SALINE TIDE PREDICTION MODEL STUDY OF THE PEARL RIVER BAYOU BASED ON BP NEURAL NETWORK

摘要

Accurate Prediction of saline tide is the strong technical support for making and implementing the emergency plans of preventing saline tide, and for water resources management of the Pearl River basin. So, the study of the forecast model for Pearl River Bayou is very important. In this paper, the BP artificial neural network(ANN) is applied to study the saline tide forecast model of the Pearl River Bayou. The adoptive function is sigmoid function which is of strong non-linear mapping advantage. This model is able to simulate and forecast the complicate mechanism of saline tide, to identify the movement principle of saline tide and forecast the process of the saline tide along the Pearl River Delta. Experiments show that the forecast process of the saline tide by the ANN model is very well.