Pore Pressure Estimation Using Artificial Neural Network
Pore pressure prediction is important in predrilling process and it is necessary in prerequisite before take a major step to start a drill a well. It also will determine the safety of drilling process and profit of the company. Realising the significance of pore pressure prediction, engineers and ex...
Saved in:
Main Author: | |
---|---|
Format: | Final Year Project |
Language: | English |
Published: |
Universiti Teknologi PETRONAS
2014
|
Subjects: | |
Online Access: | http://utpedia.utp.edu.my/14317/1/Dissertation_Muhammad%20Hafiz%20bin%20Abidin_13699_PE.pdf http://utpedia.utp.edu.my/14317/ |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
id |
my-utp-utpedia.14317 |
---|---|
record_format |
eprints |
spelling |
my-utp-utpedia.143172017-01-25T09:37:02Z http://utpedia.utp.edu.my/14317/ Pore Pressure Estimation Using Artificial Neural Network Abidin, Muhammad Hafiz T Technology (General) Pore pressure prediction is important in predrilling process and it is necessary in prerequisite before take a major step to start a drill a well. It also will determine the safety of drilling process and profit of the company. Realising the significance of pore pressure prediction, engineers and experts worldwide have done extensive work on estimation of pore pressure. The conventional method that used to predict pore pressure have their own hypothesis and some will limited on several condition, thus yield less accuracy in result. Nevertheless, the numerous of data handling to consider will bring the human error in the prediction pore pressure. Therefore, this project aims to select the best ANN modelling approach for pore pressure prediction and will show the improved result by comparing an ANN model with the conventional model. The method chosen is artificial neural network (ANN) which consist of 3 main layer which is input, hidden and output layer. The input of this model are gamma ray, density, interval transit time and depth. The optimized number of neuron are 2, 10 and 1 and the activation function selected is the log sigmoid function for first layer and hyperbolic tangent sigmoid for second and third layer. The accuracy and feasibility of this model is 5.0048% and applicable for normal and abnormal pressure field. Universiti Teknologi PETRONAS 2014-05 Final Year Project NonPeerReviewed application/pdf en http://utpedia.utp.edu.my/14317/1/Dissertation_Muhammad%20Hafiz%20bin%20Abidin_13699_PE.pdf Abidin, Muhammad Hafiz (2014) Pore Pressure Estimation Using Artificial Neural Network. 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 |
T Technology (General) |
spellingShingle |
T Technology (General) Abidin, Muhammad Hafiz Pore Pressure Estimation Using Artificial Neural Network |
description |
Pore pressure prediction is important in predrilling process and it is necessary in prerequisite before take a major step to start a drill a well. It also will determine the safety of drilling process and profit of the company. Realising the significance of pore pressure prediction, engineers and experts worldwide have done extensive work on estimation of pore pressure. The conventional method that used to predict pore pressure have their own hypothesis and some will limited on several condition, thus yield less accuracy in result. Nevertheless, the numerous of data handling to consider will bring the human error in the prediction pore pressure. Therefore, this project aims to select the best ANN modelling approach for pore pressure prediction and will show the improved result by comparing an ANN model with the conventional model. The method chosen is artificial neural network (ANN) which consist of 3 main layer which is input, hidden and output layer. The input of this model are gamma ray, density, interval transit time and depth. The optimized number of neuron are 2, 10 and 1 and the activation function selected is the log sigmoid function for first layer and hyperbolic tangent sigmoid for second and third layer. The accuracy and feasibility of this model is 5.0048% and applicable for normal and abnormal pressure field. |
format |
Final Year Project |
author |
Abidin, Muhammad Hafiz |
author_facet |
Abidin, Muhammad Hafiz |
author_sort |
Abidin, Muhammad Hafiz |
title |
Pore Pressure Estimation Using Artificial Neural Network |
title_short |
Pore Pressure Estimation Using Artificial Neural Network |
title_full |
Pore Pressure Estimation Using Artificial Neural Network |
title_fullStr |
Pore Pressure Estimation Using Artificial Neural Network |
title_full_unstemmed |
Pore Pressure Estimation Using Artificial Neural Network |
title_sort |
pore pressure estimation using artificial neural network |
publisher |
Universiti Teknologi PETRONAS |
publishDate |
2014 |
url |
http://utpedia.utp.edu.my/14317/1/Dissertation_Muhammad%20Hafiz%20bin%20Abidin_13699_PE.pdf http://utpedia.utp.edu.my/14317/ |
_version_ |
1739831988559282176 |
score |
13.160551 |