A slope stability prediction model using IHCS-BP neural network based on orthogonal experimental method

Authors

• Xiao Yarui, School of Civil Engineering , Hunan University of Science and Technology , Xiangtan , Hunan 411201 , ChinaFollow
• Ma Binhui, School of Civil Engineering , Hunan University of Science and Technology , Xiangtan , Hunan 411201 , China ;Hunan Province Key Laboratory of Geotechnical Engineering Stability Control and Health Monitoring , Hunan University of Science and Technology , Xiangtan , Hunan 411201 , ChinaFollow
• Lan Tian, School of Civil Engineering , Hunan University of Science and Technology , Xiangtan , Hunan 411201 , China
• Ma Minglei, China Construction Eighth Engineering Division Co ., Ltd., Shanghai 200000 , China
• Huang Houqing, China Communications 2nd Navigational Bureau 2nd Engineering Co ., Ltd., Chongqing 401121 , China
• Xiang Zeshi, School of Civil Engineering , Hunan University of Science and Technology , Xiangtan , Hunan 411201 , China
• Li Xiangrong, School of Civil Engineering , Hunan University of Science and Technology , Xiangtan , Hunan 411201 , China

Corresponding Author

马缤辉，男，博士，教授.E-mail：[email protected]

Abstract

To address the problems that traditional slope stability analysis methods struggle to quantify the coupling effect of multiple factors and that neural network prediction accuracy is insufficient, an IHCS-BP neural network prediction model based on orthogonal experimental design was proposed. Through orthogonal experiments, the sensitivity of six factors including cohesion and internal friction angle was analyzed, and their ranking was determined. The IHCS algorithm was employed for dynamically optimizing the weights of the BP network to improve the global optimization capability of the model, and the IHCS-BP prediction model was built. A total of 81 groups of orthogonal experimental data were used to verify the feasibility of the model, and 24 groups of actual slope cases were adopted to verify its generalization performance. The results show that the sensitivity of the influencing factors is in order of slope angle > internal friction angle > cohesion > slope height > pore pressure ratio > unit weight. In the 81 groups of orthogonal experimental data, the training set root mean square error (ERMSE, 0.084) and test set ERMSE (0.182) of the IHCS-BP model are reduced by more than 60% compared with those of the traditional BP model. The actual application results show that the prediction error of the IHCS-BP model (EMAPE = 9.365%) is significantly lower than those of the four comparison models, and the optimized model has higher accuracy in prediction. The results can provide reference for complex slope engineering.

Publication Date

10-29-2026

DOI

10.14048/j.issn.1671-2579.2026.05.002

First Page

10

Last Page

20

Submission Date

October 2026

Recommended Citation

Yarui, Xiao; Binhui, Ma; Tian, Lan; Minglei, Ma; Houqing, Huang; Zeshi, Xiang; and Xiangrong, Li (2026) "A slope stability prediction model using IHCS-BP neural network based on orthogonal experimental method," Journal of China & Foreign Highway: Vol. 46: Iss. 5, Article 2.
DOI: 10.14048/j.issn.1671-2579.2026.05.002
Available at: https://zwgl1980.csust.edu.cn/journal/vol46/iss5/2

Reference

[1]黄洪超.某高速公路巨型滑坡稳定性分析及处置方案研究[J].中外公路,2023,43(4):24-29.Huang Hongchao.Study on stability analysis and treatment of a expressway landslide [J].Journal of China & Foreign Highway,2023,43(4):24-29.

[2]王津锋,范胜通,谢海波.基于 IWOA-SVM 的边坡可靠度分析 [J].中外公路,2026,46(1):21-29.Wang Jinfeng,Fan Shengtong,Xie Haibo.Slope reliability analysis based on IWOA-SVM [J].Journal of China & Foreign Highway,2026,46(1):21-29.

[3]熊振涛.基于机器学习的边坡稳定性预测模型优化研究[D].昆明:昆明理工大学,2023.Xiong Zhentao.Research on optimization of slope stability prediction model based on machine learning [D].Kunming:Ku nming University of Science and Technology,2023.

[4]刘平,王美静,张少强,等.抗滑桩桩位对边坡稳定性影响机理的宏细观分析 [J].中外公路,2023,43(3):37-41.Liu Ping,Wang Meijing,Zhang Shaoqiang,et al.Macro and micro analysis of influence mechanism of anti-slide pile position on slope stability [J].Journal of China & Foreign Highway,2023,43(3):37-41.

[5]傅杨攀.基于大数据和卷积神经网络的边坡稳定性分析[D].广州:广东工业大学,2022.Fu Yangpan.Slope stability analysis based on big data and convolutional neural network [D].Guangzhou:Guangdong University of Technology,2022.

[6]李蕾.基于 GA-LM-BP 神经网络的降雨条件下边坡稳定性预测研究 [D].长沙:长沙理工大学,2017.Li Lei.Study on slope stability prediction under rainfall based on GA-LM-BP neural network [D].Changsha:Changsha University of Science & Technology,2017.

[7]王彪龙,孟凡利,曾超,等.基于自然选择策略的 PSO-BP神经网络的滑坡可靠性评价方法 [J].中外公路,2019,39(3):1-9.Wang Biaolong,Meng Fanli,Zeng Chao,et al.The method of landslide reliability evaluation based on natural selection strategy of PSO-BP neural network [J].Journal of China & Foreign Highway,2019,39(3):1-9.

[8]尹斐.基于改进的烟花算法在边坡工程中的应用研究[D].荆州:长江大学,2022.Yin Fei.Research on application of improved fireworks algorithm in slope engineering [D].Jingzhou:Yangtze University,2022.

[9]邓智中,田浩帆,包太.PCA和GA-BP神经网络在边坡稳定性预测中的应用 [J].采矿技术,2021,21(2):44-48.Deng Zhizhong,Tian Haofan,Bao Tai.Application of PCA and GA-BP neural network in slope stability prediction [J].Mining Technology,2021,21(2):44-48.

[10]董建华.基于改进粒子群优化算法的边坡稳定性分析[D].鞍山:辽宁科技大学,2016.Dong Jianhua.Slope stability analysis based on improved particle swarm optimization algorithm [D].Anshan:University of Science and Technology Liaoning,2016.

[11]胡卫东,曹文贵.基于改进粒子群算法的 BP神经网络在边坡稳定性评价中的应用 [J].湖南理工学院学报 (自然科学版 ),2014,27(2):71-76.Hu Weidong,Cao Wengui.Application of BP neural network in the slope stability evaluation based on improved particle swarm optimization algorithm [J].Journal of Hunan Institute of Science and Technology (Natural Sciences ),2014,27(2):71-76.

[12]李龙,高永涛,吴顺川,等.综合改进 BP神经网络在边坡稳定性预测中的应用 [J].煤炭技术,2015,34(9):202-205.Li Long,Gao Yongtao,Wu Shunchuan,et al.Application of comprehensive improvement of BP neural network in stability prediction of slope [J].Coal Technology,2015,34(9):202-205.

[13]Wang Y W,Liu D N,Dong H Y,et al.Research on a BP neural network slope safety coefficient prediction model based on improved sparrow algorithm optimization [J].Applied Sciences,2023,13(14):8446.

[14]胡少伟,李原昊,单常喜,等.基于改进的 PSO-BP 神经网络的边坡稳定性研究 [J].防灾减灾工程学报,2023,43(4):854-861.Hu Shaowei,Li Yuanhao,Shan Changxi,et al.Research on slope stability based on improved PSO-BP neural network [J].Journal of Disaster Prevention and Mitigation Engineering,2023,43(4):854-861.

[15]莫勇刚,丁德馨,肖猛.改进的 BP神经网络在边坡稳定性评价中的应用 [J].矿冶,2006,15(2):9-12.Mo Yonggang,Ding Dexin,Xiao Meng.Application of improved back propagation neural network on stability evaluation of slopes [J].Mining and Metallurgy,2006,15(2):9-12.

[16]韦立德,安少鹏,李鹏云,等.输电线路塔基土质边坡稳定性评价的 BP网络模型 [J].电力勘测设计,2013 (2):17-20.Wei Lide,An Shaopeng,Li Pengyun,et al.BP network model of stability evaluation of tower foundation soil slope in transmission line [J].Electric Power Survey & Design,2013 (2):17-20.

[17]衣瑛,孙厚广,李经科.基于正交设计方法排土场边坡稳定性敏感性分析 [C]//中国矿业科技文汇— 2014.合肥,2014:510-512.Yi Ying,Sun Houguang,Li Jingke.Sensitivity analysis of slope stability of waste dump based on orthogonal design method [C]// China Mining Science and Technology Literature- 2014.Hefei,2014:510-512.

[18]翟明洋,盛建龙,董舒,等.基于正交设计和灰色关联分析法的边坡敏感性耦合分析 [J].化工矿物与加工,2017,46(2):69-72.Zhai Mingyang,Sheng Jianlong,Dong Shu,et al.Coupled analysis of slope sensitivity based on orthogonal design and gray correlation analysis method [J].Industrial Minerals & Processing,2017,46(2):69-72.

[19]Ghodrati A,Lotfi S.A hybrid CS/PSO algorithm for global optimization [C]// Intelligent Information and Database Systems.Berlin,Heidelberg:Springer Berlin Heidelberg,2012:89-98.

[20]马技,刘彩霞,丁琳琳.基于改进布谷鸟搜索算法的矿震震源定位方法研究 [J].地球物理学进展,2024,39(3):1070 -1080.Ma Ji,Liu Caixia,Ding Linlin.Research on mining earthquake source localization method based on improved cuckoo search algorithm [J].Progress in Geophysics,2024,39(3):1070 -1080.

[21]俞社鑫.基于 GA-PSO-RBF 模型的均质土坡稳定性及可靠度预测 [D].赣州:江西理工大学,2022.Yu Shexin.Stability and reliability prediction of homogeneous soil slope based on GA-PSO-RBF model [D].Ganzhou:Jiangxi University of Science and Technology,2022.

[22]方开泰,马长兴.正交与均匀试验设计 [M].北京:科学出版社,2001.Fang Kaitai,Ma Changxing.Orthogonal and uniform experimental design [M].Beijing:Science Press,2001.

[23]刘子宁.基于产品族 DNA的自动化物流设备识别设计方法研究 [D].武汉:华中科技大学,2021.Liu Zining.Research on identification design method of automatic logistics equipment based on product family DNA [D].Wuhan:Huazhong University of Science and Technology,2021.

[24]赵梓景.基于 FAGSO-BP 神经网络模型的软土地基沉降预测研究 [D].湘潭:湖南科技大学,2023.Zhao Zijing.Study on settlement prediction of soft soil foundation based on FAGSO-BP neural network model[D].Xiangtan:Hunan University of Science and Technology,2023.

[25]王超.工程高边坡稳定性预测方法研究 [D].北京:北京交通大学,2009.Wang Chao.Study on stability prediction method of engineering high slope [D].Beijing:Beijing Jiaotong University,2009.

[26]孙荣琦.基于 GA-PSO-BP 神经网络的边坡稳定性评价研究 [D].西安:西安理工大学,2024.Sun Rongqi.Study on slope stability evaluation based on GA-PSO-BP neural network [D].Xi’an:Xi’an University of Technology,2024.

[27]李昆仲.基于 RBF神经网络的边坡稳定性评价研究:以西藏夏比公路 K85段边坡为例 [D].西安:长安大学,2010.Li Kunzhong.Study on slope stability evaluation based on RBF neural network:Taking K 85 slope of Xiabi highway in Tibet as an example [D].Xi’an:Chang’an University,2010.

[28]苏国韶,宋咏春,燕柳斌.高斯过程机器学习在边坡稳定性评价中的应用 [J].岩土力学,2009,30(3):675-679,687.Su Guoshao,Song Yongchun,Yan Liubin.Application of Gaussian process machine learning to slope stability evaluation[J].Rock and Soil Mechanics,2009,30(3):675-679,687.

[29]冯夏庭,王泳嘉,卢世宗.边坡稳定性的神经网络估计[J].工程地质学报,1995,3(4):54-61.Feng Xiating,Wang Yongjia,Lu Shizong.Neural network estimation of slope stability [J].Journal of Engineering Geology,1995,3(4):54-61.

[30]程纬华,乔登攀,张磊,等.BP神经网络在露天矿边坡稳定性分析中的应用 [J].矿冶,2012,21(2):10-14,36.Cheng Weihua,Qiao Dengpan,Zhang Lei,et al.Application of BP networks in the stability analysis of slopes in the open-pit mine [J].Mining and Metallurgy,2012,21(2):10-14,36.