路堤下伏溶洞受力模式和变形破坏的数值模拟

为进一步揭示溶洞的受力模式和变形破坏过程，根据勘察结果建立考虑地层分布、溶洞形状、溶洞位置等因素的模型，模拟路堤分层填筑的过程，采用力法、强度折减法分析溶洞受力模式及其对破坏的影响。通过预埋钻孔多点位移计监测路堤填筑过程中溶洞及其上覆土层的变形破坏过程，并建立数值模型重现其破坏过程。根据应力状态和Hoek-Brown强度包络线，将破坏形式划分为张拉破坏、拉伸剪切破坏和压缩剪切破坏，并分析顶板倾角和洞穴形状对破坏形式、变形及稳定性的影响。基于抗弯理论，推导路堤容许填筑高度的解析解，并根据地应力和拱效应进行修正。结果表明：弯拉应力在拱效应和地应力的挤压作用下减弱，导致溶洞顶板进入剪切塑性状态，而非拉伸塑性状态；由于未能考虑拱效应和地应力的挤压作用，以往按照简支梁假设计算的弯拉应力结果偏大；矩形溶洞的顶板受力状态以拉剪为主，而椭圆形溶洞的顶板由于拱效应受力状态主要为压剪，实际形状溶洞的稳定性介于二者之间；顶板倾角（25°以内）对变形和稳定性影响不明显，但溶洞顶板的受力模式由压剪变为拉剪；为防止岩溶失稳，应控制路堤填筑高度，但现行规范中厚跨比大于0.8的规定过于保守，抗弯估算法优于厚跨比评价法，但仍偏保守，考虑地应力和拱效应的修正抗弯估算法最接近工程实际和数值计算结果。

Numerical Simulation of the Mechanical Characteristic and Failure Mode of Karst Subgrade

To reveal the mechanical characteristic and failure mode of the Karst subgrade, a Karst cave model was established that considers the formation distribution, shape of the cave, and location of the cave based on the study results. The staged modeling method was adopted to simulate the layered filling process of the subgrade. The force method and the strength reduction method were used to analyze the mechanical characteristic and its effect on the failure mode. A multipoint displacement meter was used to monitor the deformation and failure process of the Karst subgrade during the subgrade filling process. Then, a numerical model was established to reproduce the failure process. According to stress state and the Hoek-Brown strength envelope, the failure modes are divided into tensile failure, tensile shear failure, and compression shear failure. The influences of roof tilt angle and cave shape were analyzed. Based on the bending theory, the analytical solution of the allowable subgrade filling height was derived. Then, corrections were made according to the ground stress and the arch effect. It is observed that the flexural tensile stress in the cave roof may be counteracted due to the arching effect and the compression of ground stress, leading to the formation of a shear plastic state, rather than a stretched plastic state. In the past, the calculation results of bending stress based on the simply supported beam assumptions are relatively large, due to the failures of considering the arch effect and the compressive effect of the ground stress. The stress state of the roof of the rectangular cave is mainly tensile shear, whereas the roof of the elliptical cave is mainly compression shear, and the stability of the actual shape of the cave is between the two. The inclination of the roof (within 25°) has no obvious influence on the size and stability of the roof, but the stress model of the roof of the cave may shift from the compression shear zone to the stretching shear zone. To prevent Karst instability, the subgrade filling height should be reasonably controlled; however, the current regulations are too conservative to allow the thickness-to-span ratio to be smaller than 0.8. The antibend estimation method was observed to be better than the thickness-to-span ratio evaluation method, but it is still somewhat conservative. The revised antibend estimation method that considers the ground stress and arch effect is closest to the actual engineering fact and numerical calculation results.