A Discrimination Method of Saturated Sand Liquefaction Possibility Based on Support Vector Machine

Abstract:

Based on the analysis of influence factors of saturated sand, this paper expounds the limitations of traditional evaluation of liquefaction, and introduces the criterion of support vector machine (SVM) based on the principle of structural risk minimization. According to the main influence factors of sand liquefaction, a SVM discriminant model of sand liquefaction with different kernel functions is established. Through studying small sample data, this model can establish nonlinear mapping relationship between influence factors and liquefaction type. On the basis of seismic data, a radial based kernel function is selected to predict sand liquefaction type. The research results show that the predicted magnitude is identical with the actual result, to prove that it is effective to apply this SVM model to evaluate the level of sand liquefaction.

• Periodical: Applied Mechanics and Materials (Volume 509)
• Pages: 38-43
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.509.38
• Online since: February 2014
• Authors: Zhong Jie Fan*, Yan Qiu Leng, Yong Long Xu, Zheng Jiang Meng, Ji Wei Xu
• Keywords: Earthquake, Evaluation of Liquefaction, Prediction Model, Saturated Sand, Support Vector Machine (SVM)

References

[1] Fang Zhi, Lu Hao-liang, Wang Long: An overview on Research of Dynamic Soil-Structure Interaction. Word Earthquake Engineering(2006). 22(1): 57-63.

[2] Chen Yue-qing, Lv Xi-lin: The revelation of Several Foundation Damaged Cases in Massive Earthquake. Engineering seismology(2001). 17(2): 8-15.

[3] Qian Jia-huan, Yin Zhong-ze. in: Earthwork Principle and Calculation, edtied by China water conservancy and hydropower Publising, Beijing(1996), in press.

[4] Ren Jin-gang, Wang Yu-fang: Overview of Saturated Sandy Soil Seismic Liquefaction Research Method. Haihe Water Resources(2006), pp.51-53.

[5] Li Fang-ming: Artificial Neural Network in The Sandy Soil Liquefaction Discrimination and Settlement of The Application of The Forecast. Journal of Nanjing industrial university(2005).

[6] HANNA. A. M, DERIN. U, SAYGILI. G: Neural Network Model for Liquefaction Potential in Soil Deposits Using Turkey and Taiwan Earthquake Data. Soil Dynamics and Earthquake Engineering(2007). 27( 6) : 521-540.

DOI: 10.1016/j.soildyn.2006.11.001

[7] Shi Xu-chao, Fan Liang, Han Yang: Analysis on Sand Seismic Liquefaction Based on Support Vector Machine. Journal of Henan University of Science and Technology (2004). 25(3): 74-77.

[8] Chen Rong-lin, Lin Jian-hua, Huang Qun-xian: Application of the support vector machine in prediction of sand liquefaction. The Chinese Journal of Geological Hazard and Control(2005) 16(2): 15-19.

[9] Xia Jian-zhong, Luo Zhan-you, Gong Xiao-nan: Support Vector Machine Model for Predicting Sand Liquefaction. Chinese Journal of Rock Mechanics and Engineering(2005). 24(22): 4139-4144.

[10] Cai Yu-dong, Gong Jia-wen, Yao Lin-sheng. Artificial Neural Network Model for Prediction of Liquefaction of Sandy Soil. Journal of geotechnical engineering(1993). 15(6): 53-58.