论文部分内容阅读
引入Newton-PCG算法,编制了算法程序,建立了流–固耦合砂土模型,实现了饱和砂土边坡液化问题的有限元数值模拟。采用一次牛顿步,和多步预条件共轭梯度子迭代的Newton-PCG复合算法,编制了算法的FORTRAN语言程序。基于临界状态塑性模型框架,给出了饱和状态下砂土固–液耦合模型的本构关系,再现了某饱和砂土边坡液化的力学状态,探讨了水压力分布与流动特性。结果表明:Newton-PCG算法能更有效地进行大规模非线性求解;建立的模型能较好地反映砂土–流体耦合状态下的力学行为;多孔介质中由于流体的流动,孔隙水压力对砂土液化的力学行为影响较大,在局部剪切带即将形成的区域,水压力值明显增大,而且方向指向剪切带的外法线方向;根据液化发生准则,画出了边坡发生液化的滑动面,对于预测边坡的发生具有重要的现实意义。
The Newton-PCG algorithm is introduced and the algorithm program is compiled. The fluid-solid coupling sand model is established and the finite element numerical simulation of liquefaction of saturated sand slope is realized. A Newton-PCG composite algorithm with a Newtonian step and a multi-step preconditioned conjugate gradient sub-iteration was used to develop the FORTRAN language program. Based on the critical plastic model framework, the constitutive relation of the solid - liquid coupling model of saturated sand is presented. The liquefaction mechanics state of a saturated sand slope is reconstructed, and the distribution and flow characteristics of hydraulic pressure are discussed. The results show that the Newton-PCG algorithm can solve large-scale nonlinear problems more efficiently. The established model can better reflect the mechanical behavior under the condition of sand-fluid coupling. Due to the fluid flow and pore water pressure in porous media, Soil liquefaction has a great influence on the mechanics behavior. In the area where the local shear zone is to be formed, the water pressure increases obviously and the direction points to the outer normal of the shear zone. According to the liquefaction criterion, The sliding surface has important practical significance for predicting the occurrence of slope.