Oil–water relative permeability is a critical parameter for reservoir fluid property evaluation and productivity prediction, playing a vital role in reservoir identification and development. However, the correlation between relative permeability and resistivity remains unclear for complex conductive mineral-bearing argillaceous sandstone reservoirs. In this study, twenty-eight artificial core samples were prepared and compacted under simulated reservoir temperature and pressure conditions. Simultaneous oil–water relative permeability and resistivity measurements were performed on these samples at two brine salinities (7000 ppm and 3000 ppm). The analysis reveals that under the same resistivity index, both conductive-mineral content and shale content exhibit a negative correlation with water-phase relative permeability but a positive correlation with oil-phase relative permeability. In contrast, porosity–permeability conditions and injected-water salinity show a positive correlation with water-phase relative permeability and a negative correlation with oil-phase relative permeability. Furthermore, the effect of salinity on the relative permeability–resistivity relationship intensifies as porosity–permeability conditions improve. Based on the consistency between movable water seepage and conductive paths, a novel relative permeability–resistivity model was developed for the target reservoirs by integrating seepage and conductive theories, with a generalized Archie’s equation adopted to calibrate the effects of conductive minerals and shale on rock conductivity. The model parameters are optimized using the Newton–Raphson iteration algorithm, and empirical correlations between the undetermined coefficients and reservoir parameters are developed. Validated against two independent cores with the most complex mineral compositions (SH2-3 and SH3-3), the model yields average relative errors of 9.8% and 13.4% for water-phase relative permeability, and 4.4% and 5.4% for oil-phase relative permeability, respectively, demonstrating its high calculation accuracy. It enables efficient prediction of reservoir relative permeability using resistivity data, providing a new theoretical and technical support for low-resistivity oil reservoir identification and productivity assessment.

In high-temperature and high-salinity reservoirs, HPAM is prone to thermal degradation, salt-induced chain shrinkage and shear instability, which weakens its mobility-control ability. In view of the above problems, this paper prepared bifunctional modified nano-SiO₂ (KH570@SiO₂-AMPS/ODMA, DF-SiO₂), and constructed a composite gel system with HPAM. Its performance was evaluated by structural and rheological characterization, aging and shear recovery, and heterogeneous core displacement. At 90 °C and about 20,000 mg/L salinity, the viscosity retention rate exceeded 70% after 90 days of aging, and recovered to 94.6% of the initial value after 72 h of standing following shearing. Core displacement shows that the system can improve the flow distribution of heterogeneous reservoirs, and the recovery of high and low permeability layers is 6.95 and 4.41 percentage points higher than that of HPAM, respectively. DF-SiO₂ constructs a dynamic network through interfacial interaction and hydrophobic association to improve the stability and structural recovery ability of the system, which provides a new idea for the application of high-temperature and high-salinity heterogeneous reservoirs.