A near infrared-based downwhole water-cut meter using neural network

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

Abstract

In this paper, near infrared-based technique of oil-water mixture for water-cut measurement using neural network technique is presented. It uses a multivariate (MDA) algorithm which comprises the Partial least Square regression (PLS), Polynomial PLS, and an Artificial Neural Network (ANN) for spectrum analysis. The NIR spectra is postprocessed using the principal component analysis (PCA).Experimental results indicate that an accurate water-cut measurement can be achieved with less than 0.5% error in the range of [90 to 100%] water-cut. This interesting result, in addition to the fact that the NIR array device is non-invasive, non-intrusive and can be easily inserted into deep oil wells using optical fiber would lead to concluded that near-infrared spectroscopy can be a good candidate for downhole accurate water-cut measurement.

Original languageBritish English
Title of host publicationInfrared Remote Sensing and Instrumentation XXII
EditorsGonzalo Paez, Marija Strojnik Scholl
PublisherSPIE
ISBN (Electronic)9781628412468
DOIs
StatePublished - 2014
EventInfrared Remote Sensing and Instrumentation XXII - San Diego, United States
Duration: 18 Aug 2014 → …

Publication series

NameProceedings of SPIE - The International Society for Optical Engineering
Volume9219
ISSN (Print)0277-786X
ISSN (Electronic)1996-756X

Conference

ConferenceInfrared Remote Sensing and Instrumentation XXII
Country/TerritoryUnited States
CitySan Diego
Period18/08/14 → …

Keywords

  • Downhole measurement
  • Near infrared instrument
  • Neural network

Fingerprint

Dive into the research topics of 'A near infrared-based downwhole water-cut meter using neural network'. Together they form a unique fingerprint.

Cite this