ANALYZING OF GROUNDWATER LEVEL FLUCTUATION USING ARTIFICIAL NEURAL NETWORK WITH DIFFERENT TRAINING ALGORITHMS

ABSTRACT

The estimation of groundwater levels in a basin is a very important factor for planning integrated management of groundwater and surface water resources. In this study, artificial neural network models have been developed for predicting and forecasting of groundwater level. The reliability of the computational models was analyzed based on simulation results and using three statistical tests including Pearson correlation coefficient, coefficient of determination and root-mean-square error. The artificial neural network (ANN) with different training algorithms is applied for prediction of groundwater fluctuation. The process was implemented for six input combinations in order to find the most optimal input combination for groundwater fluctuation prediction. As the performance evaluation criteria of the ANN models the following statistical indicators were used: the root mean squared error (RMSE), Pearson correlation coefficient (r) and coefficient of determination (R2).

Keywords: groundwater fluctuation; artificial neural network; estimation