The Application of BP Neural Network in Settlement Prediction of Highway Soft Foundation

Abstract:

In view of the settlement of highway soft foundation, the paper proposes a method to predict soft foundation settlement based on BP neural network model, taking advantage of the strong non-linear mapping and learning ability of BP neural network. Then it is compared with the three-point method, obtaining some useful conclusions. Since the BP neural network model is directly based on real samples, it could avoid the mistakes due to factitiousness in the three-point method. It is proved that the BP neural network model is accurate and the settlement has least error.

• Periodical: Advanced Materials Research (Volumes 250-253)
• Pages: 3440-3443
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.250-253.3440
• Online since: May 2011
• Authors: Yi Xue, Zheng Zheng Cao, Shan Liu
• Keywords: BP Neural Network (BPNN), Highway, Settlement, Soft Foundation

References

[1] People's Republic of industry standards: Highway Embankment Design and Construction Specifications (China Communications Press, China 1997).

[2] Xvesen Zhen: Calculation of Soft Ground (China Communications Press, China 1998).

[3] Peiyi Lu: Soil Mechanics (China Building Materials Industry Press, China 2000).

[4] Yongjian Liu: Artificial Neural Network with the Final Settlement of Soft Highway (Highway and Transportation Research, China 2000).

[5] Fan Li, Zenglai Zhou, Mei Wu: Wavelet neural network with the final settlement of soft ground expressway (Hefei University of Technology, China 2001).