CLASSIFICATION OF SENSOR MEASUREMENTS FROM NON-NEWTOWNIAN FLUIDS USING BATCH AND ONLINE ANALYSIS OF DATA

M. Ziyan Sheriff, Mohamed N. Nounou, Mohammad A. Rahman, Ibrahim Hassan, Hazem N. Nounou

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

Abstract

With recent technological advancements most industrial processes collect an abundance of process data from sensors. Sensible utilization and interpretation of collected data is increasingly important. A primary objective of this work is to evaluate the potential of using only differential pressure measurements to classify operating flow rates of non-Newtonian fluids for single phase and multiphase flow conditions using both batch and online methods of data analysis. This was accomplished by using statistical tests and charts to monitor changes in the process mean and variance. For batch analysis, the two-sample t-test and the F-test for equality of variance provided classification accuracies of 100%, despite the presence of measurement noise, while advanced statistical hypothesis testing techniques, namely the generalized and exponential likelihood ratio charts, performed well for online analysis.

Original languageEnglish
Title of host publicationOffshore Geotechnics; Petroleum Technology
PublisherAmerican Society of Mechanical Engineers (ASME)
ISBN (Electronic)9780791886915
DOIs
Publication statusPublished - 2023
Externally publishedYes
EventASME 2023 42nd International Conference on Ocean, Offshore and Arctic Engineering, OMAE 2023 - Melbourne, Australia
Duration: 11 Jun 202316 Jun 2023

Publication series

NameProceedings of the International Conference on Offshore Mechanics and Arctic Engineering - OMAE
Volume9

Conference

ConferenceASME 2023 42nd International Conference on Ocean, Offshore and Arctic Engineering, OMAE 2023
Country/TerritoryAustralia
CityMelbourne
Period11/06/2316/06/23

Keywords

  • Data classification
  • flow analysis
  • generalized likelihood ratio (GLR)
  • statistical hypothesis testing

Fingerprint

Dive into the research topics of 'CLASSIFICATION OF SENSOR MEASUREMENTS FROM NON-NEWTOWNIAN FLUIDS USING BATCH AND ONLINE ANALYSIS OF DATA'. Together they form a unique fingerprint.

Cite this