A New SPT-Based Model for Liquefaction Probability Discrimination of Fine-Grained Sandy Soil
Fine-grained sandy soil is widespread in nature.When the standard penetration test(SPT)is used for liquefaction discrimination of sandy soil,the influence of fine grain content in sandy soil is re-flected by modifying the standard penetration blow count.Based on 230 sets of measured data collect-ed and collated from 25 major earthquakes worldwide,using three correction formulas for SPT blow counts proposed by Idriss and Boulanger,Cetin et al.and NCEER,this study established three lique-faction probability calculation models using logistic regression.A comparison of the three models showed that the one based on correction formula proposed by Cetin et al.demonstrated the best good-ness-of-fit and prediction accuracy.According to the conversion relationship between blow counts in the Chinese Code for Investigation of Geotechnical Engineering(GB 50021-2001)and Standard Test Method for Standard Penetration Test(SPT)and Split-Barrel Sampling of Soils(ASTM D1586-11),a new model for liquefaction probability discrimination of fine-grained sandy soil suitable for the Chinese Code was derived.Back-analysis indicated that the new model performed well in dis-criminating liquefaction data measured using the American Standard.Using 288 sets of data from the Chi-Chi earthquake in Taiwan,the new model was compared with the discrimination method from the Code for Seismic Design of Buildings(GB 50011-2010)(hereinafter referred to as"Building Code method").The results showed that the overall discrimination success rate of the new model(94.1%)exceeded that of the Building Code method(77.4%),with a success rate of liquefaction sites of 93.9%,slightly lower than that of the Building Code method(99.4%),but the success rate of non-liq-uefaction sites(86.3%)was significantly higher than that of the Building Code method(48.4%).
万方数据
维普
