Abstract: This study applies Mann-Kendall trend test method to analyze the precipitation change characteristics and trend over years at Huangtaiqiao station, and predicts annual precipitation using univariate artificial neural network nonlinear time series model. It is discovered that the annual precipitation of Huangtaiqiao has a slight trend of increase (1.4 mm/10 a), with increasing in spring and summer and decreasing in fall and winter. The precipitations in 2018 and 2019 are predicted by artificial neural network nonlinear time series model, it is discovered that the value is higher than the annual average precipitation of huangtaiqiao station.