基于套索与优化算法的矿山岩质边坡可靠度研究
Study on the Reliability of Mine Rock Slopes Based on Lasso and Optimization Algorithm
郑阿鑫1
作者信息
- 1. 江西理工大学土木与测绘工程学院,江西 赣州 341000
- 折叠
摘要
[目的]为解决拟合边坡极限状态方程时存在的过拟合与共线性等问题,且为获得更符合实际的安全系数,在考虑剪胀角的基础上,建立基于LASSO-GA-PSO的矿山岩质边坡可靠度分析模型.[方法]通过将Slide与FLAC3D相结合的方法,计算考虑剪胀角变化的边坡安全系数;采用LASSO对数据进行拟合,获得相应的极限状态方程;编写用遗传算法与LHS优化的自适应粒子群算法程序,建立基于LASSO-GA-PSO的边坡可靠度分析模型,进而确定边坡的可靠度指标.[结果]二次带交叉项的LASSO-GA-PSO模型所得的可靠度指标为 2.325 7,略小于LASSO-Monte-Carlo的2.387 9,误差为2.6%.[结论]模型具有较好的优化结果,可为类似工程提供参考.
Abstract
[Purposes]In order to solve the problems of overfitting and collinearity that exist in fitting the slope limit state equations,and to obtain more realistic safety factors,a mine rock slope reliability analy-sis model based on LASSO-GA-PSO is established based on the consideration of dilation angles.[Meth-ods]By combining Slide with FLAC3D,the slope safety factor considering the change of dilation angle is calculated the data were fitted by LASSO to obtain the corresponding limit state equation;The adaptive particle swarm optimization algorithm program based on genetic algorithm and LHS optimization is com-piled,and the slope reliability analysis model based on LASSO-GA-PSO is established to determine the reliability index of the slope.[Findings]The reliability index obtained from the LASSO-GA-PSO model of quadratic polynomial with cross terms is 2.325 7,which is slightly smaller than that of 2.387 9 of LASSO-Monte-Carlo,and the error is 2.6%.[Conclusions]The model has a good optimization result and can provide reference for similar projects.
关键词
矿山岩质边坡/剪胀角/FLAC3D/LASSO算法/可靠度指标Key words
mine rock slope/dilation angle/FLAC3D/LASSO algorithm/reliability index引用本文复制引用
出版年
2024
NETL
NSTL
维普
万方数据