ارزیابی رویکرد کرنش چرخه ای برای تخمین رهاسازی روانگرایی Assessment of the cyclic strain approach for evaluating liquefaction triggering

1. Introduction The primary objective of the study presented herein is to evaluate the efficacy of the strain-based liquefaction triggering evaluation procedure implemented using a pragmatic variant of the procedure originally proposed by Dobry et al. [11]. Liquefaction is a phenomenon that results from the contractive tendencies of loose to medium dense soils when sheared. For saturated cohesionless soils, this tendency results in the transfer of the overburden stress to the pore fluid, with the commensurate increase in pore water pressure and decrease in effective confining stress. Liquefaction has occurred in most major earthquakes and has caused significant damage to infrastructure (e.g., Cubrinovski and Green [8]; Cubrinovski et al. [9]; Green et al. [15]; Olson et al. [33]; Stringer et al. [40]; among many others). The most widely used procedure for evaluating liquefaction triggering potential is the simplified stress-based procedure originally proposed by Whitman [44] and Seed and Idriss [37]. This procedure is semi-empirical and has undergone periodic updates as a result of findings from new laboratory studies and/or the collection and analysis of additional field case history data (e.g., Youd et al. [46]; Cetin et al. [6]; Idriss and Boulanger [16]). Inherent to this procedure is the quantification of the seismic demand imposed on the soil expressed in terms of cyclic shear stress. Despite the popularity of the stress-based procedures, multiple studies have shown that excess pore water pressure better correlates to cyclic strain than to cyclic stress (e.g., Fig. 1) (e.g., Martin et al. [25]; Dobry et al., [11]; Byrne [3]). The reason for this is the relative movement of soil particles, which is requisite for excess pore water pressure generation, relates to the induced strain, regardless of amplitude of the stress applied to soil. As a result, Dobry et al. [11] proposed a strain-based liquefaction triggering evaluation procedure. Although the Dobry et al. [11] procedure generally received a positive reception by liquefaction researchers, it has failed to be adopted into practice. One reason for this is likely the requirement to perform strain-controlled cyclic laboratory tests on undisturbed and/or reconstituted specimens. This is in contrast to the simplified stress-based procedures wherein in-situ test metrics are the primary parameters used to evaluate liquefaction potential, with laboratory index tests and grain size distribution analyses having supporting roles if their performance is deemed necessary (e.g., use of measured fines content, FC, versus apparent FC in conjunction with the Cone Penetration Test, CPT, stressbased simplified procedure). Herein an alternative approach to implementing the Dobry et al. [11] strain-based procedure is proposed which circumvents the need for performing strain-controlled cyclic laboratory tests. Per this procedure, a strain-based numerical excess pore pressure generation model is used in lieu of developing analogous relationships from laboratory tests. The soil parameters required to implement the procedure include: relative density (Dr), secant shear modulus (G), and grain size distribution characteristics of the soil (i.e., FC and coefficient of uniformity: Cu); note that focus herein is on soils that are susceptible to liquefaction (i.e., non-plastic soils) and thus Plasticity Index (PI) is not needed. These required parameters are not too different from those required to implement the stress-based simplified procedures and can be estimated using simple relationships or conservative assumptions.