Article: Efficient prediction for porosity only using logging data: a case study of lacustrine carbonate reservoirs of IARA oilfield Journal: International Journal of Oil, Gas and Coal Technology (IJOGCT) 2020 Vol.25 No.2 pp.133 - 160 Abstract: Statistical methods are capable of revealing variation trends of study cases by only processing source data. N-way analysis of variance and multivariate linear fitting are a few excellent examples. Then in order to cast off the reliance on using physical model parameters, a statistical method combined N-way analysis of variance and multivariate linear fitting is proposed. This article discloses the function of proposed method in terms of porosity prediction when only logging data is available. The data used for verification derives from the lacustrine carbonate reservoirs of IARA oilfield. Five experiments are well-designed to analyse capability of the method on porosity prediction. Experiment results prove that the proposed method can only utilise logs to predict porosity, and the predicted outcomes are accurate. The proposed method can be viewed as a new tool to obtain porosity data when only logs are handled. [Received: March 8, 2018; Accepted: February 14, 2019] Inderscience Publishers - linking academia, business and industry through research

Title: Efficient prediction for porosity only using logging data: a case study of lacustrine carbonate reservoirs of IARA oilfield

Authors: Yufeng Gu; Zhidong Bao; Zhenhua Rui; Xinmin Song

Addresses: College of Geosciences, China University of Petroleum-Beijing, Beijing, China; PetroChina Research Institute of Petroleum Exploration and Development, Xuefu Road, Beijing, China ' College of Geosciences, China University of Petroleum-Beijing, Beijing, China ' University of Alaska Fairbanks, South Chandalar Drive, Fairbanks, Alaska, USA ' PetroChina Research Institute of Petroleum Exploration and Development, Xuefu Road, Beijing, China

Abstract: Statistical methods are capable of revealing variation trends of study cases by only processing source data. N-way analysis of variance and multivariate linear fitting are a few excellent examples. Then in order to cast off the reliance on using physical model parameters, a statistical method combined N-way analysis of variance and multivariate linear fitting is proposed. This article discloses the function of proposed method in terms of porosity prediction when only logging data is available. The data used for verification derives from the lacustrine carbonate reservoirs of IARA oilfield. Five experiments are well-designed to analyse capability of the method on porosity prediction. Experiment results prove that the proposed method can only utilise logs to predict porosity, and the predicted outcomes are accurate. The proposed method can be viewed as a new tool to obtain porosity data when only logs are handled. [Received: March 8, 2018; Accepted: February 14, 2019]

Keywords: logging interpretation; porosity prediction; N-way analysis of variance; multivariate linear fitting; fitting correction.

DOI: 10.1504/IJOGCT.2020.109446

International Journal of Oil, Gas and Coal Technology, 2020 Vol.25 No.2, pp.133 - 160

Received: 08 Mar 2018
Accepted: 14 Feb 2019
Published online: 09 Sep 2020 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article

Title: Efficient prediction for porosity only using logging data: a case study of lacustrine carbonate reservoirs of IARA oilfield

Keep up-to-date