初始组构影响砂土液化势的细观数值模拟

Discrete element modeling on the effect of initial fabric on liquefaction potential of sands

基于PFC2D椭圆颗粒单元的二次开发,通过制备3种不同初始颗粒定向的数值试样,分别为随机定向(R试样)、水平定向(H试样)和竖直定向(V试样),研究了初始组构对砂土液化势的影响。数值试验采用常体积(面积)循环等应变幅加荷条件,试验过程在对比分析3种不同颗粒定向试样液化宏观力学响应的同时,从颗粒尺度层面探讨了不同颗粒定向试样的初始组构特征,揭示了初始组构影响砂土液化势的细观力学机理。结果表明,颗粒规则定向的试样其抗液化强度要高于随机定向试样,在细观机理上与不同颗粒定向试样的初始平均接触数有关,砂土的液化过程是一个组构各向异性的演化过程。

The main objective of this paper is to investigate the effect of initial fabric on liquefaction potential of granular materials by discrete element simulations.Three cases of initial arrangement of particlesw ere considered：（1）the particles created with a random distribution of the major axis orientation,namely R samples;（2）the particles created with their major axes parallel to the horizontal direction,namely H samples;and（3）the particles created with their major axes parallel to the vertical direction,namely V samples.The strain-controlled undrained（constant volume）cyclic biaxial shear tests were carried out to explore the liquefaction behavior of numerical samples with different initial particle orientations.It was found that the samples with regular distributions of major axis orientation have a higher liquefaction resistance than those with random distributions,and the micro mechanism of this phenomenon is related to the initial averaged co-ordination numbers of different samples.Numerical samples significantly exhibit stress-induced fabric anisotropy during liquefaction regardless of their initial particle arrangements.