基于一阶线性应变软化理论的边坡稳定性研究

在持续降雨或开挖卸荷作用下, 土体的强度指标会发生劣化, 但现今采用的边坡稳定性计算大多直接将其视为一个常数。为接近真实的边坡失稳破坏模式, 基于瑞典条分法以及一阶线性应变软化机制, 提出了一种边坡渐进破坏分析新方法, 推导出应变软化型边坡极限平衡表达式, 并获得了各破坏进度下边坡的安全系数。通过对模拟算例进行分析, 证明了条分-软化法的可靠性, 且计算结果表明渐进破坏过程中安全系数不仅取决于边坡的破坏方式与强度参数, 还与岩土体的软化模量密切相关。同时, 通过与滑坡实际案例的对比验算, 证实其强度指标存在不同的衰减系数, 即黏聚力的衰减系数大于摩擦角。从理论到应用, 最终获得的条分-软化法, 不仅考虑了岩土体的强度劣化效应以及滑动面的渐进发展, 还能有效地服务于实际工程背景下边坡的稳定性分析, 可以为滑坡的预防与治理提供指导建议。 

Study on slope stability considering first-order linear strain-softening theory

Under the action of continuous rainfall or excavation unloading, the strength index of soil deteriorates, but most of the current slope stability calculations directly regard it as a constant. To approach the real failure mode of slope instability, based on the Swedish slice method and the first-order linear strain-softening theory, this paper proposes a new method for slope progressive failure analysis, deduces the limit equilibrium expression of the strain-softening slope, and obtains the safety factor under the corresponding damage progress. Through the analysis of simulation examples, the feasibility of the slice-softening method is illustrated, and the calculation results show that the safety factor in the progressive failure process not only depends on the failure mode and strength parameters of the slope but is also closely related to the softening modulus of the rock and soil mass. At the same time, by comparing and checking the actual cases of the Danba landslide, it is confirmed that there are different attenuation coefficients for its strength index, that is, the attenuation coefficient of cohesion is greater than the friction angle. From theory to application, the finally obtained slice-softening method not only considers the strength degradation effect of rock and soil mass under the action of rainfall and the gradual development of sliding surface but can also effectively serve the stability analysis of slope under the background of practical engineering and provide guidance and suggestions for the prevention and treatment of landslides.