Abstract：The Tibetan Railway spans more than 550 kilometers of permafrost regions, and the key to the safe operation of the railway depends on the stability of the permafrost layer. Therefore, this paper selects Sentinel-1A ascending and descending orbits data along the railway permafrost section from October 2022 to December 2024, uses the SBAS-InSAR technique to invert deformation characteristics and analyze the driving factors; combines the advantages of multiple models to build a prediction model and conducts an analysis. The results show: ① The LOS deformation rates along the railway for ascending and descending orbits are -35~ 42 mm/a and -36.5~30 mm/a, respectively; neglecting north-south surface deformation, the average vertical and east-west rates are -1.62 mm/a and 2.31 mm/a, showing subsidence and eastward movement; ② Based on ‘thermal-hydrological-topographic’ coupling analysis, all factors are positively correlated with the deformation rate, with surface temperature and active layer thickness being the main controlling factors, contributing 62.4%; slope (21.4%) determines the direction of deformation; temperature (16.2%) indirectly drives deformation through surface energy conversion. Temperature and ground temperatures cause thickening of the active layer leading to thaw settlement, and under the influence of slope, this leads to slope-oriented creep displacement. ③ Using a layered threshold method, three levels of deformation intensity zones are identified, where high deformation zones threaten the stability of the railway foundation, medium deformation zones have permafrost creep risks, and low deformation zones are controlled by river effects. Feature point time series analysis reveals the typical seasonal frost heave and thaw settlement patterns in permafrost regions and the superimposed effects of hydrothermal deposits. ④ Prediction analysis of deformation trends for seven feature points, P1~P7, shows that the hybrid model outperforms single models in prediction accuracy at each feature point. The predicted deformation trends over the next six months for periodically deforming points closely match actual deformation, demonstrating good predictive performance at the selected feature points. The research findings confirm the potential of this method for monitoring ground deformation along railway lines in permafrost regions, and can serve as a reference for railway stability assessments and subgrade maintenance.