碎石土滑坡稳定性一元多重属性回归模型分析

为了能在地质勘查和试验的基础上对碎石土滑坡稳定性进行评价，构建了一元多重属性回归模型。基于官家滑坡的14个工程地质剖面的实测资料，选取滑体重量、滑面倾角、滑面长度、水力坡度、浸水面积、内聚力、内摩擦角7个影响因素，采用模型对影响因素和稳定性系数进行回归分析和影响因素显著性研究，得到计算稳定性系数的回归方程，并利用新昌下山滑坡进行模型准确性验证。研究结果表明：根据模型建立的线性回归方程回归性显著，能够用模型对滑坡进行稳定性计算分析；通过模型得出对稳定性系数有显著性影响的因素，综合滑坡实际地质状况确定地下水对稳定性有显著的影响，有助于开展滑坡灾害预警预报工作和采取有效的工程治理措施；新昌下山滑坡介于稳定与较不稳定状态之间，在降雨量比较大的时段应加强监测。

Stability Analysis of Gravelly Soil Landslide Using Multiple Properties Regression Model with One Variable

In order to evaluate the stability of gravel soil landslide based on the geological exploration information and experiment result, a multiple property regression model with one variable was created. Seven influencing factors such as the gravity of sliding body, angle and length of sliding surface, hydraulic gradient, immersion area, cohesion and internal friction angle were selected to conduct the regression analysis of stability coefficient and the significant research on these factors by using the model based on observed data of 14 engineering geological profiles of Guanjia landslide. The regression equation for calculating the stability coefficient of slopes was obtained, and Xinchang Xiashan landslide was taken to evaluate the accuracy of the model. Research results indicated that there was an obvious regression of the equation which was established according to the linear regression model. So the model can be used to calculate and analyze the stability of landslides. Factors that have significant influence on stability coefficient were determined by this model, and combined with the actual geological conditions, it was found that groundwater has a significant impact on the stability of landslides, which provides some useful references for the early-warning and prediction of landslide hazards and effective engineering control measures. The condition of Xinchang Xiashan landslide is between stability and instability,  intensive monitoring should be carried out during heavy rainfall periods.