PSO optimizing neural network for the Yangtze river sediment entering estuary prediction

Author

Guo, Wenxian ; Wang, Hongxiang

Author_Institution

North China Univ. of Water Resources & Electr. Power, Zhengzhou, China

Volume

4

fYear

2010

fDate

10-12 Aug. 2010

Firstpage

1769

Lastpage

1772

Abstract

The artificial neural network method is used to study the sediment entering the estuary prediction in the Yangtze River. Particle swarm optimization is applied to optimize the node numbers of the hidden layers in the ANN model and overcome the over-fitting problem. Datong hydrological station is the control station as the sediment entering the estuary. Based on the monitoring sediment load data of from 1956 to 2005 year, PSORBF neural network was applied to predict river sediment. The study indicates that the model is practical and has better prediction accuracy.

Keywords

environmental science computing; neural nets; particle swarm optimisation; rivers; sediments; Datong hydrological station; PSO; PSORBF neural network; Yangtze river sediment; artificial neural network method; estuary prediction; over-fitting problem; particle swarm optimzation; Artificial neural networks; Biological system modeling; Particle swarm optimization; Predictive models; Rivers; Sediments; Water resources; Particle Swarm Optimization; artificial neural network; sediment load; the Yangtze River;

fLanguage

English

Publisher

ieee

Conference_Titel

Natural Computation (ICNC), 2010 Sixth International Conference on

Conference_Location

Yantai, Shandong

Print_ISBN

978-1-4244-5958-2

Type

conf

DOI

10.1109/ICNC.2010.5584412

Filename

5584412