Existing evaluations of sandstone-hosted uranium deposits commonly describe ore-controlling troughs and faults qualitatively, which limits the quantitative characterization of structural ore-controlling elements and the prediction of favorable mineralization areas. This study proposes a relative geological time (RGT)-constrained method for quantitatively computing intra-stratal trough attributes and a 3D seismic-based workflow for trough–fault coupled structural favorability prediction in sandstone-hosted uranium exploration and applies this framework to the Qianjiadian uranium mining area. An interpreted-horizon-constrained RGT volume was constructed to extract stratal two-way travel time (TWT) slices, and multi-scale trough attributes were calculated from local TWT relief differences. Fault attributes were constructed from laterally smoothed fault-scan results to improve fault-response continuity. Known mineralized wells were used to calibrate trough and fault responses and to construct trough ore-controlling index (TOI) and fault ore-controlling index (FOI) volumes. The calibrated TOI and FOI volumes were then integrated to generate a coupled favorability index (CFI) volume for structural favorability prediction. High CFI values occur mainly near major fault zones and in areas with moderate trough responses, indicating favorable trough–fault coupled structural conditions for uranium mineralization. Blind-well validation shows that 15 of 20 blind wells agree with the CFI prediction, yielding an overall accuracy of 75.0% and an AUC of 0.85. The proposed method converts qualitative trough and fault ore-control knowledge into computable, mineralization-calibrated, and blind-well-validated 3D structural favorability results, supporting favorable-area selection and drilling deployment in sandstone-hosted uranium exploration.