路基工程2024,Issue(5) :26-31.DOI:10.13379/j.issn.1003-8825.202309010

基于PSO-BP神经网络的膨胀土边坡稳定性分析

Stability Analysis of Expansive Soil Slope Based on PSO-BP Neural Network

唐淼 桂红生 李选正 李勇义 鲁俊
路基工程2024,Issue(5) :26-31.DOI:10.13379/j.issn.1003-8825.202309010

基于PSO-BP神经网络的膨胀土边坡稳定性分析

Stability Analysis of Expansive Soil Slope Based on PSO-BP Neural Network

唐淼 1桂红生 2李选正 3李勇义 4鲁俊4
扫码查看

作者信息

  • 1. 中交第二公路勘察设计研究院有限公司武汉 430056
  • 2. 中国港湾工程有限责任公司,北京 100010
  • 3. 山东大学基建部,济南 250061
  • 4. 深圳大学深地科学与绿色能源研究院,广东 深圳 518060;深圳大学广东省深地科学与地热能开发利用重点实验室,广东 深圳 518060;深圳大学土木与交通工程学院,广东 深圳 518060
  • 折叠

摘要

结合具体工程案例,运用Geo-Studio软件对边坡渗透特性和稳定性进行分析;采用基于粒子群优化算法(PSO)的反向传播BP神经网络算法对降雨条件下边坡最小安全系数进行预测,同时对影响边坡稳定性的参数进行优化.结果表明:坡顶孔隙水压力呈现先增大后减小的规律,坡底孔隙水压力在降雨期间逐渐增大,停雨后保持不变;不同非饱和参数对土体孔隙水压力及边坡稳定性有一定影响;基于PSO的BP神经网络算法,能够较好地对降雨工况下边坡的最小安全系数进行模拟预测和验证.

Abstract

In relation to specific project case,Geo-Studio software was used to analyze permeability and stability of slope,and particle swarm optimization(PSO)based back propagation(BP)neural network algorithm was used to predict the minimum safety factor of slope under rainfall conditions,at the same time,the parameters influencing the slope stability were optimized.The results show that,the pore water pressure on the slope top presents a law of first increasing and then decreasing,and the pore water pressure at the slope bottom gradually increases in the course of rainfall,and remains unchanged after rainfall.Different unsaturated parameters have certain influence on the pore water pressure of soil mass and the stability of slope.PSO-based BP neural network algorithm may do well in simulation prediction and verification on minimum safety factor of slope under rainfall conditions.

关键词

降雨/粒子群优化算法/BP神经网络/最小安全系数/非饱和参数

Key words

rainfall/particle swarm optimization/BP neural networks/minimum safety factor/unsaturated parameters

引用本文复制引用

出版年

2024
路基工程
中铁二局集团有限公司,西南交通大学,中铁二院工程集团有限责任公司

路基工程

影响因子:0.36
ISSN:1003-8825
段落导航相关论文