Neural Network Analog on Dynamic Variation of the Karst Water and the Prediction for Spewing Tendency of Springs in Jinan

Abstract

Considering the factors that affect the karst water level, the improved Neural Network Model has been applied to construct the random model that analogs the dynamic change of karst water. The accuracy of our analog has been greatly improved compared with that of multi-line recurrence model; moreover, BP model has strong functions of study, fault tolerance and association. In a word, BP model is an effective tool to predict the dynamic change of karst water. In addition, the spewing tendency of springs in Jinan is analyzed based on our prediction results in this paper.