Hydrocarbon saturation in shale oil reservoirs by inversion of dielectric dispersion logs. (15th April 2020)

• Record Type: Journal Article
• Authors: Zhao, Peiqiang; Fu, Jinhua; Shi, Yujiang; Li, Gaoren; Ostadhassan, Mehdi; Luo, Miao; Mao, Zhiqiang
• Is Part Of: Fuel. Volume 266(2020)
• Publication Date: 2020-04-15
• Subjects: Shale oil reservoirs -- Hydrocarbon saturation -- Dielectric dispersion logs -- Simulated annealing -- Ordos Basin; Fuel -- Periodicals; Coal -- Periodicals; Coal; Fuel; Periodicals; 662.6
• DOI: 10.1016/j.fuel.2019.116934
• Languages: English
• ISSNs: 0016-2361

Abstract

Abstract: Hydrocarbon saturation is an important petrophysical parameter in reservoir evaluation and reserve calculation. Dielectric dispersion logs have been widely used to estimate the hydrocarbon saturation, especially in high salinity shale formations. However, there is still room to improve the accuracy of interpretation of dielectric dispersion logs. In this paper, we used two interpretation models combined four dielectric permittivity and four conductivity logs measured by Array Dielectric Tool (ADT) at four frequencies to predict the hydrocarbon saturation in shale oil reservoirs, Triassic Yanchang Formation, Ordos Basin, China. The logs measured at the highest frequency are interpreted by the Complex Refractive Index (CRIM) model. And the other three frequencies measured logs are processed by the shaly sandstone (SHSD) model. Based on the two models, simulated annealing algorithm is selected to calculate water saturation and rock textual parameter, the dispersive phase fraction, and water salinity. Besides, accurate porosity, dielectric permittivity of rock matrix, and temperature are input information to the proposed method. The proposed method is applied to shale oil reservoirs of Triassic Yanchang Formation in two wells. The effectiveness and reliability of proposed model is verified by synthetic log responses. The mean relative error between synthetic and field logs is small. The predicted hydrocarbon saturation is consistent with core data, and its variation is consistent with characteristics of NMR T2 distributions, verifying the accuracy and reliability of proposed method and inversion results. Our estimated hydrocarbon saturation of the two wells is about 30%, showing a good oil generation potential of the studied shale oil reservoirs.