Salinity Prediction Model in Estuary Area Based on Neural Network

The salinity problem in the estuary area is related to industrial, agricultural and domestic water use, and is an important indicator of the water environment in the estuary area. In order to improve the prediction accuracy of salinity in the estuary area, based on the collection of a large amount of actual measured data and the neural network method, taking the daily maximum chlorine concentration at Qibao Station in the Qiantang River Estuary as an example, the estuary area salinity prediction model was established and the algorithm was optimized. The results show that it is feasible to predict the salinity in the estuary area based on the neural network model. As far as the prediction effect is concerned, the segmented multi-model BP neural network is the best, followed by the Narx neural network. The GA-BP neural network is slightly improved compared to the BP neural network, but the improvement effect is not as good as the segmented multi-model BP neural network.