A Non-Empirical Fractal Permeability Model for EOR in Hydrate-Bearing Reservoirs: Coupling Effects of Effective Stress, Temperature, and Particle Heterogeneity

Permeability is a critical parameter for evaluating the production potential of natural gas hydrate reservoirs, and its accurate prediction is essential for enhanced oil recovery (EOR). However, existing permeability models often assume a uniform particle distribution, neglecting the inherent heterogeneity of natural sediments, and rarely fully couple the effects of effective stress and temperature variations induced by EOR operations. To address that gap, this study develops a novel non-empirical fractal permeability model that incorporates particle heterogeneity through an offset angle (θ) and an aspect ratio (m), and couples these with thermoelastic theory to describe the evolution of the pore structure under coupled thermo-mechanical conditions. The model accounts for two hydrate growth habits (grain-coating and pore-filling) and allows for their coexistence via weighting coefficients. Using this model, we systematically investigate the individual and combined effects of effective stress, temperature, particle heterogeneity, and hydrate saturation on permeability. Model predictions are validated against independent experimental data from multiple sources, showing good agreement. The results reveal that permeability decreases with increasing effective stress and temperature, with stress playing a more dominant role; moreover, the transition between hydrate growth habits under stress is captured. The proposed model provides a theoretical tool to understand permeability evolution in heterogeneous hydrate reservoirs under varying thermo-mechanical conditions, thereby supporting EOR strategy optimization.