【摘要】 强降雨作用下，公路路基容易发生滑塌灾害，不仅造成结构物与车辆的财产损失，而且会产生严重的人员伤亡。以宁波市“812”暴雨中山区农村公路的路基边坡滑塌实例为研究对象，将渗流与应力耦合的有限元方法和强度折减法相结合，模拟了该路基边坡在暴雨下的土体渗流、变形与应变情况，分析了边坡稳定系数的变化规律，再现了路基边坡在暴雨条件下发生滑塌的过程，模拟得到的圆弧滑移面与实际滑移面一致，且有限元模拟得到的塑性区贯通时间与实际滑坡都发生在暴雨第一天。结果表明：（1）“812”暴雨中的雨水入渗导致路基边坡的土体饱和度升高，孔隙水压力增大，坡顶和坡脚的基质吸力分别下降了62.3%和47.0%，土体抗剪强度降低而发生软化；（2）路基边坡的坡脚首先出现塑性变形并向上发展，特大暴雨持续17.43 h后塑性区出现贯通现象，最终导致沿着滑移面的整体滑塌，这揭示了路基边坡的滑塌机理；（3）暴雨中路基边坡的稳定系数由1.16降至0.93，因此已处于不稳定状态，可以通过预测瞬态的稳定系数对边坡滑塌进行及时预警。更多还原

【Abstract】 Under the action of heavy rainfall, highway roadbeds are prone to landslide disasters, which not only cause property losses of structures and vehicles, but also produce serious casualties. This paper takes the landslide case of the roadbed slope of a rural road in the mountainous area of Ningbo City in the “812” rainstorm as the research object, combines the finite element method of seepage and stress that is coupled with the strength reduction method, simulates the seepage, deformation and strain of the soil body of this roadbed slope, analyzes the change rule of the stability coefficient of the slope, and reproduces the process of landslide under the heavy rainfall. The simulated circular landslide surface is consistent with the actual landslide surface, and the penetration time of the shaped zone obtained by finite element simulation is the same as that of the actual landslide on the first day of the rainstorm. The results show that:(1) The infiltration of rainwater in the “812” rainstorm led to the increase of soil saturation and pore water pressure, and the suction of the substrate at the top and foot of the slope decreased by 62.3% and 47.0%, respectively, and the softening occurred due to the decrease of the shear strength of the soil body;(2) The foot of the roadbed slope firstly showed plastic deformation and developed upward, and the plastic zone appeared to penetrate after 17.43 h of the heavy rainstorm, which finally led to the overall landslide along the slip surface, revealing the landslide mechanism of the roadbed slope;(3) The stability coefficient of the roadbed slope was reduced from 1.16 to 0.93 during the rainstorm, and thus was in an unstable state, and the transient stability coefficient could be predicted to give timely warning of the sliding of the slope.更多还原