Abstract: The rapid intensification (RI) of tropical cyclones (TCs) has always been a challenge in operational forecasting. This study is based on the Tropical Regional Atmosphere Model for the South China Sea (CMA-TRAMS) and selects some typical cases of RI TCs, such as "HATO (2017)", "DOKSURI" (2017), "HIGOS" (2007), and "NESAT" (2220), to investigate the impacts of initial vortex and environmental fields on RI forecasts. Vortices in the initial fields with reported (unreported) RI simulations are replaced with vortices in the unreported (reported) RI simulations, and a series of numerical sensitivity experiments are conducted. The results show that the initial fields after vortex replacement retain the environmental characteristics of the background field, while also "inheriting" the main characteristics of different replacing vortices, and reflecting the structural differences of vortices at different stages, such as before RI, at the onset of RI, and during weakening. The results also indicate that replacing the initial vortex has no significant impact on the forecast of TC RIs, while the environmental field has a significant impact on RI forecast of TCs. Furthermore, it is also explored how the same environmental field leads to similar development trends for different initial vortices, and how different environmental fields cause different development for the same initial vortex. The environmental conditions facilitated intense heat transport in the initial stage of forecasting, leading to rapid adjustments in the vortex"s intensity and structure. This was a crucial factor for the RI of the vortex in Experiment V4B (using vortex in weakened phase as initial vortex), which was in the weakening phase. The insufficient heat and water vapor transport, coupled with the failure to form a favorable configuration with vertical wind shear, were the primary reasons why the RI of the vortex in Experiment C1 (the control experiment) did not occur as in Experiment V4 (employing an environment favorable for RI). The results of this study provide clues for improving model capability on RI prediction and lay a foundation for promoting operational forecasting level.