Extreme precipitation events have a significant impact on economic and social development and agricultural production. In order to analyze the statistical characteristics of extreme precipitation, this paper takes the 24-hour precipitation total data from CenGong County, Guizhou Province from 1960 to 2022 as the research object and constructs the generalized Pareto model. Firstly, the historical data of 24-hour precipitation total in CenGong County is preliminarily analyzed to argue that the distribution of precipitation is long-tailed and the precipitation process is stable. Secondly, based on the POT model, the MRL map and Hill map are used to determine the candidate threshold range, and the KS test is used to verify the fitting effect of the Generalized Pareto Distribution and select the optimal threshold. Then, the maximum likelihood estimation method is used to obtain the estimated values of the two-parameter Pareto distribution model parameters, and the exceedance distribution fitting diagram and residual Q-Q diagram are used to analyze the model fitting effect. Finally, by analyzing different return periods, the possibility of different precipitation intensity grades occurring and the range of precipitation amounts are demonstrated.