运用神经网络估算黄土边坡的安全系数

Estimate safety factors of loess slope by using neural network

应用人工神经网络BP模型的非线性动力学、自学习和实时处理等特性,以黄土边坡实例作为输入样本,建立神经网络BP模型,并以安全系数作为输出值,估算黄土边坡稳定性的安全系数.通过试验数据对比,所得安全系数误差在6%以内,表明该计算方法可用于边坡稳定性分析.

Application of neural network theory for estimation of safety factors for loess slope,using loess slope cases as input samples,establish BP model,with safety factors as output values,carry experiment for estimation of safety factor for loess slope.The results of experiment indicate that the safety factors from this method is rather credible,so it is a practical method for loess slope stability analysis.