论文部分内容阅读
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 realize the real-time warning on the stability of the non-clay soil highway slope, a neural network model was established to establish the relationship between the safety factor FS and the deformation value of the highway slope stability. The method overcomes the shortcomings that the FS can not be obtained in real time, calculates the FS from the deformation values measured in real time and realizes real-time warning of the slope stability of the non-cohesive soil highway by the FS, which avoids the need of empirically setting the traditional real- Various deformation threshold value of the problem. Experimental study on a non-cohesive soil highway slope shows that the neural network model is more accurate than other empirical models and can meet the needs of real-time engineering monitoring.