Large amount of satellite observations have been used in numerical weather prediction and improving the forecasts greatly. Most works focus on the case study for influence of microwave satellite data on regional numerical weather prediction, while the study for long time is very little. The motivation of this paper is to improve the utilization of microwave satellite data in assimilation system and validating the influence of satellite microwave data on regional numerical prediction. In this paper, the study time of numerical prediction is from 1st to 31st August, 2008. Rapid Update Cycle (RUC) assimilation system of microwave satellite data is designed and constructed, which is based on the regional mesoscale model WRF and WRF 3DVAR system. Therefore the improvement effect of RUC assimilation system on the regional numerical prediction can be analyzed. The results show that, the RUC assimilation scheme is better than single time assimilation scheme. Compared with the former, the RUC assimilation scheme can improve the correlation coefficient of every forecast element to a certain extent. The correlation coefficients of wind field, temperature field and potential height field on the bottom level are improved remarkably. Except the 400 hPa humidity field correlation coefficient is reduced, other various levels of humidity field correlation coefficients are increased, especially in the 700 hPa and 200 hPa, the humidity field increased significantly. RUC assimilation scheme can also reduce the average square errors on the most levels. The forecast effect of microwave satellite data application on land and sea area is consistent with the characteristics; There are slightly differences in detail. On the whole, the improvement on sea area is better than the land surface area, because the surface effect is more complex. Furthermore, in the case of heavy rain and typhoon, the analyses reveal that, RUC assimilation of satellite data scheme can improve the simulation of the precipitation and typhoon track, and can simulate the attribution of heavy rainfall better, especially to improve the strong precipitation center in the simulation, and can also improve the typhoon path prediction to some extent, especially the path trend after the typhoon landing. Overall, it shows that on regional numerical weather prediction, RUC assimilation system can fully use the assimilation of microwave satellite data, and is a useful and necessary method to improve the forecast accuracy.