摘要
为了实现对非黏性土公路边坡的稳定性实时预警,采用神经网络方法建立了公路边坡稳定性安全系数FS和变形值的关系模型。该方法克服了FS不能实时获取的弊端,由实时测量的变形值计算出FS,并通过FS实现无黏性土公路边坡稳定性的实时预警,避免了传统实时预警方法中需要根据经验设定各种变形值阀值的问题。对某无黏性土公路边坡的试验研究表明,神经网络模型计算精度优于其他经验模型,且能够满足工程实时监测的需要。
In order to solve the problem of early warning to eohesionless soil highway slop real-time stability analysis, a mathmatical model between safety factor of highway slope stability Fs and deformation value based on artificial neural network (ANN)is built. It can be work out Fs with real-time deformation value to highway slope, overcome the disadvantages that Fs can't be obtained in time. It's can be realize real-time warning of cohesionless soil highway slope stability by Fs, to avoid the traditional real-time warning methods in highway slope stability that have to set the threshold value of all deformation value. The experiment on the sample data of one cohesionless soil highway slop demonstrates that this model is superior to others in accuracy and adaptability, also can be able to meet the need of real-time monitoring engineering.
出处
《公路》
北大核心
2013年第7期32-37,共6页
Highway
基金
"十一五"国家科技支撑计划项目
项目编号2009BAG13A02
北京市交通委项目
关键词
无黏性土公路边坡
稳定性安全系数Fs
变形值
神经网络法
cohesionless soil highway slope
safety factor of stability Fs
deformation value
Artifi-cial Neural Network (ANN)