青连高铁K152+300～K152+470边坡滑坡稳定性预测研究

Stability prediction of Qingdao-Lianyungang high-speed railway K152+300～K152+470 slope landslide

边坡稳定性是影响高速铁路安全运行的重要因素之一,开展高速铁路边坡滑坡稳定性预测研究对提高高铁运行安全具有一定的工程意义.以青岛-连云港高速铁路K152+300~K152+470边坡滑坡为例,通过设计合理的监测方案开展了边坡地表水平位移和土体深部水平位移监测,总结了滑坡位移规律;基于监测数据采用GM(1,1)模型、神经网络模型和灰色神经网络模型进行了滑坡位移预测,在验证预测效果的基础上基于灰色神经网络模型构建了滑坡位移预测模型,预测了该滑坡未来3年的累计位移并建立了高铁边坡滑坡位移预警方法.研究结果表明:DB2和GY2监测点的地表水平累计位移分别为17.23 cm和21.69 cm,SB2、SB4和GY2监测点的土体深部水平累计位移分别为13.42 cm、16.05 cm和18.37 cm;DB2、SB2、SB4、GY2监测点未来3年的累计位移最大预测值分别为47.05 cm、42.53cm、46.01 cm、52.36 cm(地表)和48.15 cm(深部),该滑坡将继续保持临界稳定状态;根据滑坡累计位移量将高速铁路边坡滑坡位移预警等级分为Ⅰ、Ⅱ和Ⅲ级.

Slope stability is one of the most important factors affecting the safe operation of high-speed railway, research on stability prediction of slope landslide has a certain engineering significance to improve high-speed railway safe operation. With Qingdao-Lianyungang high-speed railway K152 ＋300 - K152 ＋470 slope landslide as an example, through the reasonable design of monitoring scheme, the slope surface horizontal deformation and soil deep horizontal deformation monitoring were managed, and the law of landslide deformation was also summarized; Landslide deformation was predicted using the GM （ 1, 1 ） model, the neural network model and the grey ueural network model based on the monitoring data, landslide deformation prediction model based on the gray neural network model was constructed on the basis of prediction accuracy authentication, and the total deformation of the landslide over the next 3 years were predicted. The research results show that the surface horizontal deformation of DB2 and GY2 monitoring sites are 17.23 cm and 21.69 cm respectively, the soil deep horizontal deformation of SB2, SB4 and GY2 monitoring sites are 13.42 cm, 16.05 cm and 18.37 cm respectively. The total deformation of DB2,SB2,SB4,GY2 prediction sites over the next 3 years are 47.05 cm, 42.53 cm, 46.01 cm, 52.36 cm （slope surface） and 48.15 cm （soil deep）, the landslide will continue to keep the critical stable state. According to the prediction cumulative deformation, high speed railway landslide warning is divided into Ⅰ, Ⅱ and Ⅲ levels.