Prediction of well production event using machine learning algorithms

Yara Alatrach, Carlos Mata, Pejman Shoeibi Omrani, Luigi Saputelli, Ram Narayanan, Mohammad Hamdan

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In this paper, a new approach was identified and tested to detect abnormal events in producing wells when a labeled dataset is unavailable or the number of instances are below 10% and are insufficient for conventional modelling methods. Autoencoders (AE), a type of unsupervised learning, are trained to learn normal behavior by trying to reconstruct the input data that is fed into the model. When run in prediction mode, low reconstruction errors are classified as Normal behavior whilst higher errors are classified as anomalous behavior. Different model structures were tested. An average accuracy of 94% with a precision and recall rate of 70% was achieved using a 6-Layered AE-NN model. The results of the models created show encouraging results and can help detect events and notify engineers when the well is deviates from expected behavior.

Original languageEnglish
Title of host publicationAbu Dhabi International Petroleum Exhibition and Conference 2020
PublisherSociety of Petroleum Engineers
ISBN (Electronic)9781613997345
DOIs
Publication statusPublished - 9 Nov 2020
EventAbu Dhabi International Petroleum Exhibition and Conference 2020 - Abu Dhabi, United Arab Emirates
Duration: 9 Nov 202012 Nov 2020

Conference

ConferenceAbu Dhabi International Petroleum Exhibition and Conference 2020
Abbreviated titleADIP 2020
Country/TerritoryUnited Arab Emirates
CityAbu Dhabi
Period9/11/2012/11/20

ASJC Scopus subject areas

  • Geotechnical Engineering and Engineering Geology
  • Fuel Technology
  • Geochemistry and Petrology

Fingerprint

Dive into the research topics of 'Prediction of well production event using machine learning algorithms'. Together they form a unique fingerprint.

Cite this