A BackPropagation (BP) Model of Artificial Neural Network (ANN) is used for th e partitioning of radar reflectivity into convective and stratiformcloud preci pitation classifications with the CINRADSA data from 2001 to 2003 in Hefei. Th e trained ANN is applied in a precipitation process. It is proved that the singl e hidelayer BP model of ANN can be used to classify the different precipitatio n echoes with a high successrate. It is also validated that: the successrate is influenced by following factors: the amount and the inputorder of the tr ainingdatabase, the nerve cell number of the hidedlayer and the choice of th e learning rate.