Quality control （QC） methods for Automatic Weather Stations （AWS） precipitation data are combined to design a fast and realtime QC Algorithm. Abnormal precipitation data are classified firstly. The Grubbs test is then used to screen out suspicious stations in a local and finite space preliminarily. Three factors are considered in the second step of the QC procedure, which are the spatial distribution of precipitation per hour, the timehistory equation of precipitation per minute, and the related characteristics between temperature and humidity. Practical test shows that the integrated quality control algorithm has a better balance between performance and accuracy, which is especially effective in summer.