Abstract: Fengyun-4B(FY-4B) is the latest generation of China geostationary meteorological satellite, and it has higher spatial and temporal resolution. In order to further promote the application of FY-4B atmospheric motion vector(AMV) in numerical models, this study compares FY-4B AMV data with FY-4A AMV data. Firstly, the study analyzes the vertical distribution of AMV observational error and applies it to the Weather Research and Forecasting Model Data Assimilation System (WRFDA) instead of the default observational error. Then, the impact of AMV data on the assimilation and forecast of Typhoon Muifa in 2022 is investigated. The analysis indicates that the observational error of the U and V components of the FY-4B AMV data are smaller than those of the FY-4A AMV data. Additionally, after assimilating the FY- 4A AMV data, the simulated typhoon gradually moves westward. Owing to the better information on the wind field after assimilating the FY-4B AMV data, the forecast track of the typhoon is closer to the observation track. Furthermore, the assimilation of the FY-4B AMV data improves better the precipitation forecasts at a certain extent than FY-4A AMV.