Artificial Neural Network and Regression Models for Predicting Intrusion of Non-Reacting Gases into Production Pipelines

Wax deposition and gelation of waxy crude oil in production pipelines are detrimental to crude oil transportation from offshore fields. A waxy crude oil forms intra-gel voids in pipelines under cooling mode, particularly below the pour point temperature. Consequently, intrusion of non-reacting gas i...

Full description

Saved in:
Bibliographic Details
Published in:Energies (Basel) 2022-03, Vol.15 (5), p.1725
Main Authors: Chala, Girma Tadesse, Negash, Berihun Mamo
Format: Article
Language:English
Subjects:
Algorithms
ANN
Cooling
Crude oil
Experiments
Gas pipelines
Gases
Intrusion
multilinear regression
Natural gas
Neural networks
Normal distribution
Ocean floor
Oil
Petroleum pipelines
Pipelines
Prediction models
Pressure
Regression analysis
restart pressure
restart time
Trial and error methods
Waxes
waxy crude oil
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Wax deposition and gelation of waxy crude oil in production pipelines are detrimental to crude oil transportation from offshore fields. A waxy crude oil forms intra-gel voids in pipelines under cooling mode, particularly below the pour point temperature. Consequently, intrusion of non-reacting gas into production pipelines has become a promising method to lessen the restart pressure required and clear the clogged gel. A trial-and-error method is currently employed to determine the required restart pressure and restart time in response to injected gas volume. However, this method is not always accurate and requires expert knowledge. In this study, predictive models based on an Artificial Neural Network (ANN) and multilinear regression are developed to predict restart pressure and time as a function of seabed temperature and non-reacting gas injected volume. The models’ outcomes are compared against experimental results available from the literature. The empirical models predicted the response variables with an absolute error of below 5% compared to the experimental studies. Thus, such models would allow accurate estimation of restart pressure, thereby improving transportation efficiency in offshore fields.
ISSN:1996-1073
1996-1073
DOI:10.3390/en15051725