理论与试验研究

基于机器学习的间断级配土应力—应变预测

  • 胡凌恺 ,
  • 徐山琳 ,
  • 王渤
展开
  • 浙江工业大学 土木工程学院,杭州 310014
胡凌恺(1999—),男,浙江慈溪人,博士,研究方向为数值模拟在岩土工程中的应用。E-mail:221122060134@zjut.edu.cn
徐山琳(1993—),女,西安人,博士,副研究员,主要从事多相颗粒流数值模拟。E-mail:xuxiaoshan7@126.com

收稿日期: 2025-10-20

  网络出版日期: 2026-09-07

基金资助

浙江省教育厅科研项目(Y202352704)

Stress-Strain Prediction for Gap-Graded Soils Based on Machine Learning

  • Hu Lingkai ,
  • Xu Shanlin ,
  • Wang Bo
Expand
  • College of Civil Engineering, Zhejiang University of Technology, Hangzhou 310014, P.R. China

Received date: 2025-10-20

  Online published: 2026-09-07

摘要

间断级配土的宏观力学性能因级配特性而显著变化,准确建立级配参数与力学性能关系表达式一直是工程领域重大挑战。为应对这一问题,采用离散元法生成了间断级配土的应力—应变数据集,提出了一种创新本构模型框架。该框架将细粒含量、粒径比和应变作为输入参数,偏应力为输出参数。基于贝叶斯优化方法(BO),分别采用随机森林(RF)模型和反向传播神经网络(BPNN)进行模型训练与验证。结果表明:RF模型和BPNN模型均展现出高决定系数;在预测完整间断级配土的应力—应变曲线时,RF模型的表现优于BPNN模型,其预测值与模拟值的一致性更高。研究成果可为解决级配参数与力学性能的复杂问题提供新思路。

本文引用格式

胡凌恺 , 徐山琳 , 王渤 . 基于机器学习的间断级配土应力—应变预测[J]. 地下空间与工程学报, 2026 , 22(4) : 1188 -1196 . DOI: 10.20174/j.JUSE.2026.04.08

Abstract

The macroscopic mechanical properties of gap-graded soils vary significantly due to gradation characteristics, and accurately establishing relationship expressions between gradation parameters and mechanical properties has long been a major challenge in engineering. To address this issue, stress-strain datasets of gap-graded soils were obtained using the discrete element method and an innovative constitutive model framework was proposed. This framework takes fine content, particle size ratio, and strain as input parameters, with deviatoric stress as the output parameter. Based on Bayesian Optimization (BO), the Random Forest (RF) model and the Back Propagation Neural Network (BPNN) were respectively adopted for model training and validation. Research results show that both RF and BPNN models demonstrate high coefficients of determination. When predicting the complete stress-strain curves of gap-graded soils, the RF model outperforms the BPNN model, with higher consistency between predicted and simulated values. This study successfully applies machine learning techniques to predict the mechanical properties of gap-graded soils, offering new insights into modeling the complex relationship between gradation parameters and mechanical properties.

参考文献

[1] Thevanayagam S, Shenthan T, Mohan S, et al. Undrained fragility of clean sands, silty sands, and sandy silts[J]. Journal of geotechnical and geoenvironmental engineering, 2002, 128(10): 849-859.
[2] Vallejo L E. Interpretation of the limits in shear strength in binary granular mixtures[J]. Canadian Geotechnical Journal, 2001, 38(5): 1097-1104.
[3] Peng M, Zhang L M. Breaching parameters of landslide dams[J]. Landslides, 2012, 9(1): 13-31.
[4] Yasuda S, Shimizu Y, Deguchi K. Investigation of the mechanism of the 2015 failure of a dike on Kinu River[J]. Soils and Foundations, 2016, 56(4): 581-592.
[5] Zhu Y, Nie Z, Gong J, et al. An analysis of the effects of the size ratio and fines content on the shear behaviors of binary mixtures using DEM[J]. Computers and Geotechnics, 2020, 118: 103353.
[6] 张嘎, 王刚, 尹振宇, 等. 土的基本特性及本构关系[J]. 土木工程学报, 2020, 53(2): 105-118. (Zhang Ga, Wang Gang, Yin Zhenyu, et al. A critical review on the research of fundamental behavior and constitutive relationship of the soil[J]. China Civil Engineering Journal, 2020, 53(2): 105-118. (in Chinese))
[7] 姚仰平, 张丙印, 朱俊高. 土的基本特性、本构关系及数值模拟研究综述[J]. 土木工程学报, 2012, 45(3): 127-150. (Yao Yangping, Zhang Bingyin, Zhu Jungao. Behaviors, constitutive models and numerical simulation of soils [J]. China Civil Engineering Journal, 2012, 45(3): 127-150. (in Chinese))
[8] Roscoe K H, Schofield A N, Thurairajah A. Yielding of clays in states wetter than critical[J]. Geotechnique, 1963, 13(3): 211-240.
[9] Wang Z, Liu X, Yang X, et al. An improved duncan-chang constitutive model for sandstone subjected to drying-wetting cycles and secondary development of the model in FLAC 3 D[J]. Arabian Journal for Science and Engineering, 2017, 42(3): 1265-1282.
[10] Labuz J F, Zang A. Mohr-Coulomb Failure Criterion[J]. Springer International Publishing, 2012.
[11] Shakya S, Inazumi S. Applicability of Numerical Simulation by Particle Method to Unconfined Compression Tests on Geomaterials[J]. Civil Engineering Journal, 2024, 10(1): 1-19.
[12] Ali T, Eldin M N, H W. The Effect of Soil-Structure Interaction on the Seismic Response of Structures Using Machine Learning, Finite Element Modeling and ASCE 7-16 Methods[J]. Sensors, 2023, 23(4): 2047.
[13] Aydın Y, Işıkdağ Ü, Bekdaş G, et al. Use of Machine Learning Techniques in Soil Classification[J]. Sustainability, 2023, 15(3): 2374.
[14] 林永贵,王海波,魏立新,等. 基于机器学习的岩石节理面力学性能分析及预测[J].人民长江, 2024, 55(7): 240-246. (Lin Yonggui, Wang Haibo, Wei Lixin, et al. Analysis on mechanical properties of fractured rock mass and intelligent prediction based on machine learning[J]. Yangtze River, 2024, 55(7): 240-246. (in Chinese))
[15] Zhang N, Shen S L, Zhou A, et al. Application of LSTM approach for modelling stress-strain behaviour of soil[J]. Applied Soft Computing, 2021, 100: 106959.
[16] Sadik L. Developing Prediction Equations for Soil Resilient Modulus Using Evolutionary Machine Learning[J]. Transportation Infrastructure Geotechnology, 2024,11(4):1598-1620.
[17] 周世杰,孟上九,孙义强,等. 基于机器学习的黏土三轴应力—应变曲线预测研究[J]. 工业建筑, 2023,53(增2): 456-461. (Zhou Shijie, Meng Shangjiu, Sun Yiqiang, et al. Prediction of triaxial stress-strain curves of clay based on machine learning [J]. Industrial Construction, 2023, 53(Supp.2): 456-461. (in Chinese))
[18] Zeini H A, Al-Jeznawi D, Imran H, et al. Random Forest Algorithm for the Strength Prediction of Geopolymer Stabilized Clayey Soil[J]. Sustainability, 2023, 15(2): 1408.
[19] 王海波, 黄林冲, 林越翔,等. 基于机器学习模型的节理岩体UCS预测方法[J]. 人民长江, 2024, 55(增1): 192-197, 202. (Wang Haibo, Huang Linchong, Lin Yuexiang, et al. UCS prediction method for jointed rock masses based on machine learning models [J]. Yangtze River, 2024, 55(Supp.1): 192-197, 202. (in Chinese))
[20] Cundall P A, Strack O D L. A discrete numerical model for granular assemblies[J]. Géotechnique, 1979, 29(1): 47-65.
[21] Hazeghian M, Soroush A. DEM-aided study of shear band formation in dip-slip faulting through granular soils[J]. Computers and Geotechnics, 2016, 71: 221-236.
[22] Shi Z, Jiang T, Jiang M, et al. DEM investigation of weathered rocks using a novel bond contact model[J]. Journal of Rock Mechanics and Geotechnical Engineering, 2015, 7(3): 327-336.
[23] Xiong H, Zhang Z, Yang J, et al. Role of inherent anisotropy in infiltration mechanism of suffusion with irregular granular skeletons[J]. Computers and Geotechnics, 2023, 162: 105692.
[24] 姚志雄. 三维离散元法解析砂型泥石流细观运动特征研究[J]. 地下空间与工程学报, 2017, 13(5): 1414-1423. (Yao Zhixiong. Analysis on Mesoscopic Movement Characteristic of Sandy Debris Flow by 3D Discrete Element[J]. Chinese Journal of Underground Space and Engineering, 2017, 13(5): 1414-1423. (in Chinese))
[25] 路军富, 章春炜, 钟英哲, 等. 不同含水率砂卵石围岩宏细观力学特性研究[J]. 地下空间与工程学报, 2018, 14(6): 1564-1570. (Lu Junfu, Zhang Chunwei, Zhong Yingzhe, et al. Research on Macro-Meso Mechanical Properties of Sandy Cobble Surrounding Rock with Different Moisture Content[J]. Chinese Journal of Underground Space and Engineering, 2018, 14(6): 1564-1570. (in Chinese))
[26] Yang J, Gu X Q. Shear stiffness of granular material at small strains: does it depend on grain size?[J]. Géotechnique, 2013, 63(2): 165-179.
[27] Abbireddy C O R, Clayton C R I. Varying initial void ratios for DEM simulations[J]. Géotechnique, 2010, 60(6): 497-502.
[28] Minh N H, Cheng Y P. A DEM investigation of the effect of particle-size distribution on one-dimensional compression[J]. Géotechnique, 2013, 63(1): 44-53.
[29] Shire T, O'Sullivan C, Hanley K J, et al. Fabric and Effective Stress Distribution in Internally Unstable Soils[J]. Journal of Geotechnical and Geoenvironmental Engineering, 2014, 140(12): 04014072.
[30] Xie Y, Yang J, Zhang C, et al. A quantitative evaluation method for granular skeleton state of binary soils at arbitrary relative density[J]. Computers and Geotechnics, 2023, 153: 105083.
[31] Huang X, Hanley K J, O'Sullivan C, et al. Exploring the influence of interparticle friction on critical state behaviour using DEM[J]. International Journal for Numerical and Analytical Methods in Geomechanics, 2014, 38(12): 1276-1297.
[32] Yang J. The critical state friction angle of granular materials: does it depend on grading?[J]. Acta Geotechnica, 2018.
[33] Hu Z, Li J Z, Zhang Y D, et al. A CFD-DEM study on the suffusion and shear behaviors of gap-graded soils under stress anisotropy[J]. Acta Geotechnica, 2023, 18(6): 3091-3110.
[34] Yang Z X, Yang J, Wang L Z. On the influence of inter-particle friction and dilatancy in granular materials: a numerical analysis[J]. Granular Matter, 2012, 14(3): 433-447.
[35] Zhang L, Thornton C. DEM simulations of the direct shear test[A]//15th ASCE Engineering Mechanics Conference[C]. Columbia University New York, 2002: 2-5.
[36] Zhao T, Houlsby G T, Utili S. Investigation of granular batch sedimentation via DEM-CFD coupling[J]. Granular Matter, 2014, 16(6): 921-932.
[37] Cheng K, Wang Y, Yang Q, et al. Determination of microscopic parameters of quartz sand through tri-axial test using the discrete element method[J]. Computers and Geotechnics, 2017, 92: 22-40.
[38] Liu Y, Wang Y, Zhang J. New machine learning algorithm: Random forest[A]//Information Computing and Applications, ICICA 2012, Chengde, Chinat[C]. Springer Berlin Heidelberg, 2012: 246-252.
[39] Iwendi C, Anajemba J H, Biamba C, et al. Security of Things Intrusion Detection System for Smart Healthcare[J]. Electronics, 2021, 10(12): 1375.
[40] Kohonen T. An introduction to neural computing[J]. Neural networks, 1988, 1(1): 3-16.
[41] 单敬福, 陈欣欣, 赵忠军, 等. 利用BP神经网络法对致密砂岩气藏储集层复杂岩性的识别[J]. 地球物理学进展, 2015, 30(3): 1257-1263. (Dan Jingfu, Chen Xinxin, Zhao Zhongjun, et al. Prediction for Permeability of Tight Reservoir Based on Artificial Neural Network[J]. Progress in Geophysics, 2015, 30(3): 1257-1263. (in Chinese))
[42] 林磊, 王军, 刘行军, 等. 基于人工神经网络的致密储层渗透率预测[J]. 测井技术, 2021, 45(2): 179-184. (Lin Lei, Wang Jun, Liu Xingjun, et al. Permeability Prediction of Tight Reservoir Based on Artificial Neural Network[J]. Well Logging Technology, 2021, 45(2): 179-184.(in Chinese))
[43] Bergstra J, Bengio Y. Random search for hyper-parameter optimization[J]. Journal of machine learning research, 2012, 13(2): 281 - 305.
[44] Jones D R, Schonlau M, Welch W J. Efficient Global Optimization of Expensive Black-Box Functions[J]. Journal of Global Optimization, 1998, 13(4): 455-492.
[45] Ng T T, Zhou W, Chang X L. Effect of particle shape and fine content on the behavior of binary mixture[J]. Journal of Engineering Mechanics, 2017, 143(1): C4016008.
[46] Terzaghi K. Theoretical soil mechanics[M]. INC, 1943.
文章导航

/

〈 〉