基于改进SPH数值方法的强透水性砂层中流砂现象研究  

Study on Sand Flow Phenomenon in Strongly Permeable Sand Layer Based on Improved SPH Numerical Method

作　　者：吴丽萍[1] 陈弢锋 Wu Liping;Chen Taofeng(Changchun Engineering College,Changchun,China)

机构地区：[1]长春工程学院,吉林长春

出　　处：《科学技术创新》2025年第9期21-24,共4页Scientific and Technological Innovation

摘　　要：在深基坑支护施工过程中,饱和动态含水砂层松散,黏聚力小,自稳能力很差,极易发生流砂引发土体空腔造成塌方,由于砂土颗粒组成的复杂性,其力学特性十分复杂,因此提出了一种基于平滑粒子流体力学(SPH)与卷积神经网络(CNN)相结合的数值模拟方法,为研究流砂现象提供了一种有效的方法。In the process of deep foundation pit support construction,the saturated dynamic water-bearing sand layer is loose,with small cohesive force and poor self-stabilisation ability,which is very likely to occur in quicksand triggering the soil cavity causing collapse,and due to the complexity of the composition of the sand and soil particles,its mechanical properties are very complex,so a numerical simulation method based on the combination of smoothed particle hydrodynamics(SPH)and convolutional neural network(CNN)is proposed,which offers an an effective method.

关 键 词：砂层 神经网络 SPH 桩 数值模拟 

分 类 号：TU476.3[建筑科学—结构工程]