An Back Propagation (BP) neural network model for water quality evaluation was established to overcome the shortcomings of the traditional methods. Levenberg-Marquardt (LM) optimization algorithm was used to make the BP neural network converging quickly, and golden section theory was used to get the reasonable number of the network's hidden nodes. The optimal BP network model was used to evaluate the water quality degree of Xindu area of Chengdu, which results were compared with the evaluation results of traditional methods: comprehensive pollution evaluation method and single factor method, it was proved that the results of the set BP network model are more objective and steady, and it makes it possible to compare the water quality of two rivers which belong to different function grades.

Date Published: 20 March 2013
PDF: 5 pages
Proc. SPIE 8768, International Conference on Graphic and Image Processing (ICGIP 2012), 87683J (20 March 2013); doi: 10.1117/12.2011063

Published in SPIE Proceedings Vol. 8768:
International Conference on Graphic and Image Processing (ICGIP 2012)
Zeng Zhu, Editor(s)