LASSO algorithm and its application in slope stability analysis
XIE Meng-long1, YE Xin-yu1, ZHANG Sheng1, SHENG Dai-chao1,2
1. School of Civil Engineering, Central South University, Changsha 400041, China; 2. School of Civil and Environmental Engineering, Sydney University of Technology Sydney, NSW 2007, Australia
Abstract：By introducing the LASSO algorithm into the slope reliability analysis, an algorithm model to predict the safety factor of a slope is established, and the function of searching the dangerous area of the slope is realized. First, using the finite element software is used to implement the Monte Carlo algorithm and to obtain the data of slope reliability analysis. Second, the acquired data is analyzed by the LASSO algorithm. A model is established to predict the safety factor of the slope under the specific intensity parameter distribution. The predicted results are compared with those by the ordinary linear regression algorithm to confirm the superiority of the LASSO algorithm, and its application suggestions in long-term slope risk monitoring are put forward. Third, the LASSO algorithm is combined with the Monte Carlo simulation to search for the most dangerous areas that affect the stability of the slope under multiple simulation results. The results indicate that compared with the ordinary linear regression algorithm, the model established by the LASSO algorithm can accurately find out the most dangerous area that affects the stability of the slope. Therefore, the LASSO algorithm can provide new ideas for the long-term slope risk monitoring and slope reinforcement.
[1] 刘树林. 黄土边坡可靠度的敏感性分析[D]. 西安: 长安大学, 2010. (LIU Shu-lin.Sensitiwity Analysis on Loess Slope Reliability[D]. Xi'an: Chang'an University, 2010. (in Chinese)) [2] 蒋水华, 杨建华, 姚池, 等. 考虑土体参数空间变异性边坡失稳风险定量评估[J]. 工程力学, 2018, 35(1): 136-147. (JIANG Shui-hua, YANG Jian-hua, YAO Chi, et al.Quantitative risk assessment of slope failure considering spatial variability of soil properties[J]. Engineering Mechanics, 2018, 35(1): 136-147. (in Chinese)) [3] 祁小辉, 李典庆, 周创兵, 等. 考虑土体空间变异性的边坡最危险滑动面随机分析方法[J]. 岩土工程学报, 2013, 35(4): 745-753. (QI Xiao-hui, LI Dian-qing, ZHOU Chuang-bing, et al.Stochastic analysis method of critical slip surfaces in soil slopes considering spatial variability[J]. Chinese Journal of Geotechnical Engineering, 2013, 35(4): 745-753. (in Chinese)) [4] 蒋水华, 刘贤, 黄发明, 等. 考虑多参数空间变异性的降雨入渗边坡失稳机理及可靠度分析[J]. 岩土工程学报, 2020, 42(5): 900-907. (JIANG Shui-hua, LIU Xian, HUANG Fa-ming, et al.Failure mechanism and reliability analysis of soil slopes under rainfall infiltration considering spatial variability of multiple soil parameters[J]. Chinese Journal of Geotechnical Engineering, 2020, 2020, 42(5): 900-907. (in Chinese)) [5] 张光鹏. 基于蒙特卡罗法的黄土滑坡稳定性研究[D]. 兰州:兰州大学, 2013. (ZHANG Guang-peng.Research on the Stability of Loess Landslide Based on Monte Carlo Method[D]. Lanzhou: Lanzhou University, 2013. (in Chinese)) [6] JI J, LOW B K.Stratified response surfaces for system probabilistic evaluation of slopes[J]. Journal of Geotechnical & Geoenvironmental Engineering, 2012, 138(11): 1398-1406. [7] 傅旭东, 赵善锐. 用蒙特卡洛(Monte—Carlo)方法计算岩土工程的可靠度指标[J]. 西南交通大学学报, 1996(2): 164-168. (FU Xu-dong, ZHAO Shan-rui.Using Monte Carlo method to calculate the reliability index of geotechnical engineering[J]. Journal of Southwest Jiaotong University, 1996(2): 164-168. (in Chinese)) [8] 李猛. 边坡稳定可靠度的蒙特卡罗数值模拟及其应用研究[D]. 郑州: 郑州大学, 2004. (LI Meng.Monte Carlo Numerical Simulation of Slope Stability Reliability and Its Application Research[D]. Zhengzhou: Zhengzhou University, 2004. (in Chinese)) [9] CHING J, PHOON K K, HU Y G.Efficient evaluation of reliability for slopes with circular slip surfaces using importance sampling[J]. Journal of Geotechnical & Geoenvironmental Engineering, 2009, 135(6): 768-777. [10] GRIFFITHS D V, HUANG J S, FENTON G A.Influence of spatial variability on slope reliability using 2-D random fields[J]. Journal of Geotechnical and Geoenvironmental Engineering, 2009, 135(10): 1367-1378. [11] KANG F, XU Q, LI J.Slope reliability analysis using surrogate models via new support vector machines with swarm intelligence[J]. Applied Mathematical Modelling, 2016, 40(11/12): 6105-6120. [12] LANE P A, GRIFFITHS D V.Slope stability analysis by finite elements[J]. Géotechnique, 1999, 49(3): 387-403. [13] HE X, XU H, SABETAMAL H, et al.Machine learning aided stochastic reliability analysis of spatially variable slopes[J]. Computers and Geotechnics, 2020, 126: 103711. [14] TREVOR H, ROBERT T, JEROME F.The Elements of Statistical Learning[M]. New York: Springer New York, 2009. [15] 崔琨鹏. 基于Lasso的变量选择方法及其在贝叶斯网络中的应用[D]. 济南: 山东师范大学, 2019. (CUI Kun-peng.Methods of Variable Selection Based on Lasso and Its Application in Bayesian Networks[D]. Jinan: Shandong Normal University, 2019. (in Chinese)) [16] 黄小城. 基于共线性问题的岩质边坡可靠度分析及其工程应用[D]. 重庆: 重庆大学, 2018. (HUANG Xiao-cheng.Reliability Analysis of Rock Slopes with Multicollinearity and Its Engineering Application[D]. Chongqing: Chongqing University, 2018. (in Chinese)) [17] LOMBARDO L G, DANIELA C C, MARTIN M, et al.High predictor dimensionality in slope-unit-based landslide susceptibility models through LASSO-penalized Generalized Linear Model[C]// Egu General Assembly Conference, 2017. [18] TIBSHIRANI R.Regression shrinkage and selection via the lasso[J]. Journal of the Royal Statal Society, Series B, 1996, 58(1): 267-288. [19] 熊颖. 基于Lasso和二进制粒子群优化的基因选择方法研究[D]. 镇江: 江苏大学, 2019. (XIONG Ying.A Study of Gene Selection Method Based on Lasso and Binary Particle Swarm Optimization[D]. Zhenjiang: Jiangsu University, 2019. (in Chinese))
