Prediction of PVT properties in crude oil systems using support vector machines
Calculation of reserves in an oil reservoir and the determination of its performance and economics require good knowledge of its physical properties. Accurate determination of the pressure-volume-temperature (PVT) properties such as the bubble point pressure (Pb) and the oil formation volume factor...
Saved in:
| Main Authors: | , , , |
|---|---|
| Other Authors: | |
| Format: | Conference paper |
| Published: |
2023
|
| Subjects: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| id |
my.uniten.dspace-30758 |
|---|---|
| record_format |
dspace |
| spelling |
my.uniten.dspace-307582023-12-29T15:52:41Z Prediction of PVT properties in crude oil systems using support vector machines Nagi J. Kiong T.S. Ahmed S.K. Nagi F. 25825455100 15128307800 25926812900 56272534200 Bubble point pressue Oil formation volume factor PVT properties Support vector machine Support vector regression Crude oil Forecasting Gears Learning algorithms Mathematical models Multilayer neural networks Petroleum analysis Petroleum reservoir engineering Petroleum reservoirs Regression analysis Support vector machines Sustainable development Vectors Artificial Neural Network Bubble point pressure Bubble points Comparative studies Crude oil system Data sets Empirical correlations Gas oil ratios Input features Machine learning techniques Non-linear regression Oil formation Oil gravity Oil reservoirs Prediction model Pressure-volume-temperature properties PVT properties Relative density Reservoir temperatures Support vector regressions Training and testing Oil field development Calculation of reserves in an oil reservoir and the determination of its performance and economics require good knowledge of its physical properties. Accurate determination of the pressure-volume-temperature (PVT) properties such as the bubble point pressure (Pb) and the oil formation volume factor (Bob) are important in the primary and subsequent development of an oil field. This paper proposes Support Vector Machines (SVMs) as a novel machine learning technique for predicting outputs in uncertain situations using the ?-Support Vector Regression (?-SVR) method. The objective of this research is to investigate the capability of SVRs in modeling PVT properties of crude oil systems and solving existing Artificial Neural Network (ANN) drawbacks. Three datasets used for training and testing the SVR prediction model were collected from distinct published sources. The ?-SVR model incorporates four input features from the datasets: (1) solution gas-oil ratio, (2) reservoir temperature, (3) oil gravity and, (4) gas relative density. A comparative study is carried out to compare ?-SVR performance with ANNs, nonlinear regression, and different empirical correlation techniques. The results obtained reveal that the ?-SVR once successfully trained and optimized is more accurate, reliable, and outperforms the other existing approaches such as empirical correlation for estimating crude oil PVT properties. �2009 IEEE. Final 2023-12-29T07:52:41Z 2023-12-29T07:52:41Z 2009 Conference paper 10.1109/ICEENVIRON.2009.5398681 2-s2.0-77949584632 https://www.scopus.com/inward/record.uri?eid=2-s2.0-77949584632&doi=10.1109%2fICEENVIRON.2009.5398681&partnerID=40&md5=190c3284d24e5862c183e62bc90bd28f https://irepository.uniten.edu.my/handle/123456789/30758 5398681 1 5 Scopus |
| institution |
Universiti Tenaga Nasional |
| building |
UNITEN Library |
| collection |
Institutional Repository |
| continent |
Asia |
| country |
Malaysia |
| content_provider |
Universiti Tenaga Nasional |
| content_source |
UNITEN Institutional Repository |
| url_provider |
http://dspace.uniten.edu.my/ |
| topic |
Bubble point pressue Oil formation volume factor PVT properties Support vector machine Support vector regression Crude oil Forecasting Gears Learning algorithms Mathematical models Multilayer neural networks Petroleum analysis Petroleum reservoir engineering Petroleum reservoirs Regression analysis Support vector machines Sustainable development Vectors Artificial Neural Network Bubble point pressure Bubble points Comparative studies Crude oil system Data sets Empirical correlations Gas oil ratios Input features Machine learning techniques Non-linear regression Oil formation Oil gravity Oil reservoirs Prediction model Pressure-volume-temperature properties PVT properties Relative density Reservoir temperatures Support vector regressions Training and testing Oil field development |
| spellingShingle |
Bubble point pressue Oil formation volume factor PVT properties Support vector machine Support vector regression Crude oil Forecasting Gears Learning algorithms Mathematical models Multilayer neural networks Petroleum analysis Petroleum reservoir engineering Petroleum reservoirs Regression analysis Support vector machines Sustainable development Vectors Artificial Neural Network Bubble point pressure Bubble points Comparative studies Crude oil system Data sets Empirical correlations Gas oil ratios Input features Machine learning techniques Non-linear regression Oil formation Oil gravity Oil reservoirs Prediction model Pressure-volume-temperature properties PVT properties Relative density Reservoir temperatures Support vector regressions Training and testing Oil field development Nagi J. Kiong T.S. Ahmed S.K. Nagi F. Prediction of PVT properties in crude oil systems using support vector machines |
| description |
Calculation of reserves in an oil reservoir and the determination of its performance and economics require good knowledge of its physical properties. Accurate determination of the pressure-volume-temperature (PVT) properties such as the bubble point pressure (Pb) and the oil formation volume factor (Bob) are important in the primary and subsequent development of an oil field. This paper proposes Support Vector Machines (SVMs) as a novel machine learning technique for predicting outputs in uncertain situations using the ?-Support Vector Regression (?-SVR) method. The objective of this research is to investigate the capability of SVRs in modeling PVT properties of crude oil systems and solving existing Artificial Neural Network (ANN) drawbacks. Three datasets used for training and testing the SVR prediction model were collected from distinct published sources. The ?-SVR model incorporates four input features from the datasets: (1) solution gas-oil ratio, (2) reservoir temperature, (3) oil gravity and, (4) gas relative density. A comparative study is carried out to compare ?-SVR performance with ANNs, nonlinear regression, and different empirical correlation techniques. The results obtained reveal that the ?-SVR once successfully trained and optimized is more accurate, reliable, and outperforms the other existing approaches such as empirical correlation for estimating crude oil PVT properties. �2009 IEEE. |
| author2 |
25825455100 |
| author_facet |
25825455100 Nagi J. Kiong T.S. Ahmed S.K. Nagi F. |
| format |
Conference paper |
| author |
Nagi J. Kiong T.S. Ahmed S.K. Nagi F. |
| author_sort |
Nagi J. |
| title |
Prediction of PVT properties in crude oil systems using support vector machines |
| title_short |
Prediction of PVT properties in crude oil systems using support vector machines |
| title_full |
Prediction of PVT properties in crude oil systems using support vector machines |
| title_fullStr |
Prediction of PVT properties in crude oil systems using support vector machines |
| title_full_unstemmed |
Prediction of PVT properties in crude oil systems using support vector machines |
| title_sort |
prediction of pvt properties in crude oil systems using support vector machines |
| publishDate |
2023 |
| _version_ |
1806427481250463744 |
| score |
13.222552 |