Application of BP neural network model in water environmental carrying capacity research of Xiangshan Bay

Abstract:

In order to study the water environmental carrying capacity (WECC) of Xiangshan Bay in recent years, the thresholds of the water quality parameters DO, COD, DIN and DIP were obtained according to the statistics of Xiangshan Bay from 2010 to 2013. Then BP neural network technology was applied to establish a WECC model of Xiangshan Bay. The input of the model are the monitoring data of DO, COD, DIN and DIP. The output of the model was the Water Environmental Carrying Capacity Index (WECCI). The model was applied in the study of the WECC of Xiangshan Bay in the four seasons of 2014. The results show that:the seasonal-averaged WECCI of Xiangshan Bay in 2014 is always bellow 0.4, so the WECC of Xiangshan Bay is not ideal; The WECC of Xiangshan Bay is higher in the inshore than offshore area; The WECC is low all through the year in the bay mouth, influenced by the offshore water; The seasonal variation of WECC is complicated in the inner bay, and it is mainly locally influenced; The WECC in the central bay is high in spring, and low in summer, which is influenced by biological activity; The structure of BP neural network is simple, and the results are intuitive and reliable. Therefore, BP neural network could be used in the study of the WECC of Xiangshan Bay.