Artificial Neural Network and Its Application in Regime Prediction of Groundwater Quality
Guo Qing Zhong · Journal of Eastchina Normal University · 2001
According to previous study, the variety of groundwater quality is mostly influenced by human activities of exploitation and reinjection in Shanghai. The correlation between these two factors is nonlinear, and, artificial neural network has the ability to solve this problem. Therefore, based on collected data of groundwater quality, quantity of exploitation and reinjection,this article predicts the variety of salinity in III confined aquifer in Shanghai area using artificial neural network. The results indicate that the relative error is only 2.07% in single well prediction, and less 1.5% in multi-well prediction.