库水变动下堆积层滑坡加卸载响应规律与稳定性预测

LOAD-UNLOAD REPONSE OF COLLUVIAL LANDSLIDES WITH RESERVOIR WATER LEVEL FLUCTUATION AND STABILITY PREDICTION

三峡库区堆积层滑坡稳定性受库水位变动影响十分明显,库水变动下堆积层滑坡的演化过程与稳定性预测研究对防灾减灾具有重要的指导意义。基于库水变动与滑坡变形的响应关系,建立库水动力加卸载与位移速率响应耦合的加卸载响应比预测模型;建立库水变动与滑坡稳定系数的响应关系,进而确定库水变动下滑坡体的渗流场类型,并以滑坡稳定系数的变化率的正负来判断库水变动的加卸载作用。以黄莲树滑坡为例,预测其稳定性,并对预测结果进行验证。结果表明:黄莲树滑坡水平方向位移变化与库水变动存在响应关系,且响应具有明显的滞后性;库水变动下该滑坡的渗流场属于动水压力型,每个水文年中库水动力对滑坡有6个月为加载过程,1个月为卸载过程;滑坡监测点的加卸载响应比在2011年出现整体上升并大于1,揭示滑坡趋于失稳,对库水变动加卸载作用的响应加强。结论得到了宏观变形破坏迹象的验证,说明改进的加卸载响应比预测模型具有良好的预测效果。

The stability of colluvial landslides in the Three Gorges Reservoir is obviously influenced by the fluctuation of water level. So studying the evolutionary process and stability prediction of colluvial landslides under the reservoir water level fluctuation is significant to provide references for disaster management and construction planning. The load-unload response prediction model in which reservoir water level was treated as the load-unload parameter and landslide displacement rate,as the load-unload response parameter was established. The model was based on the relationship between landslide deformation and reservoir water level fluctuation. And the type of landslide seepage field under the reservoir water level fluctuation was determined by revealing the relationship. Then the load- unload interval differences were evaluated by the rate of stability coefficient. Taking Huanglianshu landslide for example,the model was used to predict the development trend. The results show that the relationshipbetween horizontal displacement of the landslide and reservoir water level fluctuation has an obvious hysteretic nature. The hydrodynamic pressure landslide has a loading process of six months and an unloading process of one month. The load-unload response ratio of monitoring sites raises and is more than 1. It reveals that the landslide will be instable and the load-unload response is more obvious due to reservoir water level fluctuation. The application results show that the prediction is well coincident with the macroscopic deformation evidence. Thus the improved model is feasible in the stability prediction of landslides.