Groundwater level prediction using ARMA-ANN model

At present, classic methods are often used to predict groundwater level, but the result is not ideal. Recent studies show that the combinational prediction methods have higher precision than single prediction methods. A combinational prediction model is presented based on ARMA and ANN neural network. And it is applied to comprehensive analysis and prediction of groundwater level. Case study indicates that precision of the model is rather high and its popularization significance is better than the other models, and has some practical value when being used in the dynamic groundwater level analysis.