High-precision calculation of gas saturation in organic shale pores using an intelligent fusion algorithm and a multi-mineral model

Shale gas reservoirs have been the subject of intensifying research in recent years. In particular, gas saturation has received considerable attention as a key parameter reflecting the gas-bearing properties of reservoirs. However, no mature model exists for calculating the saturation of shale gas reservoirs due to the difficulty in calculating the gas saturation. This paper proposes a new gas saturation prediction method that combines model-driven and data-driven approaches. A multi-mineral petrophysical model is applied to derive the apparent saturation model. Using the calculated apparent saturation, matrix parameters and porosity curve as inputs, an intelligent fusion algorithm composed of five regression algorithms is employed to predict the gas saturation. The gas saturation prediction results in the Yongchuan block, Sichuan Basin, reveal that the model proposed in this paper boasts good reliability and a greatly improved prediction accuracy. The proposed model can greatly assist in calculating the gas saturation of shale gas reservoirs.