Prediction by Machine Learning in Nanoparticles-Based Enhanced Oil Recovery

Aug 3, 2026

Upload by Ophelia

Nanotechnology is positioned to reshape industrial processes, with the petroleum sector actively adopting nano-enhanced oil recovery. A key limitation is insufficient data on nanoparticle transport in porous media. This research applies machine learning to predict nanoparticle transport using finite-difference simulations derived from a modified linear adsorption model. Synthetic datasets train models including artificial neural networks, decision trees, and random forests, enabling accurate estimates of nanoparticle concentration and pore volume.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages18

Reading Time45 min

AccessFree to download

Introduction

Enhanced oil recovery with nanoparticles

Background

Traditional and nano-EOR recovery context

Methodology

Finite difference dataset generation and modified adsorption model

Machine Learning Modeling

ANN, decision tree, and random forest approaches

Results and Discussion

Predictive performance evaluation

Conclusion and Future Research Scope

Mathematical Modelling and Numerical Simulation with Applications, 2024, 4(4), 544–561 [https://dergipark.org.tr/en/pub/mmnsa](https://dergipark.org.tr/en/pub/mmnsa) ISSN Online: 2791-8564 / Open Access [https://doi.org/10.53391/mmnsa.1498986](https://doi.org/10.53391/mmnsa.1498986) R E S EA R C H PA P E R Prediction by machine learning in nanoparticles-based enhanced oil recovery Pavan Patel 1,*,‡, Saroj R. Yadav 1,‡, Mohamed F. El-Amin 2,‡ and Mustafa Yıldız 3,‡ 1Department of Mathematics, Sardar Vallabhbhai National Institute of Technology, Surat, Gujarat, India, 2Energy Research Lab., College of Engineering, Effat University, Jeddah, Saudi Arabia, 3Department of Mathematics, Bartın University, Bartın 74100, Türkiye *Corresponding Author ‡[pavanpatel704@gmail.com](pavanpatel704@gmail.com) (Pavan Patel); [sry@amhd.svnit.ac.in](sry@amhd.svnit.ac.in) (Saroj R.Yadav); [momousa@effatuniversity.edu.sa](momousa@effatuniversity.edu.sa) (Mohamed F. El-Amin); [myildiz@bartin.edu.tr](myildiz@bartin.edu.tr) (Mustafa Yıldız) | Abstract
Nanotechnology is on the brink of transforming numerous industrial sectors, and the petroleum industry stands as a front-runner in embracing these revolutionary advancements. In recent years, a growing interest has occurred in leveraging nanotechnology within the petroleum industry. Extensive research studies on nano-enhanced oil recovery (nano-EOR) have consistently delivered promising outcomes, underscoring its potential to elevate oil production substantially. However, a notable challenge persists within this domain due to the limited data availability concerning nanoparticle transport in porous media. This paper uses machine learning techniques to predict nanoparticle transport in porous media. This study uses the finite difference method to generate simulated datasets from a modified linear adsorption model. These simulated datasets are used to train machine learning models for prediction by considering artificial neural network (ANNs), decision tree (DT), and random forest (RF) . We achieve mean squared values for ANN as 0.0478 (training), 0.0496 (testing), 0.0509 (validation), and R-squared values as 0.9798 (training), 0.9780 (testing), 0.9773 (validation), and for DT and RF mean squared values are 0.014683, 0.009807, and R squared values are 0.928775, 0.952425 .
Keywords: Enhanced oil recovery; nanoparticles; machine learning; fluid flow
AMS 2020 Classification: 35Q35; 68T07 | | --- | | 1 Introduction
Enhanced oil recovery (EOR) is a technique employed to augment the production of hydrocarbons from a well. The selection of an appropriate EOR method is contingent upon the data amassed | ➤ Received: 10.06.2024 ➤ Revised: 05.12.2024 ➤ Accepted: 30.12.2024 ➤ Published: 30.12.2024 during the reservoir assessment phase. Nanoparticles (NPs) are utilized in EOR because a significant portion, approximately 60% to 70% of the hydrocarbons in most oil reservoirs remain unrecovered through primary and secondary recovery methods. Nanometer-sized particles can migrate, for instance, through the process of diffusion, into the confined pore throats within reservoir rock formations. Within these narrow spaces, these particles can engage in various physical and chemical interactions with the components of the reservoir fluid-rock system. These interactions play a pivotal role in altering the multiphase flow parameters within the reservoir. Ultimately, this phenomenon contributes significantly to enhancing oil recovery from the reservoirs, leading to more efficient and effective extraction processes [1, 2] . Silica-based NPs stand out as very promising materials in a variety of applications, owing to the ease of manufacture and the diversity in designing their surface features [3–5] . This paper employs numerical modeling, specifically based on filtration theory, to generate datasets essential for the application of machine learning techniques aimed at predicting nanopart

FAQs

What problem does the paper address in nano-enhanced oil recovery?

The study addresses the lack of publicly available data describing nanoparticle transport in porous media, which limits predictive modeling for nano-EOR.

How are the training datasets for machine learning generated?

Datasets are generated via finite difference simulations based on a modified linear adsorption model, producing synthetic data for model training.

Which machine learning models are used to predict nanoparticle transport?

The paper uses artificial neural networks (ANNs), decision trees (DT), and random forests (RF) to predict nanoparticle concentration and pore volume.

Aug 3, 2026

Upload by Ophelia