With the successful launch of FengYun-3, more satellite observations of radiances are directly assimilated. The satellite microwave all-sky assimilation technology can make better use of the microwave observation data in clear, cloudy and precipitating conditions. It can not only increase assimilated radiances, but also show positive impacts on numerical weather prediction. As a result, the all-sky approach gets a lot of attention now. In this study, a typhoon case called Maria in July 2018 is selected and all-sky assimilation of FY-3C/MWHS-2 data were used in Weather Research and Forecasting Model, based on the 3-dimensional variational data assimilation technique within the WRFDA (WRF data assimilation) system, in order to study the impacts of assimilating FY-3C/MWHS-2 data under the all-sky conditions and using different data types of the initial fields on the forecasts in the regional numerical model. The comparison of the clear-sky and all-sky assimilation experiments shows that under the all-sky conditions, more clouds and precipitation data are assimilated and the data usage is increased. Improvements are displayed in the analysis of the vertical structures of Maria, such as the warm core and the symmetric wind speed structures, and the humidity fields. The average reduction rates of the track error are about 34% to 62%. These positive impacts can be demonstrated in all the all-sky experiments with different data types for initializing WRF.