Share Email Print
cover

Proceedings Paper

Use of genetic-algorithm-optimized back propagation neural network and ordinary kriging for predicting the spatial distribution of groundwater quality parameter
Author(s): Jian Li; Zheng Zhang; Yu Feng; Xiumei Zhang
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

This study focuses on applying the model of Genetic Algorithm optimized Back Propagation Neural Networks (GABPNN) to formulate the forecast methodology of groundwater quality parameter with spatial data as parameters. Back Propagation neural network (BPNN) is the basic arithmetic of this method. Genetic Algorithm (GA) was used to optimize the weights and biases of BPNN. The GABPNN was firstly introduced into spatial prediction of groundwater quality parameter in Langfang city (China) and it was compared with ordinary Kriging(OK). The groundwater quality parameter such as chloride was selected for the study, the performance of the models was found to be very good. The results of the GABPNN model application were that the regional prediction map of the optimal GABPNN model could describe the spatial distribution situation of groundwater quality parameter and that the predictive validity of the GABPNN model was better than that of ordinary Kriging. The result shows that the GABPNN model is a reasonable and feasible method for spatial distribution of groundwater quality parameter in Langfang city (China).

Paper Details

Date Published: 4 March 2013
PDF: 8 pages
Proc. SPIE 8768, International Conference on Graphic and Image Processing (ICGIP 2012), 87684V (4 March 2013); doi: 10.1117/12.2011857
Show Author Affiliations
Jian Li, Tsinghua Univ. (China)
Zheng Zhang, Beijing Forestry Univ. (China)
Yu Feng, Beijing Forestry Univ. (China)
Xiumei Zhang, Beijing Forestry Univ. (China)


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

© SPIE. Terms of Use
Back to Top