Prediction of Ground Water Level Based on DE-BP Neutral Network

摘要

The continuous decline of ground water level is one of the important factors that affect development of national economy and society. Based on the DE-BP (back propagation-differential evolution) neutral network, the predicting model of ground water level is presented. The precision of the model is checked using the monitoring data in Zhangjiakou area. The comparisons between the predicted results of the three models (BP model, GA-BP model and DE-BP model) and the monitoring data show that the precision of the present algorithm is high with the maximum relative error being 0.17%.