Seepage and Thermal Conductivity Characteristics of Fractured Hydrate-Bearing Sediments

Abstract

Natural gas hydrate is a strategic energy source with great potential in the future. Elucidating the seepage and thermal conductivity characteristics of hydrate reservoirs is critical for evaluating economic acceptability. However, limited research considers the existence of fracture in natural and artificial hydrate reservoirs. In this study, a digital core generation method for fractured hydrate-bearing sediments (HBSs) based on CT images was proposed; 120 models with different crack openings, inclination angles, and growth ratios were reconstructed. The main findings were as follows: as fractures mature, the permeability of the reservoir will increase, and the effective thermal conductivity will greatly decrease. The increase in the fracture inclination angle will increase the permeability of HBS reservoirs and reduce the effective thermal conductivity. The inclination angle and growth ratio of fractures have the most significant impact on the permeability characteristic, and the growth ratio has the most significant impact on the thermal characteristic. In terms of predicting the permeability and thermal conductivity of fractured HBS, the back propagation and support vector machine are the relatively optimal neural networks compared with random forest and long−short-term memory.