周鹏 王霜 王恺

摘 要：利用有限元分析软件模拟双层堤基、三层堤基以及多层堤基内部的水力梯度分布，将堤基内部各区域的水力梯度与相应的允许水力梯度进行比较，最终确定堤基管涌发生的临界上游水头。三层堤基和多层堤基中，堤基内部存在细砂夹层，随着细砂夹层埋深的增加，管涌发生的临界上游水头逐渐增大，通道深度不断增加，管涌破坏发生的标志是细砂夹层被冲破，发生竖向流土破坏;当细砂夹层埋深达到一定深度时，细砂上部砂砾石层的渗水量增加，增加到足以将砂砾石表面的颗粒带出，管涌破坏过程与双层堤基类似，通道主要形成在黏土下部的砂砾石表面，临界水头有所减小。将该数值模拟结果与前人管涌试验结果进行比较，两者吻合度较好。

关键词：堤基结构;管涌;数值模拟;临界水头

Abstract：A finite element software was used to simulate the distribution of hydraulic gradient in two-stratum， three-stratum and multi-stratum foundations. The hydraulic gradient in different areas was compared to the critical hydraulic gradient of soil erosion in the corresponding area and then the critical hydraulic head of piping in the dike foundation was determined. In three-stratum and multi-stratum foundations， the critical hydraulic head gradually increased with the burial depth of the fine sand layer and the depth of the piping passage was gradually increased. Piping happened when the sand layer broke through with the vertical soil flow. When the burial depth of the fine sand layer came to a certain value， the piping passage formed on the surface of the sandy gravel layer below the clay layer， which was similar with that in two-stratum foundation. The critical hydraulic head decreased. Comparisons between the simulation results and the experimental value were conducted and the results showed that the goodness of fit was high.

Key words： structure of dike foundation; piping; numerical modeling; critical hydraulic head

历史上的河流冲刷、淤积和改道，以及沉积条件的千变万化，导致不同时期、不同河段所形成的堤基结构有很大差异。堤基大多由黏土、粉质黏土、壤土、淤泥质土、砂壤土、粉细砂和砂砾石等组成，有多种组合类型，大体上可以概括为单层堤基、双层堤基、三层堤基和多层堤基4种类型[1-3]。对于不同结构的堤基，渗流场也是不同的，应该对具体条件下的渗流场进行分析。随着计算机技术的迅速发展，数值模拟已经渐渐成为管涌研究的一个重要手段和方法。殷建华[4]在国内首次采用有限元方法来研究管涌问题，计算了管涌发生区域的长度和渗透系数对堤身渗流的影响。……