数据与物理力学机制双驱动的边坡土体参数高效反分析方法&#x0002A;

刘奕镳; 贾西贝

doi:10.7523/j.ucas.2026.041

中国科学院大学学报 >

0 2026041

DOI: https://doi.org/10.7523/j.ucas.2026.041

数据与物理力学机制双驱动的边坡土体参数高效反分析方法^*

刘奕镳 ,
贾西贝

展开

1.中国科学院武汉岩土力学研究所岩土力学与工程安全全国重点实验室,湖北武汉 430071;
2.中国科学院大学工程科学学院,北京 100049

收稿日期: 2026-04-21

修回日期: 2026-08-24

网络出版日期: 2026-08-26

基金资助

^*湖北省自然科学基金资助（2025AFB100）资助

收起

Data-driven and physics-informed approach for efficient back-analysis of slope soil parameters

LIU Yibiao ,
JIA Xibei

Expand

1. State Key Laboratory of Geomechanics and Geotechnical Engineering Safety, Institute of Rock and Soil Mechanics, Chinese Academy of Sciences, Wuhan Hubei 430071, China;
2. School of Engineering Science, University of Chinese Academy of Sciences, Beijing 100049, China

Received date: 2026-04-21

Revised date: 2026-08-24

Online published: 2026-08-26

Fold

摘要

针对边坡土体参数反分析中传统方法依赖经验、数据驱动方法缺乏物理机理等问题,提出一种融合弹塑性变形分析理论的物理信息神经网络（PINN）反分析方法。该方法将边坡弹塑性变形分析的控制方程及其定解条件,以约束形式嵌入神经网络训练过程,构建了物理力学机制与监测数据共同驱动的反分析框架。以待反分析土体参数为可训练超参数,通过与模型超参数同步优化,使模型预测结果同时满足物理机制与观测位移约束。两个典型二维数值算例验证表明：该方法仅需位移观测数据而无需预先大量数值样本,即可高效反分析出弹性模量及抗剪强度参数,且计算效率显著高于传统数据驱动方法。训练后的模型可同时作为高精度代理模型,直接用于位移场预测,实现了参数反分析与正演分析的一体化。

关键词： 边坡; 反分析; 物理信息神经网络（PINN）; 弹塑性分析; 数据与物理力学机制双驱动

本文引用格式

刘奕镳 , 贾西贝 . 数据与物理力学机制双驱动的边坡土体参数高效反分析方法^*[J]. 中国科学院大学学报, 0 : 2026041 . DOI: 10.7523/j.ucas.2026.041

Abstract

Conventional methods for back-analysis of slope soil parameters relied on subjective experience, while purely data-driven approaches tended to neglect underlying physical mechanisms. To address these limitations, a back-analysis method based on physics-informed neural networks (PINNs) that integrated elastoplastic deformation analysis theory was proposed. The governing equations and boundary/initial conditions of slope elastoplastic analysis were embedded as constraints into the neural network training process. A back-analysis framework jointly driven by physical laws and monitoring data was established. The soil parameters to be back-analyzed were treated as trainable hyperparameters in the network. By simultaneously optimizing these and the network's own parameters, the model's predictions were constrained to satisfy both the physical principles and the observed displacements. Validation via two-dimensional (2D) numerical cases demonstrated that that the proposed method could efficiently back-analyze elastic modulus, cohesion, and internal friction angle using only monitored displacement data, eliminating the need for extensive pre-computed numerical simulations to generate training samples. Its computational efficiency was shown to be significantly higher than that of conventional data-driven methods. Furthermore, the trained PINN model itself served as a high-fidelity surrogate model capable of directly predicting displacement fields, thereby achieving an integration of parameter back-analysis and forward analysis.

Key words： slope; back-analysis; physics-informed neural network (PINN); elastoplastic analysis; data-driven and physics-informed

参考文献

[1] 卢坤林,朱大勇,许强,等.滑带土抗剪强度参数的三维反分析[J].岩土力学,2010, 31(10):3319-3323,3328.DOI:10.16285/j.rsm.2010.10.010.
[2] 郑明新. 论滑带土强度特征及强度参数的反算法[J].岩土力学,2003, 24(4):528-532.DOI:10.16285/j.rsm.2003.04.010.
[3] 易朋莹,王凯,任佳,等.基于传递系数法的滑坡滑带土强度参数反分析研究[J].中国地质灾害与防治学报,2008, 19(4):23-26,32.DOI:10.16031/j.cnki.issn.1003-8035.2008.04.029.
[4] Zhao L H, Zuo S, Lin Y L, et al.Reliability back analysis of shear strength parameters of landslide with three-dimensional upper bound limit analysis theory[J]. Landslides, 2016, 13(4): 711-724.DOI:10.1007/s10346-015-0604-3.
[5] 黄帅,陶志刚,汪建明,等.基于雷视融合的边坡深部三维滑移面非接触判定方法研究[J/OL].地球科学, (2026-04-15)[2026-04-18].https://link.cnki.net/urlid/42.1874.p.20260414.1635.020.
[6] 徐卫亚,陈世壮,张贵科,等.基于PSO-LSSVM-BP模型的高边坡力学参数反分析及稳定性评价[J].河海大学学报(自然科学版),2024, 52(5):52-59.DOI:10.3876/j.issn.1000-1980.2024.05.007.
[7] 顾忠卫, 李国维, 贺冠军, 等. 基于三维数值模拟反分析的临危边坡锚固研究[J]. 土木工程学报, 2015, 48(S2): 202-207.
[8] 王震豪, 聂闻, 许汉华, 等.基于EEMD-Prophet-LSTM的滑坡位移预测[J]. 中国科学院大学学报, 2023, 40(4): 514-522.DOI:10.7523/j.ucas.2022.002.
[9] 冯夏庭, 张治强, 杨成祥. 位移反分析的进化神经网络方法研究[J]. 岩石力学与工程学报, 1999,18(5): 529-533.DOI:10.3321/j.issn:1000-6915.1999.05.008.
[10] Zhao H B, Yin S D.Geomechanical parameters identification by particle swarm optimization and support vector machine[J]. Applied Mathematical Modelling, 2009, 33(10):3997-4012.DOI:10.1016/j.apm.2009.01.011.
[11] 孙钱程,李邵军,赵洪波,等.基于位移和松弛深度的岩体参数概率反分析方法[J].岩石力学与工程学报,2019, 38(9):1884-1894.DOI:10.13722/j.cnki.jrme.2019.0003.
[12] 杨朝晖,姜清辉,薛利军,等.高边坡开挖变形多目标全过程反分析方法[J].武汉大学学报(工学版),2024, 57(8):1065-1076.DOI:10.14188/j.1671-8844.2024-08-006.
[13] 毛伟琦,李小珍,王翔,等.基于LSSVM和GWOPSO算法的桥岸边坡位移反演方法研究[J].铁道科学与工程学报,2023, 20(11):4299-4310.DOI:10.19713/j.cnki.43-1423/u.t20231026.
[14] 揭鸿鹄, 蒋水华, 万建宏, 等. 基于微调DeepONet模型的非饱和边坡参数贝叶斯反分析[J]. 岩土工程学报, 2026, 48(4): 825-834.DOI: 10.11779/CJGE20250067.
[15] Liu Y B, Ren W Z, Liu C C, et al.Displacement-based back-analysis frameworks for soil parameters of a slope: Using frequentist inference and Bayesian inference[J]. International Journal of Geomechanics, 2022, 22(4): 04022026. DOI:10.1061/(asce)gm.1943-5622.0002318.
[16] Wang Z Z.Deep learning for geotechnical reliability analysis with multiple uncertainties[J]. Journal of Geotechnical and Geoenvironmental Engineering, 2022, 148(4): 06022001. DOI:10.1061/(asce)gt.1943-5606.0002771.
[17] 窦杰,向子林,许强,等.机器学习在滑坡智能防灾减灾中的应用与发展趋势[J].地球科学,2023, 48(5):1657-1674.DOI:10.3799/dqkx.2022.419.
[18] Kamali A, Sarabian M, Laksari K.Elasticity imaging using physics-informed neural networks: Spatial discovery of elastic modulus and Poisson's ratio[J]. Acta Biomaterialia, 2023, 155: 400-409.DOI:10.1016/j.actbio.2022.11.024.
[19] Cai S Z, Mao Z P, Wang Z C, et al.Physics-informed neural networks (PINNs) for fluid mechanics: A review[J]. Acta Mechanica Sinica, 2021, 37(12): 1727-1738.DOI:10.1007/s10409-021-01148-1.
[20] 兰鹏, 李海潮, 叶新宇, 等. PINNs算法及其在岩土工程中的应用研究[J]. 岩土工程学报, 2021, 43(3): 586-592.DOI:10.11779/CJGE202103023.
[21] Bekele Y W.Physics-informed deep learning for one-dimensional consolidation[J]. Journal of Rock Mechanics and Geotechnical Engineering, 2021, 13(2): 420-430.DOI:10.1016/j.jrmge.2020.09.005.
[22] Haghighat E, Raissi M, Moure A, et al.A physics-informed deep learning framework for inversion and surrogate modeling in solid mechanics[J]. Computer Methods in Applied Mechanics and Engineering, 2021, 379: 113741.DOI:10.1016/j.cma.2021.113741.
[23] 张子龙,潘秋景,仉文岗,等.基于物理信息神经网络的盾构隧道诱发地表沉降预测[J].工程力学,2024, 41(4):161-173.DOI:10.6052/j.issn.1000-4750.2022.04.0369.
[24] 张宏涛, 候宇翔, 赵宇飞. 边坡稳定下限分析的物理信息神经网络方法[J/OL]. 工程力学, (2025-11-17)[2026-04-18].https://kns.cnki.net/kcms2/article/abstract?v=BsQQ9aL8NZtVnNWgQZghjrVFsbzmY0ZibqHMTvzEoNPXoyxXG9KkGxP14Cyn6pEvCJIQTuRxzRvh_GSDUoVgR6oo_Z7lbgsv0aMItqpMSwszuCUaw-5vivCrQhiczpOcsj_gGIHCHEFISQSyDterf7Yj6eyeFMfE90Sab18dlY5H21Fw4-041A==&uniplatform=NZKPT&language=CHS.
[25] 李野, 陈松灿. 基于物理信息的神经网络:最新进展与展望[J]. 计算机科学, 2022, 49(4): 254-262.DOI:10.11896/jsjkx.210500158
[26] Hu Y D, Wang X H, Zhou H, et al.A more general electromagnetic inverse scattering method based on physics-informed neural network[J]. IEEE Transactions on Geoscience and Remote Sensing, 2023, 61: 4505109. DOI:10.1109/TGRS.2023.3301455.
[27] Huang F M, Xiong H W, Chen S X, et al.Slope stability prediction based on a long short-term memory neural network: Comparisons with convolutional neural networks, support vector machines and random forest models[J]. International Journal of Coal Science & Technology, 2023, 10(1): 18. DOI:10.1007/s40789-023-00579-4.
[28] Hsiao C H, Chen A Y, Ge L, et al.Performance of artificial neural network and convolutional neural network on slope failure prediction using data from the random finite element method[J]. Acta Geotechnica, 2022, 17(12): 5801-5811. DOI:10.1007/s11440-022-01520-w.
[29] Bai J S, Rabczuk T, Gupta A, et al.A physics-informed neural network technique based on a modified loss function for computational 2D and 3D solid mechanics[J]. Computational Mechanics, 2023, 71(3): 543-562.DOI:10.1007/s00466-022-02252-0.
[30] Imambi S, Prakash K B, Kanagachidambaresan G R.PyTorch[M]//Programming with TensorFlow: Solution for Edge Computing Applications. Cham: Springer International Publishing, 2021: 87-104. DOI:10.1007/978-3-030-57077-4_10.
[31] Kingma D P, Ba J. Adam: A method for stochastic optimization[EB/OL]. arXiv2014: 1412.6980.(2014-12-22)[2026-04-18]. https://doi.org/10.48550/arXiv.1412.6980.
[32] Simo J C, Taylor R L.A return mapping algorithm for plane stress elastoplasticity[J]. International Journal for Numerical Methods in Engineering, 1986, 22(3): 649-670.DOI:10.1002/nme.1620220310.
[33] Griffiths D V, Lane P A.Slope stability analysis by finite elements[J]. Géotechnique, 1999, 49(3): 387-403.DOI:10.1680/geot.1999.49.3.387.

Options

文章导航

摘要

本文引用格式

Abstract

参考文献

访问统计

联系我们

模态框（Modal）标题

摘要

本文引用格式

Abstract

参考文献

访问统计

联系我们