BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Statistical Change Detection for Prognosis and Diagnosis - Profess
 or Mogens Blanke\, Technical University of Denmark
DTSTART:20141104T140000Z
DTEND:20141104T150000Z
UID:TALK55945@talks.cam.ac.uk
CONTACT:Tim Hughes
DESCRIPTION:Statistical change detection plays a key role in prognosis and
  diagnosis of faults. Design of change detection algorithms is well known 
 in theory\, but practice often violates the idealised perquisites that the
 ory requires. This talk provides tutorial insight in existing methods for 
 change detection in the presence of Gaussian or Non-Gaussian noise with fo
 cus on reliable diagnosis / prognosis where accurate knowledge of detectio
 n and false alarm probabilities is required. As these features are determi
 ned by properties of the test statistic of the particular problem\, the ta
 lk discusses the consequences of correlation in real life\, and compares t
 heoretical results with those obtained in actual applications. A methodolo
 gy for change detection is suggested that adapts to real-life conditions. 
 Using estimation of distribution parameters for the actual test statistic\
 , threshold for hypothesis testing and test sequence length are shown to b
 e parameters in an optimization problem that can guarantee prescribed prop
 erties. Detection with multiple detectors\, based on different indicators 
 of change\, is then discussed\, and it is shown how a Copula description o
 f joint probability can be employed to assess properties of combined detec
 tors. Selected industrial applications from unmanned aircraft and drilling
  illustrate the methods.
LOCATION: Cambridge University Engineering Department\, LR12
END:VEVENT
END:VCALENDAR