Predicting wax deposition using robust machine learning techniques

Menad Nait Amar

doi:10.1016/j.petlm.2021.07.005

Return

Predicting wax deposition using robust machine learning techniques

2022-06-01

Menad Nait Amar *

Ashkan Jahanbani Ghahfarokhi

Cuthbert Shang Wui Ng

DOI：10.1016/j.petlm.2021.07.005

Original

Save

View PDF

Abstract

En 中文

Accurate prediction of wax deposition is of vital interest in digitalized systems to avoid many issues that interrupt the flow assurance during production of hydrocarbon fluids. The present investigation aims at establishing rigorous intelligent schemes for predicting wax deposition under extensive production conditions. To do so, multilayer perceptron (MLP) optimized with Levenberg-Marquardt algorithm (MLPLMA) and Bayesian Regularization algorithm (MLP-BR) were taught using 88 experimental measurements. These latter were described by some independent variables, namely temperature (in K), specific gravity, and compositions of C1-C3, C4-C7, C8-C15, C16-C22, C23-C29 and C30 thorn . The obtained results showed that MLP-LMA achieved the best performance with an overall root mean square error of 0.2198 and a coefficient of determination (R2) of 0.9971. The performance comparison revealed that MLP-LMA outperforms the prior approaches in the literature. (c) 2021 Southwest Petroleum University. Publishing services by Elsevier B.V. on behalf of KeAi Communications Co. Ltd. This is an open access article under the CC BY-NC-ND license (http:// creativecommons.org/licenses/by-nc-nd/4.0/).

Keywords:

Wax deposition

Multilayer perceptron

Levenberg-marquardt algorithm

Flow assurance

AI Summary

Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.