Abstract: Machine learning (ML) has strong potential for soil settlement prediction, but determining hyperparameters for ML models is often intricate and laborious. Therefore, we apply Bayesian optimization to determine the optimal hyperparameter combinations, enhancing the effectiveness of ML models for soil parameter inversion. The ML models are trained using numerical simulation data generated with the modified Cam-Clay (MCC) model in ABAQUS software, and their performance is evaluated using ground settlement monitoring data from an airport runway. Five optimized ML models—decision tree (DT), random forest (RF), support vector regression (SVR), deep neural network (DNN), and one-dimensional convolutional neural network (1D-CNN)—are compared in terms of their accuracy for soil parameter inversion and settlement prediction. The results indicate that Bayesian optimization efficiently utilizes prior knowledge to identify the optimal hyperparameters, significantly improving model performance. Among the evaluated models, the 1D-CNN achieves the highest accuracy in soil parameter inversion, generating settlement predictions that closely match real monitoring data. These findings demonstrate the effectiveness of the proposed approach for soil parameter inversion and settlement prediction, and reveal how Bayesian optimization can refine the model selection process.

基于贝叶斯优化增强的机器学习方法在土体参数反演中的应用研究

[1]AbbasA, VantasselJP, CoxBR, et al., 2023. A frequency-velocity CNN for developing near-surface 2D vs images from linear-array, active-source wavefield measurements. Computers and Geotechnics, 156:105305.

[2]AkibaT, SanoS, YanaseT, et al., 2019. Optuna: a next-generation hyperparameter optimization framework. Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, p.2623-2631.

[3]AsaokaA, 1978. Observational procedure of settlement prediction. Soils and Foundations, 18(4):87-101.

[4]AsterisPG, SkentouAD, BardhanA, et al., 2021. Predicting concrete compressive strength using hybrid ensembling of surrogate machine learning models. Cement and Concrete Research, 145:106449.

[5]AsterisPG, KaroglouM, SkentouAD, et al., 2024. Predicting uniaxial compressive strength of rocks using ANN models: incorporating porosity, compressional wave velocity, and Schmidt hammer data. Ultrasonics, 141:107347.

[6]BardhanA, OzcanNT, AsterisPG, et al., 2024. Hybrid ensemble paradigms for estimating tunnel boring machine penetration rate for the 10-km long Bahce-Nurdagi twin tunnels. Engineering Applications of Artificial Intelligence, 136:108997.

[7]BenzaamiaA, GhriciM, RebouhR, et al., 2024a. Predicting the shear strength of rectangular RC beams strengthened with externally-bonded FRP composites using constrained monotonic neural networks. Engineering Structures, 313:118192.

[8]BenzaamiaA, GhriciM, RebouhR, et al., 2024b. Predicting the compressive strength of CFRP-confined concrete using deep learning. Engineering Structures, 319:118801.

[9]BergstraJ, BengioY, 2012. Random search for hyper-parameter optimization. The Journal of Machine Learning Research, 13:281-305. https://dl.acm.org/doi/10.5555/2188385.2188395

[10]BoserBE, GuyonIM, VapnikVN, 1992. A training algorithm for optimal margin classifiers. Proceedings of the Fifth Annual Workshop on Computational Learning Theory, p.144-152.

[11]BreimanL, 2001. Random forests. Machine Learning, 45(1):5-32.

[12]ChaabanM, HeiderY, SunW, et al., 2024. A machine-learning supported multi-scale LBM-TPM model of unsaturated, anisotropic, and deformable porous materials. International Journal for Numerical and Analytical Methods in Geomechanics, 48(4):889-910.

[13]ChenRP, ZhangP, KangX, et al., 2019. Prediction of maximum surface settlement caused by earth pressure balance (EPB) shield tunneling with ANN methods. Soils and Foundations, 59(2):284-295.

[14]ChengY, WangJF, HeY, 2023. Prediction models of newmark sliding displacement of slopes using deep neural network and mixed-effect regression. Computers and Geotechnics, 156:105264.

[15]FayedHA, AtiyaAF, 2019. Speed up grid-search for parameter selection of support vector machines. Applied Soft Computing, 80:202-210.

[16]GlabK, WehrmeyerG, ThewesM, et al., 2024. Predictive machine learning in earth pressure balanced tunnelling for main drive torque estimation of tunnel boring machines. Tunnelling and Underground Space Technology, 146:105642.

[17]GuX, WangL, OuQ, et al., 2023. Reliability assessment of rainfall-induced slope stability using Chebyshev–Galerkin–KL expansion and Bayesian approach. Canadian Geotechnical Journal, 60(12):1909-1922.

[18]HasanipanahM, Noorian-BidgoliM, Jahed ArmaghaniD, et al., 2016. Feasibility of PSO-ANN model for predicting surface settlement caused by tunneling. Engineering with Computers, 32(4):705-715.

[19]HuAF, LiT, ChenY, et al., 2021. Deep learning for preprocessing of measured settlement data. Journal of Hunan University (Natural Sciences), 48(9):43-51 (in Chinese).

[20]HuAF, XieSL, LiT, et al., 2023. Soil parameter inversion modeling using deep learning algorithms and its application to settlement prediction: a comparative study. Acta Geotechnica, 18(10):5597-5618.

[21]HuangZK, ZhangDM, XieXC, 2022. A practical ANN model for predicting the excavation-induced tunnel horizontal displacement in soft soils. Underground Space, 7(2):278-293.

[22]JinYF, YinZY, ZhouWH, et al., 2019a. Bayesian model selection for sand with generalization ability evaluation. International Journal for Numerical and Analytical Methods in Geomechanics, 43(14):2305-2327.

[23]JinYF, YinZY, ZhouWH, et al., 2019b. Identifying parameters of advanced soil models using an enhanced transitional Markov chain Monte Carlo method. Acta Geotechnica, 14(6):1925-1947.

[24]JonesDR, 2001. A taxonomy of global optimization methods based on response surfaces. Journal of Global Optimization, 21(4):345-383.

[25]LesterAM, KouretzisGP, SloanSW, 2019. Finite element modelling of prefabricated vertical drains using 1D drainage elements with attached smear zones. Computers and Geotechnics, 107:235-254.

[26]LewisRJ, 2000. An introduction to classification and regression tree (CART) analysis. Annual Meeting of the Society for Academic Emergency Medicine in San Francisco, California.

[27]LiuMP, ZhuangPZ, LaiFW, 2024. A Bayesian optimization-genetic algorithm-based approach for automatic parameter calibration of soil models: application to clay and sand model. Computers and Geotechnics, 176:106717.

[28]MaZC, HeXH, YanPC, et al., 2023. A fast and flexible algorithm for microstructure reconstruction combining simulated annealing and deep learning. Computers and Geotechnics, 164:105755.

[29]MahmoodzadehA, MohammadiM, DaraeiA, et al., 2020. Forecasting maximum surface settlement caused by urban tunneling. Automation in Construction, 120:103375.

[30]MengJJ, MattssonH, LaueJ, 2021. Three-dimensional slope stability predictions using artificial neural networks. International Journal for Numerical and Analytical Methods in Geomechanics, 45(13):1988-2000.

[31]PanY, WuMZ, ZhangLM, et al., 2023. Time series clustering-enabled geological condition perception in tunnel boring machine excavation. Automation in Construction, 153:104954.

[32]QiCC, TangXL, 2018. Slope stability prediction using integrated metaheuristic and machine learning approaches: a comparative study. Computers & Industrial Engineering, 118:112-122.

[33]SnoekJ, LarochelleH, AdamsRP, 2012. Practical Bayesian optimization of machine learning algorithms. Proceedings of the 26th International Conference on Neural Information Processing Systems, p.2951-2959.

[34]SridharanA, MurthyNS, PrakashK, 1987. Rectangular hyperbola method of consolidation analysis. Géotechnique, 37(3):355-368.

[35]TanBK, WangD, ShiJL, et al., 2024. Temperature field prediction of steel-concrete composite decks using TVFEMD-stacking ensemble algorithm. Journal of Zhejiang University-SCIENCE A, 25(9):732-748.

[36]TangLB, NaS, 2021. Comparison of machine learning methods for ground settlement prediction with different tunneling datasets. Journal of Rock Mechanics and Geotechnical Engineering, 13(6):1274-1289.

[37]VictoriaAH, MaragathamG, 2021. Automatic tuning of hyperparameters using Bayesian optimization. Evolving Systems, 12(1):217-223.

[38]WangJJ, VanapalliS, 2024. A framework for estimating the matric suction in unsaturated soils using multiple artificial intelligence techniques. International Journal for Numerical and Analytical Methods in Geomechanics, 48(11):2854-2879.

[39]WangLQ, WangL, ZhangWG, et al., 2024. Time series prediction of reservoir bank landslide failure probability considering the spatial variability of soil properties. Journal of Rock Mechanics and Geotechnical Engineering, 16(10):3951-3960.

[40]WijesingheDR, DysonA, YouG, et al., 2022. Simultaneous slope design optimisation and stability assessment using a genetic algorithm and a fully automatic image-based analysis. International Journal for Numerical and Analytical Methods in Geomechanics, 46(15):2868-2892.

[41]WuSB, YangZF, DingXL, et al., 2020. Two decades of settlement of Hong Kong International Airport measured with multi-temporal InSAR. Remote Sensing of Environment, 248:111976.

[42]XieSL, HuAF, XiaoZR, et al., 2024. PINN-based approach to the consolidation analysis of visco-elastic soft soil around twin tunnels. Tunnelling and Underground Space Technology, 153:105981.

[43]XieSL, HuAF, WangMH, et al., 2025. 1DCNN-based prediction methods for subsequent settlement of subgrade with limited monitoring data. European Journal of Environmental and Civil Engineering, 29(4):759-784.

[44]YaoYP, HuangJ, WangND, et al., 2020. Prediction method of creep settlement considering abrupt factors. Transportation Geotechnics, 22:100304.

[45]YeXW, JinT, ChenYM, 2022. Machine learning-based forecasting of soil settlement induced by shield tunneling construction. Tunnelling and Underground Space Technology, 124:104452.

[46]YeXW, ZhangXL, ChenYB, et al., 2024. Prediction of maximum upward displacement of shield tunnel linings during construction using particle swarm optimization-random forest algorithm. Journal of Zhejiang University-SCIENCE A, 25(1):1-17.

[47]YinZY, JinYF, ShenJS, et al., 2018. Optimization techniques for identifying soil parameters in geotechnical engineering: comparative study and enhancement. International Journal for Numerical and Analytical Methods in Geomechanics, 42(1):70-94.

[48]ZhangDM, ShenYM, HuangZK, et al., 2022. Auto machine learning-based modelling and prediction of excavation-induced tunnel displacement. Journal of Rock Mechanics and Geotechnical Engineering, 14(4):1100-1114.

[49]ZhangLM, WuXG, JiWY, et al., 2017. Intelligent approach to estimation of tunnel-induced ground settlement using wavelet packet and support vector machines. Journal of Computing in Civil Engineering, 31(2):04016053.

[50]ZhangP, YinZY, JinYF, et al., 2021. Intelligent modelling of clay compressibility using hybrid meta-heuristic and machine learning algorithms. Geoscience Frontiers, 12(1):441-452.

[51]ZhangQ, SuQ, ZhangZY, et al., 2024. Prediction of high-embankment settlement combining joint denoising technique and enhanced GWO-ν-SVR method. Journal of Rock Mechanics and Geotechnical Engineering, 16(1):317-332.

[52]ZhangRH, LiYQ, GohATC, et al., 2021. Analysis of ground surface settlement in anisotropic clays using extreme gradient boosting and random forest regression models. Journal of Rock Mechanics and Geotechnical Engineering, 13(6):1478-1484.

[53]ZhangWG, WuCZ, ZhongHY, et al., 2021. Prediction of undrained shear strength using extreme gradient boosting and random forest based on Bayesian optimization. Geoscience Frontiers, 12(1):469-477.

[54]ZhangWG, LiuSL, WangLQ, et al., 2024. The overall stability of a partially unstable reservoir bank slope to water fluctuation and rainfall based on Bayesian theory. Landslides, 21(8):2021-2032.

[55]ZhouZ, ChenY, LiuZZ, et al., 2020. Theoretical prediction model for deformations caused by construction of new tunnels undercrossing existing tunnels based on the equivalent layered method. Computers and Geotechnics, 123:103565.