School of Water Conservancy Engineering, Jiangxi University of Water Resources and Electric Power, Nanchang 330099, China
Show less
文章历史+
Received
Published
Revised
2025-10-15
2026-06-18
2025-12-18
Online Date
2026-08-04
PDF (5824K)
摘要
为实现对边坡安全系数(factor of safety,FOS)的高精度预测,提出一种异质集成框架SHEN(slope-stability hybrid ensemble network)能够捕捉边坡稳定性分析中复杂的非线性物理关系。该框架首先对极端梯度提升(extreme gradient boosting,XGBoost)与快速梯度提升(light gradient boosting machine,LightGBM)进行贝叶斯优化,并基于验证集性能进行加权集成,然后与深度神经网络(deep neural networks,DNN)进行二级动态融合。基于整理的302个开挖土质边坡工程案例数据库(特征包括重度γ、黏聚力c、内摩擦角ϕ、边坡角β、边坡高度H、孔隙压力比ru等),在10折交叉验证与独立测试集上评估模型的回归与分类[FOS(factor of safety)≥1.0为稳定]性能。结果表明,利用SHEN在测试集计算可得R2=0.907 0、RMSE=0.148 9、MAE=0.107 4,并在分类任务上计算得到F1为0.937 5,这些结果均优于XGBoost、LightGBM与单一DNN模型。该方法兼具树模型的特征选择能力与神经网络的非线性表征能力,为岩土工程中的稳定性评估与风险管理提供了可复用的技术路径。
Abstract
A heterogeneous ensemble framework, termed the slope-stability hybrid ensemble network (SHEN), was proposed for high-precision prediction of the factor of safety (FOS) in slope stability. The framework was engineered to model the complex nonlinear physical relationships inherent in slope stability analysis. Within this framework, Bayesian optimization was first applied to tune the hyperparameters of extreme gradient boosting (XGBoost) and light gradient boosting machine (LightGBM), followed by weighted integration based on their performance on the validation set. Subsequently, a secondary dynamic fusion was performed using a deep neural network (DNN). Based on a database of 302 excavation soil slope engineering cases (including characteristics such as heavy γ, cohesion c, internal friction angle ϕ, slope angle β, slope height H, pore pressure ratio ru, etc.), the regression and classification of the model were evaluated on a 10 fold cross validation and independent test set, with FOS (factor of safety) ≥ 1.0 as the stability performance. The assessment was carried out through 10-fold cross-validation and an independent test set. The results demonstrate that superior performance was demonstrated by SHEN on the test set. R2of 0.907 0, RMSE of 0.148 9, and MAE of 0.107 4 are achieved for regression, an F1 of 0.937 5 is obtained for classification. All individual benchmark models are outperformed. Individual XGBoost, LightGBM, and DNN models are outperformed by SHEN. The feature selection capabilities of tree-based models are integrated with the nonlinear representation abilities of neural networks in this approach. A reusable technical framework for stability assessment and risk management in geotechnical engineering is provided by this method.
WANG Zhan-hong, GAN Jian-jun.
Prediction of Slope Factor of Safety Based on SHEN Model[J].
Science Technology and Engineering, 2026, 26(17): 7165-7173 DOI:10.12404/j.issn.1671-1815.2507238
边坡稳定性分析是岩土工程中的关键问题之一,广泛应用于土木工程、交通基础设施、矿山开采等领域[1-2]。边坡失稳不仅可能导致严重的经济损失和人员伤亡,还会对周围环境造成不可逆的破坏[3-4]。因此,精确评估边坡稳定性、预测边坡安全系数(factor of safety, FOS)并提前发出预警,对于确保工程安全、降低灾害风险具有重要意义[5]。
传统的边坡稳定性评价方法,如工程类比法、极限平衡法和数值分析法等,虽然在一定程度上能够提供理论支持,但这些方法在处理复杂的非线性关系和多因素影响时存在一定局限性[6]。例如,极限平衡法假设边坡材料为均质、各向同性,且忽略了实际边坡中复杂的力学行为;数值分析法则依赖于大量的现场数据和复杂的计算,往往难以适用于动态变化的实际工程[7-8]。在这种背景下,基于人工智能和机器学习(artificial intelligence and machine learning,AI/ML)技术的边坡稳定性预测方法日益受到关注,因为它们能够通过挖掘大量数据中潜在的非线性关系,克服传统方法的局限性[9]。
ArifA, ZhangC, SajibM H, et al. Rock slope stability prediction: a review of machine learning techniques[J]. Geotechnical and Geological Engineering, 2025, 43(1): 1-25.
ZhangYongshun, ChangMingjun, LiuYuhang, et al. Cut slope stability prediction in permafrost region based on ensemble learning[J]. Journal of Highway and Transportation Research and Development, 2025, 42(1): 131-139.
TianJunwei, JiangXingyuan, LiYangchun, et al. Analysis of disaster mechanism and instability process of bedding landslide with weak inter-layer in Triassic strata of Guizhou[J]. Science Technology and Engineering, 2025, 25(9): 3593-3603.
[6]
EhsanM, AneesM T, BakarA, et al. A review of geological and triggering factors influencing landslide susceptibility: artificial intelligence-based trends in mapping and prediction[J]. International Journal of Digital Earth, 2025, 18(1): 200-225.
[7]
EhsanM, AneesM T, BakarA, et al. A review of geological and triggering factors influencing landslide susceptibility: artificial intelligence-based trends in mapping and prediction[J]. International Journal of Digital Earth, 2025, 18(1): 200-225.
[8]
NanehkaranY A, LicaiZ, ChengyongJ, et al. Comparative analysis for slope stability by using machine learning methods[J]. Applied Sciences, 2023, 13(15): 8765.
ChenLei, ZhangQiang, JiaChaojun, et al. Centrifugal modeling and numerical simulation on stability of reservoir bank accumulation slope caused by heavy rainfall[J]. Rock and Soil Mechanics, 2024, 45(5): 1423-1434.
YuanXushan, LiuJinghui, HuangLongsheng, et al. Assessment of landslide susceptibility in Xinyuan County based on machine learning models[J]. Science Technology and Engineering, 2025, 25(5): 1815-1826.
[15]
HeL, CogganJ, FrancioniM, et al. Maximizing impacts of remote sensing surveys in slope stability—a novel method to incorporate discontinuities into machine learning landslide prediction[J]. International Journal of Geo-Information, 2021, 10(9): 587.
[16]
KumarS, SinghL K, RoyL B, et al. Enhancing the sustainability of slope stability in embankment construction by leveraging smart sensors and monitoring systems for data-driven insights[J]. International Journal of Sustainable Engineering, 2025, 18(2): 134-156.
[17]
Bahadori-JahromiA, RoomS, PaknahadC, et al. The role of artificial intelligence and machine learning in advancing civil engineering: a comprehensive review[J]. Applied Sciences, 2025, 15(10): 6789.
[18]
RanjanK D, WipulanusatW. Advancements in predicting soil liquefaction susceptibility: a comprehensive analysis of ensemble and deep learning approaches[J]. Scientific Reports, 2025, 15(2): 7890.
[19]
AhmadA, HusainA. Applications of machine learning in seismic performance assessment: trends, challenges, and future directions[EB/OL]. (2025-08-04)[2026-03-31]. https://doi.org/10.20944/preprints202508.0177.v1.
SuGuoshao, SongYongchun, YanLiubin. Application of Gaussian process machine learning in slope stability evaluation[J]. Rock and Soil Mechanics, 2009, 30(3): 675-679, 687.
[22]
LiJ, WangF. Study on the forecasting models of slope stability under data mining[C]//Earth and Space 2010:Engineering, Science, Construction, and Operations in Challenging Environments. Reston: ASCE, 2010: 765-776.
[23]
ManouchehrianA, GholamnejadJ, SharifzadehM. Development of a model for analysis of slope stability for circular mode failure using genetic algorithm[J]. Environmental Earth Sciences, 2014, 71(4): 1791-1801.
[24]
XuW, ShaoJ F. Artificial neural network analysis for the evaluation of slope stability[C]// Proceedings of the International Conference on Neural Networks in Engineering. Chem: Springer, 1998: 521-524.
[25]
KudaibergenovM, NurakynovS, IskakovB, et al. Application of artificial intelligence in landslide susceptibility assessment: review of recent progress[J]. Remote Sensing, 2024, 16(12): 2103.
[26]
ArifA, ZhangC, SajibM H, et al. Rock slope stability prediction: a review of machine learning techniques[J]. Geotechnical and Geological Engineering, 2025, 43(1): 1-25.
[27]
EhsanM, AneesM T, BakarA, et al. A review of geological and triggering factors influencing landslide susceptibility: artificial intelligence-based trends in mapping and prediction[J]. International Journal of Digital Earth, 2025, 18(1): 200-225.
[28]
Waris K A, ur Rahaman M A, Basha B M. Detection and prediction of slope stability in unsaturated finite slopes using interpretable machine learning[J]. Modeling Earth Systems and Environment, 2025, 11(4): 123-145.
[29]
KurnazT F, ErdenC, DağdevirenU, et al. Comparison of machine learning algorithms for slope stability prediction using an automated machine learning approach[J]. Natural Hazards, 2024, 120(3): 567-589.
[30]
Waris K A, ur Rahaman M A, Basha B M. Detection and prediction of slope stability in unsaturated finite slopes using interpretable machine learning[J]. Modeling Earth Systems and Environment, 2025, 11(4): 123-145.
[31]
Bahadori-JahromiA, RoomS, PaknahadC, et al. The role of artificial intelligence and machine learning in advancing civil engineering: a comprehensive review[J]. Applied Sciences, 2025, 15(10): 6789.