摘要
为了合理利用土石坝边坡的监测数据,提出基于支持向量机的土石坝边坡失稳预测模型。在考虑土石坝坝体和坝基参数变异性的条件下,利用蒙特卡洛方法获得土石坝边坡的位移监测数据库,采用支持向量机方法和边坡监测点位移构建土石坝边坡失稳预测模型,通过实际案例验证了方法的有效性。计算结果表明,该方法可以有效利用多点变形监测数据建立土石坝边坡失稳预测模型,准确率令人满意;另外,监测点的数量与位置也与模型预测准确率相关,该文提出的最优监测点设计方法可为实际工程监测提供更为高效且经济的布置方案。
In order to make reasonable use of the monitoring data of embankment slope,this paper proposes a prediction model of slope failure of embankment based on the support vector machine method.Firstly,based on the Monte Carlo method,a database of displacement monitoring of embankment slope is obtained considering the variabilities of soil parameters of embankment.Then,the support vector machine combines with the monitoring database are adopted to construct the prediction model of slope failure of embankment.An actual embankment is taken as an example to verify the effectiveness of proposed method.The results show that the proposed method can effectively integrate the multi-point monitoring data of deformation to establish the high efficiency prediction model of slope failure of embankment.In addition,the number and location of monitoring points are also related to the accuracy of model predictions,the proposed method can provide a more efficient and economical optimal monitoring point design method in practice.
作者
沈金明
SHEN Jinming(China Jiangxi International Economic and Technical Cooperation Co.,Ltd.,Nanchang 330000,China)
出处
《广东水利水电》
2023年第5期21-25,共5页
Guangdong Water Resources and Hydropower
关键词
土石坝边坡
现场监测
支持向量机
预测模型
embankment slope
field monitoring
support vector machine
prediction model
作者简介
沈金明(1995-),男,本科,助理工程师,主要从事水利工作。
