Dynamic detection of anomalous regions within distributed acoustic sensing data streams using locally stationary wavelet time series

Rebecca Wilson, Idris A. Eckley, Matthew Nunes, Timothy Park

Research output: Contribution to journalArticle

Abstract

Distributed acoustic sensing technology is increasingly being used to support production and well management within the oil and gas sector, for example to improve flow monitoring and production profiling. This sensing technology is capable of recording substantial data volumes at multiple depths within an oil well, giving unprecedented insights into production behaviour. However the technology is also prone to recording periods of anomalous behaviour, where the same physical features are concurrently observed at multiple depths. Such features are called `stripes' and are undesirable, detrimentally affecting well performance modelling. This paper focuses on the important challenge of developing a principled approach to identifying such anomalous periods within distributed acoustic signals. We extend recent work on classifying locally stationary wavelet time series to an online setting and, in so doing, introduce a computationally-efficient online procedure capable of accurately identifying anomalous regions within multivariate time series.
Original languageEnglish
Pages (from-to)748-772
Number of pages25
JournalData Mining and Knowledge Discovery
Volume33
Issue number3
Early online date20 Feb 2019
DOIs
Publication statusPublished - 15 May 2019

Fingerprint

Time series
Acoustics
Data recording
Oil wells
Monitoring
Gases
Oils

Keywords

  • Coherence
  • Distributed acoustic sensing
  • Dynamic classification
  • Locally stationary time series
  • Stripe detection
  • Wavelets

ASJC Scopus subject areas

  • Information Systems
  • Computer Science Applications
  • Computer Networks and Communications

Cite this

Dynamic detection of anomalous regions within distributed acoustic sensing data streams using locally stationary wavelet time series. / Wilson, Rebecca; Eckley, Idris A.; Nunes, Matthew; Park, Timothy.

In: Data Mining and Knowledge Discovery, Vol. 33, No. 3, 15.05.2019, p. 748-772.

Research output: Contribution to journalArticle

@article{e44bf457f1944027ac94f4682fc19cf5,
title = "Dynamic detection of anomalous regions within distributed acoustic sensing data streams using locally stationary wavelet time series",
abstract = "Distributed acoustic sensing technology is increasingly being used to support production and well management within the oil and gas sector, for example to improve flow monitoring and production profiling. This sensing technology is capable of recording substantial data volumes at multiple depths within an oil well, giving unprecedented insights into production behaviour. However the technology is also prone to recording periods of anomalous behaviour, where the same physical features are concurrently observed at multiple depths. Such features are called `stripes' and are undesirable, detrimentally affecting well performance modelling. This paper focuses on the important challenge of developing a principled approach to identifying such anomalous periods within distributed acoustic signals. We extend recent work on classifying locally stationary wavelet time series to an online setting and, in so doing, introduce a computationally-efficient online procedure capable of accurately identifying anomalous regions within multivariate time series.",
keywords = "Coherence, Distributed acoustic sensing, Dynamic classification, Locally stationary time series, Stripe detection, Wavelets",
author = "Rebecca Wilson and Eckley, {Idris A.} and Matthew Nunes and Timothy Park",
year = "2019",
month = "5",
day = "15",
doi = "10.1007/s10618-018-00608-w",
language = "English",
volume = "33",
pages = "748--772",
journal = "Data Mining and Knowledge Discovery",
issn = "1384-5810",
publisher = "Springer Verlag",
number = "3",

}

TY - JOUR

T1 - Dynamic detection of anomalous regions within distributed acoustic sensing data streams using locally stationary wavelet time series

AU - Wilson, Rebecca

AU - Eckley, Idris A.

AU - Nunes, Matthew

AU - Park, Timothy

PY - 2019/5/15

Y1 - 2019/5/15

N2 - Distributed acoustic sensing technology is increasingly being used to support production and well management within the oil and gas sector, for example to improve flow monitoring and production profiling. This sensing technology is capable of recording substantial data volumes at multiple depths within an oil well, giving unprecedented insights into production behaviour. However the technology is also prone to recording periods of anomalous behaviour, where the same physical features are concurrently observed at multiple depths. Such features are called `stripes' and are undesirable, detrimentally affecting well performance modelling. This paper focuses on the important challenge of developing a principled approach to identifying such anomalous periods within distributed acoustic signals. We extend recent work on classifying locally stationary wavelet time series to an online setting and, in so doing, introduce a computationally-efficient online procedure capable of accurately identifying anomalous regions within multivariate time series.

AB - Distributed acoustic sensing technology is increasingly being used to support production and well management within the oil and gas sector, for example to improve flow monitoring and production profiling. This sensing technology is capable of recording substantial data volumes at multiple depths within an oil well, giving unprecedented insights into production behaviour. However the technology is also prone to recording periods of anomalous behaviour, where the same physical features are concurrently observed at multiple depths. Such features are called `stripes' and are undesirable, detrimentally affecting well performance modelling. This paper focuses on the important challenge of developing a principled approach to identifying such anomalous periods within distributed acoustic signals. We extend recent work on classifying locally stationary wavelet time series to an online setting and, in so doing, introduce a computationally-efficient online procedure capable of accurately identifying anomalous regions within multivariate time series.

KW - Coherence

KW - Distributed acoustic sensing

KW - Dynamic classification

KW - Locally stationary time series

KW - Stripe detection

KW - Wavelets

UR - http://www.scopus.com/inward/record.url?scp=85061910936&partnerID=8YFLogxK

U2 - 10.1007/s10618-018-00608-w

DO - 10.1007/s10618-018-00608-w

M3 - Article

VL - 33

SP - 748

EP - 772

JO - Data Mining and Knowledge Discovery

JF - Data Mining and Knowledge Discovery

SN - 1384-5810

IS - 3

ER -