.
LOCATION:MEC2405
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR