TY  - CONF
KW  - Crude oil; Data handling; Gasoline; Multiple linear regression; Neural networks; Petroleum reservoir engineering
KW  -  Fluid property; Group method of data handling; Neural-networks; Nigerians; Oil formation volume factors; PVT properties; Regression; Reservoir engineering; Reservoir fluid; Support vectors machine
KW  -  Support vector machines
N1  - cited By 1; Conference of 2022 SPE Nigeria Annual International Conference and Exhibition, NAIC 2022 ; Conference Date: 1 August 2022 Through 3 August 2022; Conference Code:181613
SN  - 9781613999547
TI  - New Oil Formation Volume Factor Correlation for Nigerian Crude Oils
ID  - scholars17556
A1  - Atthi, A.J.
A1  - Sulaimon, A.A.
A1  - Akinsete, O.K.
Y1  - 2022///
UR  - https://www.scopus.com/inward/record.uri?eid=2-s2.0-85136831724&doi=10.2118%2f211968-MS&partnerID=40&md5=591cb225ea28b03fdcac23394e7e41ca
AV  - none
PB  - Society of Petroleum Engineers
N2  - A comprehensive description of reservoir fluid properties is critical in developing solutions and resolving reservoir engineering issues. The oil formation volume factor, βo, is an indispensable reservoir fluid property in reservoir engineering calculations. In this study, we used a total of 11040 data points from 1840 oil samples to develop new βo correlations for the Nigerian crude oils specifically, and another set of correlations for the other regions herein referred to as the global crude oils. Linear regression (LR), multiple linear regression (MLR), multiple non-linear regression (MNLR), neural network (NN), support vector machine (SVM), and the group method of data handling (GMDH) techniques were used to develop several correlations. Results show that the GMDH method yielded the best correlation while the MNLR is the least accurate. The root means square error (RMSE) for the Nigerian, and Global correlations are 0.0033, and 0.0256 respectively. The two correlations are reliably better in terms of accuracy than the existing correlations. The new correlations would facilitate a more accurate reservoir characterization, and reliable design of surface equipment. © 2022, Society of Petroleum Engineers.
ER  -