Prediction of Spatial Distribution of Precipitation in Hebei Province Based on BP Neural Network and Multi-source Data
Based on the precipitation data of 78 standard meteorological stations in Hebei Province,the spatial distribution prediction model of precipitation based on terrain,sea and land location variables was established by BP neural network,and the spatial distribution characteristics of precipitation in Hebei Province were extracted.The results show that the precipitation observed by meteorological stations in Hebei Province ranges from 394.06 mm to 764.87 mm,which accords with normal distribution.The BP regression model can fit the nonlinear relationship between the covariable and the observed precipitation.The accuracy of the model R2 is 0.61,and the MAE and RMSE are 43.25 mm and 67.16 mm respectively.The forecast results show that the precipitation in Hebei Province is more in the south than in the north and less in the east than in the west,and the precipitation center is located in the southeast of Tangshan.The results can provide a basis for decision-making of regional protection and water resources utilization.
万方数据
维普
