Pressure Drop in Vertical Multiphase Flow using Neuro Fuzzy Technique; A Comparative Approach
The sole objective of this study is to develop a model for estimating the pressure drop in vertical multiphase flow using one of the artificial intelligence techniques which is Neuro Fuzzy Systems with a good and acceptable accuracy that can work for a wide range of well flowing conditions that can...
Saved in:
Main Author: | |
---|---|
Format: | Final Year Project |
Language: | English |
Published: |
IRC
2014
|
Subjects: | |
Online Access: | http://utpedia.utp.edu.my/14569/1/Dissertation-PE-Mohamed%20Alaa.pdf http://utpedia.utp.edu.my/14569/ |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
id |
my-utp-utpedia.14569 |
---|---|
record_format |
eprints |
spelling |
my-utp-utpedia.145692017-01-25T09:36:32Z http://utpedia.utp.edu.my/14569/ Pressure Drop in Vertical Multiphase Flow using Neuro Fuzzy Technique; A Comparative Approach AlaaElDin Mohamed, Mohamed TA Engineering (General). Civil engineering (General) The sole objective of this study is to develop a model for estimating the pressure drop in vertical multiphase flow using one of the artificial intelligence techniques which is Neuro Fuzzy Systems with a good and acceptable accuracy that can work for a wide range of well flowing conditions that can replace the rigorous empirical and mechanistic correlations. In this study a number of 206 data sets collected from some fields in the Middle East were used to develop the Neuro Fuzzy Model. Many attempts have been done to estimate the pressure drop in vertical multiphase flow starting from the homogeneous models, the empirical models and the mechanistic models. But yet, none of the traditional correlations works well for the variety of well conditions that are found in the oil industry. Thus, the accuracy of the old pressure drop correlations cannot be raised to a generally accepted level. For this purpose, one of the artificial intelligence techniques (Neuro Fuzzy System) is used to have a significant reduction in the error involved with estimating the pressure drop. The Neuro Fuzzy Model was developed through 3 stages; Training, Validation, Testing. The developed Neuro Fuzzy Model has successfully achieved the lowest Average Absolute Percentage Error (AAPE%) of 2.92% that could overcome all the empirical and mechanistic correlations when tested against the same set of data. It can be concluded that Neuro Fuzzy system has overcame the performance of the models currently used in the industry. IRC 2014-09 Final Year Project NonPeerReviewed application/pdf en http://utpedia.utp.edu.my/14569/1/Dissertation-PE-Mohamed%20Alaa.pdf AlaaElDin Mohamed, Mohamed (2014) Pressure Drop in Vertical Multiphase Flow using Neuro Fuzzy Technique; A Comparative Approach. IRC, Universiti Teknologi PETRONAS. (Unpublished) |
institution |
Universiti Teknologi Petronas |
building |
UTP Resource Centre |
collection |
Institutional Repository |
continent |
Asia |
country |
Malaysia |
content_provider |
Universiti Teknologi Petronas |
content_source |
UTP Electronic and Digitized Intellectual Asset |
url_provider |
http://utpedia.utp.edu.my/ |
language |
English |
topic |
TA Engineering (General). Civil engineering (General) |
spellingShingle |
TA Engineering (General). Civil engineering (General) AlaaElDin Mohamed, Mohamed Pressure Drop in Vertical Multiphase Flow using Neuro Fuzzy Technique; A Comparative Approach |
description |
The sole objective of this study is to develop a model for estimating the pressure drop in vertical multiphase flow using one of the artificial intelligence techniques which is Neuro Fuzzy Systems with a good and acceptable accuracy that can work for a wide range of well flowing conditions that can replace the rigorous empirical and mechanistic correlations.
In this study a number of 206 data sets collected from some fields in the Middle East were used to develop the Neuro Fuzzy Model.
Many attempts have been done to estimate the pressure drop in vertical multiphase flow starting from the homogeneous models, the empirical models and the mechanistic models. But yet, none of the traditional correlations works well for the variety of well conditions that are found in the oil industry. Thus, the accuracy of the old pressure drop correlations cannot be raised to a generally accepted level. For this purpose, one of the artificial intelligence techniques (Neuro Fuzzy System) is used to have a significant reduction in the error involved with estimating the pressure drop.
The Neuro Fuzzy Model was developed through 3 stages; Training, Validation, Testing.
The developed Neuro Fuzzy Model has successfully achieved the lowest Average Absolute Percentage Error (AAPE%) of 2.92% that could overcome all the empirical and mechanistic correlations when tested against the same set of data. It can be concluded that Neuro Fuzzy system has overcame the performance of the models currently used in the industry. |
format |
Final Year Project |
author |
AlaaElDin Mohamed, Mohamed |
author_facet |
AlaaElDin Mohamed, Mohamed |
author_sort |
AlaaElDin Mohamed, Mohamed |
title |
Pressure Drop in Vertical Multiphase Flow using Neuro Fuzzy Technique; A Comparative Approach |
title_short |
Pressure Drop in Vertical Multiphase Flow using Neuro Fuzzy Technique; A Comparative Approach |
title_full |
Pressure Drop in Vertical Multiphase Flow using Neuro Fuzzy Technique; A Comparative Approach |
title_fullStr |
Pressure Drop in Vertical Multiphase Flow using Neuro Fuzzy Technique; A Comparative Approach |
title_full_unstemmed |
Pressure Drop in Vertical Multiphase Flow using Neuro Fuzzy Technique; A Comparative Approach |
title_sort |
pressure drop in vertical multiphase flow using neuro fuzzy technique; a comparative approach |
publisher |
IRC |
publishDate |
2014 |
url |
http://utpedia.utp.edu.my/14569/1/Dissertation-PE-Mohamed%20Alaa.pdf http://utpedia.utp.edu.my/14569/ |
_version_ |
1739832018540167168 |
score |
13.209306 |