A new empirical model for enhancing well log permeability prediction, using nonlinear regression method: Case study from Hassi-Berkine oil field reservoir – Algeria

• Main Authors: H.E. Belhouchet, M.S. Benzagouta, A. Dobbi, A. Alquraishi, J. Duplay
• Format: Article
• Language: English
• Published: Elsevier 2021-02-01
• Series: Journal of King Saud University: Engineering Sciences
• Subjects: Permeability; Reservoir characterization; Rock typing; Hydraulic unite; Flow zone indicator
• Online Access: http://www.sciencedirect.com/science/article/pii/S1018363920302270

The reservoir permeability (K) factor is the key parameter for reservoir characterization. This parameter is considered as a determinant reservoir quality index. Depending on the data required and procedure availability, permeability can be defined from several methods such as; well test interpretation, wireline formation tester, and core data. These approaches can also be in assumption with permeability prediction targeting the non-cored sections. According to a similar status, well logs records can be an interesting support tool in use to reach the planned objectives. Thus, this investigation consists of finding out a model able to estimate the well log permeability and adjusting the outcome to the core permeability results.In this led research, the applied approach to the core data, to start with, was aimed to determine the reservoir rock types (RRT) using the flow zone indicator (FZI) method. The obtained classification allows stating a permeability model for each rock type.In order to calculate permeability from well logs, FZI has been founded out. A multi-regression technique was used to analyze the relationship of FZI with respect to specific logs such as Gamma-ray (GR), Density Log (RHOB), and Sonic log (DT). An objective function has been designated to minimize the quadratic error between the observed normalized FZI coming from core data, and the normalized FZI calculated from well logs. This process is carried out to identify a mathematical correlation allowing the estimation of FZI from porosity logs, leading to permeability determination. As results, permeability from logs was supporting relatively permeability defined from cores. The final results can be an accurate and real test for associating the exactitude performance of logging data records in boreholes with respect to the overall reservoir characterization sections. Thus, the applied investigation can be a genuine and quick method for essentially a specific deduction regarding the non-cored reservoir sections, with reference to rock typing, permeability and probably further reservoir factors.