Application of supervised machine learning paradigms in the prediction of petroleum reservoir properties: Comparative analysis of ANN and SVM models

Daniel Asante Otchere*, Tarek Omar Arbi Ganat, Raoof Gholami, Syahrir Ridha

*Corresponding author for this work

Research output: Contribution to journalReview articlepeer-review

70 Citations (Scopus)

Abstract

The advent of Artificial Intelligence (AI) in the petroleum industry has seen an increase in its use in exploration, development, production, reservoir engineering and management planning to accelerate decision making, reduce cost and time. Supervised machine learning has gained much popularity in establishing a relationship between complex non-linear datasets. This type of machine learning algorithm has showcased its superiority over petroleum engineering regression techniques in terms of prediction errors for high dimensional data, computational power and memory. This review focuses on the most widely used machine learning algorithm employed in the petroleum industry, the Artificial Neural Network (ANN) with different shallow models used in reservoir characterisation. The Support Vector Machine (SVM) and Relevant Vector Machine (RVM) has over the years emerged as competitive algorithms where in most cases based on this review it outperformed the ANN. This makes it preferable than the ANN when there are limited data sets. Finally, hybridisation of multiple algorithms methodologies also showed improved performance over singularly applied algorithms offering a pathway in improving reservoir characterisation based on supervised machine learning as future scope of work.

Original languageEnglish
Article number108182
JournalJournal of Petroleum Science and Engineering
Volume200
DOIs
Publication statusPublished - May 2021
Externally publishedYes

Keywords

  • Artificial intelligence (AI)
  • Artificial neural network (ANN)
  • Hybrid intelligent system
  • Relevance vector machine (RVM)
  • Supervised machine learning
  • Support vector machine (SVM)

ASJC Scopus subject areas

  • Fuel Technology
  • Geotechnical Engineering and Engineering Geology

Cite this