Prediction of Bottom-Hole Flowing Pressure using general regression neural network

This paper presents the application of Surrogate Reservoir Model (SRM) for predicting the Bottom-Hole Flowing Pressure (BHFP) on an initially undersaturated reservoir. SRM is recently introduce technology that is used to replicates the results of numerical simulation model. High computational cost a...

Full description

Main Authors: Memon, P.Q., Yong, S.-P., Pao, W., Seanl, P.J.
Format: Conference or Workshop Item
Institution: Universiti Teknologi Petronas
Record Id / ISBN-0: utp-eprints.31205 /
Published: Institute of Electrical and Electronics Engineers Inc. 2014
Online Access: https://www.scopus.com/inward/record.uri?eid=2-s2.0-84938794886&doi=10.1109%2fICCOINS.2014.6868849&partnerID=40&md5=00cb0ecb85a2b456650b914e2f58c972
http://eprints.utp.edu.my/31205/
Tags: Add Tag
No Tags, Be the first to tag this record!
id utp-eprints.31205
recordtype eprints
spelling utp-eprints.312052022-03-25T09:02:54Z Prediction of Bottom-Hole Flowing Pressure using general regression neural network Memon, P.Q. Yong, S.-P. Pao, W. Seanl, P.J. This paper presents the application of Surrogate Reservoir Model (SRM) for predicting the Bottom-Hole Flowing Pressure (BHFP) on an initially undersaturated reservoir. SRM is recently introduce technology that is used to replicates the results of numerical simulation model. High computational cost and long processing time limits our ability to perform comprehensive sensitivity analysis and quantify uncertainties associated with reservoir because reservoir model that contains large number of grids in its geological structure takes considerable amount of time for a single simulation run. And also making hundred and thousands simulation runs is considered as a cumbersome process and sometimes impractical. SRM is considered as as a solution tool to tackle this issue. SRM uses Artificial Neural Network (ANN) technique for the reservoir simulation and modeling. In this paper, the results of SRM for predicting BHFP is presented and a reservoir simulation model has been presented using Black Oil Applied Simulation Tool (BOAST). To build any SRM, it requires small number of runs to train the model. Once we train the SRM, it can generate hundred and thousands of simulation runs in a matter of seconds. As a part of this system, it is proposed to develop a SRM extraction based on ANN to enhance the realization run time. © 2014 IEEE. Institute of Electrical and Electronics Engineers Inc. 2014 Conference or Workshop Item NonPeerReviewed https://www.scopus.com/inward/record.uri?eid=2-s2.0-84938794886&doi=10.1109%2fICCOINS.2014.6868849&partnerID=40&md5=00cb0ecb85a2b456650b914e2f58c972 Memon, P.Q. and Yong, S.-P. and Pao, W. and Seanl, P.J. (2014) Prediction of Bottom-Hole Flowing Pressure using general regression neural network. In: UNSPECIFIED. http://eprints.utp.edu.my/31205/
institution Universiti Teknologi Petronas
collection UTP Institutional Repository
description This paper presents the application of Surrogate Reservoir Model (SRM) for predicting the Bottom-Hole Flowing Pressure (BHFP) on an initially undersaturated reservoir. SRM is recently introduce technology that is used to replicates the results of numerical simulation model. High computational cost and long processing time limits our ability to perform comprehensive sensitivity analysis and quantify uncertainties associated with reservoir because reservoir model that contains large number of grids in its geological structure takes considerable amount of time for a single simulation run. And also making hundred and thousands simulation runs is considered as a cumbersome process and sometimes impractical. SRM is considered as as a solution tool to tackle this issue. SRM uses Artificial Neural Network (ANN) technique for the reservoir simulation and modeling. In this paper, the results of SRM for predicting BHFP is presented and a reservoir simulation model has been presented using Black Oil Applied Simulation Tool (BOAST). To build any SRM, it requires small number of runs to train the model. Once we train the SRM, it can generate hundred and thousands of simulation runs in a matter of seconds. As a part of this system, it is proposed to develop a SRM extraction based on ANN to enhance the realization run time. © 2014 IEEE.
format Conference or Workshop Item
author Memon, P.Q.
Yong, S.-P.
Pao, W.
Seanl, P.J.
spellingShingle Memon, P.Q.
Yong, S.-P.
Pao, W.
Seanl, P.J.
Prediction of Bottom-Hole Flowing Pressure using general regression neural network
author_sort Memon, P.Q.
title Prediction of Bottom-Hole Flowing Pressure using general regression neural network
title_short Prediction of Bottom-Hole Flowing Pressure using general regression neural network
title_full Prediction of Bottom-Hole Flowing Pressure using general regression neural network
title_fullStr Prediction of Bottom-Hole Flowing Pressure using general regression neural network
title_full_unstemmed Prediction of Bottom-Hole Flowing Pressure using general regression neural network
title_sort prediction of bottom-hole flowing pressure using general regression neural network
publisher Institute of Electrical and Electronics Engineers Inc.
publishDate 2014
url https://www.scopus.com/inward/record.uri?eid=2-s2.0-84938794886&doi=10.1109%2fICCOINS.2014.6868849&partnerID=40&md5=00cb0ecb85a2b456650b914e2f58c972
http://eprints.utp.edu.my/31205/
_version_ 1741197535185207296
score 11.62408

Similar Items