Detection of unusual events and trends in complex non-stationary data streams

R. B. Perez, V. A. Protopopescu, B. A. Worley, C. L. Perez

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

The search for unusual events and trends hidden in multi-component, nonlinear, non-stationary, noisy signals is extremely important for a host of different applications, ranging from nuclear power plant and electric grid operation to internet traffic and implementation of non-proliferation protocols. In the context of this work, we define an unusual event as a local signal disturbance and a trend as a continuous carrier of information added to and different from the underlying baseline dynamics. The goal of this paper is to investigate the feasibility of detecting hidden, intermittent events inside non-stationary signal data sets corrupted by high levels of noise by using the Hilbert-Huang empirical mode decomposition method.

Original languageEnglish
Title of host publication5th International Topical Meeting on Nuclear Plant Instrumentation Controls, and Human Machine Interface Technology (NPIC and HMIT 2006)
Pages486-494
Number of pages9
StatePublished - 2006
Event5th International Topical Meeting on Nuclear Plant Instrumentation Controls, and Human Machine Interface Technology (NPIC and HMIT 2006) - Albuquerque, NM, United States
Duration: Nov 12 2006Nov 16 2006

Publication series

Name5th International Topical Meeting on Nuclear Plant Instrumentation Controls, and Human Machine Interface Technology (NPIC and HMIT 2006)
Volume2006

Conference

Conference5th International Topical Meeting on Nuclear Plant Instrumentation Controls, and Human Machine Interface Technology (NPIC and HMIT 2006)
Country/TerritoryUnited States
CityAlbuquerque, NM
Period11/12/0611/16/06

Funding

We thank Drs. J. Antonino-Daviu and J. Roger-Folch, Electrical Engineering Department, Universitat Politecnica de Valencia, Spain, for providing the induction machine data. C.C.-P. gratefully acknowledges Dr. Steven R. Long, NASA/GSFC/Wallops Flight Facility for providing the Hilbert Transform Algorithm. The Oak Ridge National Laboratory is managed by UT-Battelle, LLC, under Contract DE-AC05-00OR22725 with the US Department of Energy. The United States Government retains, and the publisher by accepting the article for publication, acknowledges that the United States Government retains, a non-exclusive, paid-up, irrevocable, world-wide license to publish or reproduce the published form of this manuscript, or allow others to do so, for United States Government purposes.

Fingerprint

Dive into the research topics of 'Detection of unusual events and trends in complex non-stationary data streams'. Together they form a unique fingerprint.

Cite this