Retraction Note: Source discrimination of mine water inrush based on Elman neural network globally optimized by genetic algorithm

To identify water inrush sources quickly and effectively, six hydrated ions, namely, Na++ K+, Ca2+, Mg2+, Cl-, SO42-, and HCO3-, were selected as discriminant indices. Considering the unobvious and fuzzy boundaries of water quality between aquifers, Elman neural network (ENN) was chosen as the basis of the discrimination model; in order to overcome the error backpropagation algorithm of ENN, genetic algorithm (GA) was selected to optimize ENN through initialization, fitness evaluation, selection, crossover, and mutation, creating an improved discrimination model called GA-ENN. On the basis of giving consideration to the principle of fairness, GA-ENN was compared with ENN, backpropagation neural network (BPNN), and the BPNN optimized by GA (GA-BPNN). The results shows that GA-ENN outperformed the other three models in convergence speed and output accuracy and GA-ENN can reflect the dynamic changes of the water quality in the aquifer through years of mining. Apart from the hydrogeological conditions of the coalmine, representative and accurate hydration data should be selected to improve the accuracy and validity of source discrimination of mine water inrush.