Probabilistic evaluation of seismic liquefaction potential in field conditions A kinetic energy approach

摘要

Purpose - In this paper, the peak kinetic energy density (KED) of soil particles during earthquake excitation is used as an intensity measure for the evaluation of liquefaction potential under field conditions. The paper seeks to discuss this measure. Design/methodology/approach - Using centrifuge tests data, it is shown that seismic pore water pressure buildup is proportional to cumulative KED at a particular soil depth. Linear relationships are found between cumulative kinetic energy and corresponding cumulative strain energy. To consider the effect of soil amplification, several equivalent linear ground response analyses are performed and the results are used to derive an equation for depth reduction factor of peak kinetic energy density. Two separate databases of liquefaction case histories are used in order to validate the proposed model. The performance of the proposed model is compared with a number of commonly used shear stress-based liquefaction assessment methods. Finally, the logistic regression method is employed to obtain probabilistic boundary curves based on the present model. Parametric study of the proposed probabilistic model is carried out to verify its agreement with the previous methods. Findings - It has been shown that the kinetic energy model works satisfactorily in classifying liquefied and non-liquefied cases compared with the existing recommendations of shear stress-based criterion. The results of the probabilistic kinetic energy model are in good agreement with those of previous studies and show a reasonable trend with respect to the variations of fines content and effective overburden pressure. The proposed model can be as used an alternative approach for assessment of liquefaction potential. Originality/value - These findings make a sound basis for the development of a kinetic energy-based method for assessment of liquefaction potential.