Javascript must be enabled for the correct page display

IPTC-19865-MS : A Machine Learning-Based System for Self-Diagnosis Multiphase FlowMeters

Barbariol, Tommaso and Feltresi, Enrico and Fiorentini, Pietro and Susto, Gian Antonio (2020) IPTC-19865-MS : A Machine Learning-Based System for Self-Diagnosis Multiphase FlowMeters. International Petroleum Technology Conference. pp. 1-14.

[img] Text
OnePetro IPTC-19865-MS.pdf - Published Version
Restricted to RUG campus

Download (1MB)

Abstract

In the oil and gas industry, the need of Multiphase Flow Meters (MPFM) continues slowly to grow. Theindividual oil, gas and water flows rates measurements are important to monitor the reservoir, to improvethe well performances and to optimize the well production. Traditionally the composition of the flow hasbeen performed separating physically the phases with large, cumbersome, time consuming and expensivetest separator vessels. A MPFM is, on the other hand, a compact multi-sensors system that provides real-time and continuous measurements of the individual oil, gas and water flows rates of a well without theneed to separate the phases. Therefore, procedures for measure quality assessment are of crucial importance.In this work it is proposed an Anomaly Detection approach for MPFM based on unsupervised MachineLearning algorithms. It is effectively able to handle the complexity and variability associated with MPFMdata and it has the capability both to detect outliers and to pinpoint the faulty sensor. The proposed approachis designed for embedded implementation.

Item Type: Article
Additional Information: This paper was selected for presentation by an IPTC Programme Committee following review of information contained in an abstract submitted by the author(s).Contents of the paper, as presented, have not been reviewed by the International Petroleum Technology Conference and are subject to correction by the author(s).
Publisher: International Petroleum Technology Conference
Status: Published
Date Deposited: 09 Jun 2020 15:33
Last Modified: 09 Jun 2020 15:33
URI: https://ebooks.ub.rug.nl/id/eprint/329

Actions (login required)

View Item View Item