Abstract: In order to predict mine water inrush disaster more accurately, provide help for water inrush prediction and rescue, and reduce the loss caused by flood, a mine water inrush prediction model based on GAPSO-RFR is proposed.Genetic particle swarm optimization algorithm(GAPSO) is used to optimize the random forest regression(RFR) model.34 samples are selected to iterate and train the GAPSO-RFR model, and the optimal parameters are obtained.The test results show the GAPSO-RFR model improved the prediction accuracy and reduced the generalization error.The model is used to predict the risk of water inrush in some mining areas of Wangjialing coalfield.The regional distribution of high risk of water inrush is gained.