Estimation of oil saturation via pseudo capillary pressure curve from nuclear magnetic resonance log data in tight conglomerate reservoirs

Tight conglomerate reservoirs have complex pore structure and strong heterogeneity which could bring great difficulties in the identification of oil and water layers. In order to solve this problem, the tight conglomerate reservoirs of Triassic Baikouquan Formation and Permian Urho Formation in Mahu Depression, Northwest Junggar Basin in China are selected as the study area. Firstly, 23 representative core plunger samples were selected from 9 wells. The experimental data of helium porosity, permeability, mercury intrusion capillary pressure (MICP) curve, and nuclear magnetic resonance (NMR) T2 spectrum under completely watered condition were measured and analyzed. Secondly, conglomerate reservoir classification criterion was built based on flow zone indicator (FZI). Its classification results conform to the classification of J function and core images. Thirdly, the piecewise power function was adopted to predict the pseudo capillary pressure curve by NMR T2 spectrum and the corresponding models were established via reservoir classification. Fourthly, based on the obtained pseudo capillary pressure curve and average J function curve, a novel method for predicting oil saturation by the ratio of median radius was proposed. Finally, the built models were used to process the experimental data not involved in modeling and log data. The predicting results are well consistent with the experimental results, which indicates the high reliability of the models. The established models are essential for helping the classification of tight conglomerate reservoirs, calculation of reservoir parameters, and identification of oil and water layers.