Simulation of non-point source COD pollution load by BP neural network

Author

Wang, Baoqing ; Ma, Qitao ; Sun, Yichao ; Liu, Honglei

Author_Institution

Coll. of Environ. Sci. & Eng., Nankai Univ., Tianjin, China

fYear

2011

fDate

24-26 June 2011

Firstpage

8488

Lastpage

8491

Abstract

By analysis of back propagation (BP) neutral network model structure and Tianjin Binhai new area non-point source COD pollution load (1997-2008), the paper established the BP neutral network of input layer neuron number 7, hidden layer neuron number 19, and output layer neuron number 1 to simulate COD pollution load. The result shows that mean error is 0.284% when the precision is 0.001 and hidden layer neuron number 19 for BP neural network. This BP neural network model has high accuracy. A simulation model of non-point source COD pollution load is formed automatically, so as to calculate the changing tendency of non-point COD pollution load.

Keywords

backpropagation; environmental science computing; geophysics computing; hydrological techniques; neural nets; rivers; simulation; water pollution; AD 1997 to 2008; BP neural network model structure; Binhai New Area; China; Tianjin; back propagation neutral network; chemical oxygen demand; nonpoint source COD pollution load simulation; Analytical models; Artificial neural networks; Joining processes; Load modeling; Neurons; Water pollution; BP neural network; COD pollution load; Tianjin Binhai new area; non-point source;

fLanguage

English

Publisher

ieee

Conference_Titel

Remote Sensing, Environment and Transportation Engineering (RSETE), 2011 International Conference on

Conference_Location

Nanjing

Print_ISBN

978-1-4244-9172-8

Type

conf

DOI

10.1109/RSETE.2011.5964140

Filename

5964140