A saturation evaluation method in tight gas sandstones based on diagenetic facies

Water saturation is one of the key parameters in reservoir evaluation and calculation. Tight sandstone reservoirs with complex pore structures caused by strong diagenesis often have complex rock conductivity and response laws, resulting in the low accuracy of the water saturation values calculated using existing models. Taking a Permian block in the eastern margin of the Ordos Basin as an example, the core samples of five wells were analysed using X-ray diffraction, scanning electron microscopy, core mercury intrusion experiments, analyses of their physical properties, and rock and electric experiments. First, research on the diagenetic fades classification method was carried out. Second, the conduction mechanisms of different diagenetic facies were studied, and a saturation model was established. Finally, Deep AutoEncoder was used to automatically abstract and express the response characteristics of different diagenetic facies. Random forest, which is sensitive to small sample sizes, was used to establish the logging criteria of different diagenetic fades, and the longitudinal parameters of the saturation parameters of the exploration wells were obtained. Studies have shown that the division schemes of diagenetic facies based on diagenesis, pore types, pore structures and damage to or construction of the reservoir are reasonable and effective. The rock electrical parameters (m, n) in the study area are controlled by the diagenetic facies: the cementation index m of various diagenetic facies is mainly controlled by the pore structure, and the cementation index m and porosity of the three types of diagenetic facies fall on different trend lines. The saturation exponent n is mainly affected by the pore structure and clay; as the pore structure deteriorates, the clay content decreases, and the n value increases. The saturation model based on diagenetic facies was applied in the saturation log evaluation of the study block, and the predicted results were compared with those of five other saturation models suitable for different reservoir types. The saturation evaluation error was reduced from 40.28% to 12.4%. The reliability of the saturation model proposed in this paper was verified, and a new classification scheme is provided for the evaluation of the saturation of tight sandstone, which can assist in the evaluation of the saturation of tight sandstone reservoirs.