$Prediction model of fluid flow behavior in the naturally fractured reservoirs by machine learning$

Prediction model of fluid flow behavior in the naturally fractured reservoirs by machine learning

The naturally fractured reservoirs are one of the products of the tectonic movements, which increases the permeability and conductivity of the fractures. The instability of the permeability and conductivity effect on the fluid's flow path causes problems during the transfer of the fluids from t...

Full description

Saved in:

Bibliographic Details
Main Author:	Al-Obaidy, Mustafa Mudhafar Shawkat
Format:	Thesis
Language:	English
Published:	2022
Subjects:	Q Science (General) TP Chemical technology
Online Access:	http://eprints.utm.my/id/eprint/102521/1/MustafaMudhafarShawkatMSChE2022.pdf
Tags:	Add Tag No Tags, Be the first to tag this record!

id	my-utm-ep.102521
record_format	uketd_dc
spelling	my-utm-ep.1025212023-09-03T06:38:26Z Prediction model of fluid flow behavior in the naturally fractured reservoirs by machine learning 2022 Al-Obaidy, Mustafa Mudhafar Shawkat Q Science (General) TP Chemical technology The naturally fractured reservoirs are one of the products of the tectonic movements, which increases the permeability and conductivity of the fractures. The instability of the permeability and conductivity effect on the fluid's flow path causes problems during the transfer of the fluids from the matrix to the fractures and fluids losses during production. In addition, these complications made it difficult for engineers to estimate fluid flow during production. The fracture properties study is essential to model the fluids flow paths such as the fracture porosity, permeability, and the shape factor which are considered essential in the stability of fluids flow. To examine this, this research introduced new models called the Decision trees model (DT), Random Forest model (RF), Ridge regression model, LASSO regression model, and K-nearest regression model. The research studied the fracture properties in naturally fractured reservoirs like the fracture porosity and the shape factor. The datasets used in this study were collected from previous studies “i.eTexas oil and gas fields” to build a predictive intelligence model for fluid flow characteristics. The prediction process was conducted based on interporosity flow coefficient, storativity ratio, wellbore radius, matrix permeability, and fracture permeability as input data and shape factor (SF) and fracture porosity (FP) as output data. This study revealed a positive finding for the adopted machine learning models and was superior in using R2 of accuracy based on the quantitative metrics. The results of the test showed an increase in the readings of the fractured porosity and the shape factor compared to the actual data for oil and gas, which improved the fracture properties. For fluid flow, fluids are designed on the basis that the flow is radial. All models exhibited similar behavior of fluid flow, as the fluids were traveling parallel and radial, which changed the fluid properties except for the LASSO model. The research results of LASSO found that the accuracy of gas flow is less although gas flow is faster than oil flow in naturally fractured reservoirs. In conclusion, the radial flow model cannot be implemented for all fluids in the naturally fractured reservoirs that prefer to assume as flow is pseudo steady state. Overall, the research emphasized implementing computer aid models for naturally fractured reservoir analysis, which gives more details on the extensive executing techniques, such as injection or the creation of artificial cracks, to minimize hydrocarbon losses or leakage. 2022 Thesis http://eprints.utm.my/id/eprint/102521/ http://eprints.utm.my/id/eprint/102521/1/MustafaMudhafarShawkatMSChE2022.pdf application/pdf en public http://dms.library.utm.my:8080/vital/access/manager/Repository/vital:152368 masters Universiti Teknologi Malaysia, Faculty of Engineering - School of Chemical & Energy Engineering Faculty of Engineering - School of Chemical & Energy Engineering
institution	Universiti Teknologi Malaysia
collection	UTM Institutional Repository
language	English
topic	Q Science (General) TP Chemical technology
spellingShingle	Q Science (General) TP Chemical technology Al-Obaidy, Mustafa Mudhafar Shawkat Prediction model of fluid flow behavior in the naturally fractured reservoirs by machine learning
description	The naturally fractured reservoirs are one of the products of the tectonic movements, which increases the permeability and conductivity of the fractures. The instability of the permeability and conductivity effect on the fluid's flow path causes problems during the transfer of the fluids from the matrix to the fractures and fluids losses during production. In addition, these complications made it difficult for engineers to estimate fluid flow during production. The fracture properties study is essential to model the fluids flow paths such as the fracture porosity, permeability, and the shape factor which are considered essential in the stability of fluids flow. To examine this, this research introduced new models called the Decision trees model (DT), Random Forest model (RF), Ridge regression model, LASSO regression model, and K-nearest regression model. The research studied the fracture properties in naturally fractured reservoirs like the fracture porosity and the shape factor. The datasets used in this study were collected from previous studies “i.eTexas oil and gas fields” to build a predictive intelligence model for fluid flow characteristics. The prediction process was conducted based on interporosity flow coefficient, storativity ratio, wellbore radius, matrix permeability, and fracture permeability as input data and shape factor (SF) and fracture porosity (FP) as output data. This study revealed a positive finding for the adopted machine learning models and was superior in using R2 of accuracy based on the quantitative metrics. The results of the test showed an increase in the readings of the fractured porosity and the shape factor compared to the actual data for oil and gas, which improved the fracture properties. For fluid flow, fluids are designed on the basis that the flow is radial. All models exhibited similar behavior of fluid flow, as the fluids were traveling parallel and radial, which changed the fluid properties except for the LASSO model. The research results of LASSO found that the accuracy of gas flow is less although gas flow is faster than oil flow in naturally fractured reservoirs. In conclusion, the radial flow model cannot be implemented for all fluids in the naturally fractured reservoirs that prefer to assume as flow is pseudo steady state. Overall, the research emphasized implementing computer aid models for naturally fractured reservoir analysis, which gives more details on the extensive executing techniques, such as injection or the creation of artificial cracks, to minimize hydrocarbon losses or leakage.
format	Thesis
qualification_level	Master's degree
author	Al-Obaidy, Mustafa Mudhafar Shawkat
author_facet	Al-Obaidy, Mustafa Mudhafar Shawkat
author_sort	Al-Obaidy, Mustafa Mudhafar Shawkat
title	Prediction model of fluid flow behavior in the naturally fractured reservoirs by machine learning
title_short	Prediction model of fluid flow behavior in the naturally fractured reservoirs by machine learning
title_full	Prediction model of fluid flow behavior in the naturally fractured reservoirs by machine learning
title_fullStr	Prediction model of fluid flow behavior in the naturally fractured reservoirs by machine learning
title_full_unstemmed	Prediction model of fluid flow behavior in the naturally fractured reservoirs by machine learning
title_sort	prediction model of fluid flow behavior in the naturally fractured reservoirs by machine learning
granting_institution	Universiti Teknologi Malaysia, Faculty of Engineering - School of Chemical & Energy Engineering
granting_department	Faculty of Engineering - School of Chemical & Energy Engineering
publishDate	2022
url	http://eprints.utm.my/id/eprint/102521/1/MustafaMudhafarShawkatMSChE2022.pdf
_version_	1776100942828011520

Prediction model of fluid flow behavior in the naturally fractured reservoirs by machine learning

Similar Items