{"title":"Data-driven probabilistic seismic demand prediction and sustainability optimization of stone columns for liquefaction mitigation in regional mildly sloping ground","authors":"Zhijian Qiu , Junrui Zhu , Ahmed Ebeido , Athul Prabhakaran , Yewei Zheng","doi":"10.1016/j.compgeo.2025.107125","DOIUrl":"10.1016/j.compgeo.2025.107125","url":null,"abstract":"<div><div>With the growing need for efficient mitigation strategies in liquefaction-prone regions, ensuring both seismic resilience and sustainability of infrastructure has become increasingly significant. This paper presents a data-driven probabilistic seismic demand model (PSDM) prediction and sustainability optimization framework to mitigate liquefaction-induced lateral deformation in regional mildly sloping ground improved with stone columns. The framework integrates finite element (FE) simulations with machine learning (ML) models, generating 1,200 ground FE models based on the key site attributes, such as ground inclination, soil properties, and stone column configurations. The performance of the selected ML models is evaluated through hyperparameter tuning by <em>k</em>-fold cross-validation, with the artificial neural network (ANN) outperforming other models in accurately predicting the PSDM. Subsequently, this framework is applied to a set of representative mildly sloping ground sites, enabling rapid PSDM prediction for each site with varying site attributes. Moreover, by incorporating cost and sustainability metrics, multi-objective optimization is performed using the developed ANN predictive model to maximize seismic performance while minimizing total carbon emissions and costs associated with ground improvement. Overall, the framework allows for rapid and accurate PSDM prediction and regional optimization, facilitating the identification of the optimal stone column configurations for efficient and sustainable liquefaction mitigation.</div></div>","PeriodicalId":55217,"journal":{"name":"Computers and Geotechnics","volume":"181 ","pages":"Article 107125"},"PeriodicalIF":5.3,"publicationDate":"2025-02-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143377331","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Fan Yang , Jieya Zhang , Mingxing Xie , Wenwen Cui , Xiaoqiang Dong
{"title":"Evaluation of compression index of red mud by machine learning interpretability methods","authors":"Fan Yang , Jieya Zhang , Mingxing Xie , Wenwen Cui , Xiaoqiang Dong","doi":"10.1016/j.compgeo.2025.107130","DOIUrl":"10.1016/j.compgeo.2025.107130","url":null,"abstract":"<div><div>The annual increase in red mud emissions necessitates the expansion of bauxite residue disposal areas (BRDAs), while the escalating land value has led to proposals for development on closed BRDAs. Therefore, understanding the compressive properties of red mud is critical for the safe management and construction of BRDAs. The process of deriving compression index (<em>C<sub>c</sub></em>) through consolidation tests to assess compression characteristics is both time-intensive and vulnerable to the quality of the sampling methods employed. Consequently, it is essential to develop predictive models for compression indices that utilize more easily measurable physical parameters. This study proposes the use of machine learning(ML) models to predict the <em>C<sub>c</sub></em> of red mud. Several machine learning models were studied, including Linear Regression (LR), Ridge Regression (RR), Support Vector Machines (SVR), Random Forest(RF), Extremely Randomized Trees (Extra Trees), K-Nearest Neighbors (KNN), Category Boosting (CatBoost), and LightGBM (Light Gradient Boosting Machine). The grid search algorithm was used to obtain the optimal parameters for each ML model, and k-fold cross-validation was employed to enhance the model’s generalization performance. Ultimately, the KNN model achieved the best performance. The SHAP method was used to describe the specific influence patterns, and to provide a quantitative contribution of each feature to the <em>C<sub>c</sub></em> of red mud. The research indicated that the <em>I<sub>L</sub></em> and <em>w<sub>n</sub></em> exerted the most substantial influence on the <em>C<sub>c</sub></em>, yielding a positive effect.</div></div>","PeriodicalId":55217,"journal":{"name":"Computers and Geotechnics","volume":"181 ","pages":"Article 107130"},"PeriodicalIF":5.3,"publicationDate":"2025-02-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143377332","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"A novel soil reaction model for continuous impact pile driving","authors":"Shihong Zhang , Lizhong Wang , Mengtao Xu , Shengjie Rui , Zhen Guo","doi":"10.1016/j.compgeo.2025.107123","DOIUrl":"10.1016/j.compgeo.2025.107123","url":null,"abstract":"<div><div>Impact pile driving is widely employed in various environments. The soil surrounding driven piles undergoes large shear displacements and highly cyclic loads, leading to significant strength degradation. This paper introduces a novel soil reaction model with easily calibrated parameters to estimate the pile penetration performance under continuous impact driving, incorporating both cyclic degradation and base gap. Soil cumulative plastic displacement is utilized to quantity the degradation, enabling more accurate simulation of cyclic pile response. The model is integrated into the pile driving system and applied in multiple-blow analysis. Non-linear cumulative displacement-blow count curves are analyzed and the development of residual stress varies between the pile upper and lower sections. It is found that lower blow counts are required when cyclic degradation is considered, although the increased rebound effect may counterbalance this benefit. Comparative analyses for degradation constants further demonstrate that early-stage degradation has a more pronounced impact. Finally, the proposed model is also adopted to predict blow count in field practice, offering valuable insights for driveability analysis.</div></div>","PeriodicalId":55217,"journal":{"name":"Computers and Geotechnics","volume":"181 ","pages":"Article 107123"},"PeriodicalIF":5.3,"publicationDate":"2025-02-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143350407","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Hypoplastic model for gas hydrate-bearing sediments considering pore morphology","authors":"Sahil Wani , Ramesh Kannan Kandasami , Wei Wu","doi":"10.1016/j.compgeo.2025.107115","DOIUrl":"10.1016/j.compgeo.2025.107115","url":null,"abstract":"<div><div>A novel hypoplastic constitutive model is proposed which accurately captures the geomechanical behavior of gas hydrate-bearing sediments. This model explicitly accounts for the effects of hydrate saturation, temperature, and pore pressure across different pore morphologies. A cementation coefficient (<span><math><msub><mrow><mi>α</mi></mrow><mrow><mi>h</mi></mrow></msub></math></span>) is introduced which represents the influence of pore morphology on the sediment’s mechanical response. The predictive capabilities of the proposed model are thoroughly validated against an extensive dataset from the literature, covering various initial stress, hydrate saturation, and temperature conditions. The model’s generality is demonstrated by its ability to predict the mechanical behavior of sediments containing both methane (<span><math><msub><mrow><mi>CH</mi></mrow><mrow><mn>4</mn></mrow></msub></math></span>) and carbon dioxide (<span><math><msub><mrow><mi>CO</mi></mrow><mrow><mn>2</mn></mrow></msub></math></span>) hydrates. Furthermore, the model responses are validated using data obtained from testing natural hydrate cores from the Nankai Trough.</div></div>","PeriodicalId":55217,"journal":{"name":"Computers and Geotechnics","volume":"181 ","pages":"Article 107115"},"PeriodicalIF":5.3,"publicationDate":"2025-02-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143349931","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Probabilistic analysis of ground surface settlement induced by super large diameter shield tunneling based on 3D random finite element method","authors":"Jing-Kang Shi , Jin-Zhang Zhang , Shuai Zhao , Zhen-Chang Guan , Hong-Wei Huang","doi":"10.1016/j.compgeo.2025.107111","DOIUrl":"10.1016/j.compgeo.2025.107111","url":null,"abstract":"<div><div>The accurate prediction of ground surface settlement remains a crucial undertaking in shield tunnelling, particularly for tunnels with super large diameters. Instead of employing conventional deterministic analysis, this study employed 3D random finite element method (RFEM) to probabilistically investigate the longitudinal ground surface settlement induced by super large diameter shield tunneling. Through large quantities of numerical simulations, a new modified Logistic function with parameters of maximum ground settlement <em>S</em><sub>max</sub>, settlement upon face arrival <em>S</em><sub>0</sub>, and settlement speed during face passage <em>v</em> was proposed to represent the longitudinal ground surface settlement curves. The probabilistic distribution of longitudinal ground settlement was transformed into a multivariate Gaussian distribution of <em>S</em><sub>max</sub>, <em>S</em><sub>0</sub>, and <em>v</em>. A randomness transfer coefficient <em>η</em> was innovatively proposed to evaluate the effects of randomness transferred from modulus random field to the ground settlements. It was found that the mean values of <em>S</em><sub>max</sub>, <em>S</em><sub>0</sub>, and <em>v</em> were rarely affected by random field statistics. The randomness transferring coefficient <em>η</em> was determined by the vertical and horizontal scales of fluctuation (<em>δ</em><sub>v</sub> and <em>δ</em><sub>h</sub>). The correlation coefficients between <em>S</em><sub>max</sub>, <em>S</em><sub>0</sub>, and <em>v</em> were only sensitive to <em>δ</em><sub>h</sub>.</div></div>","PeriodicalId":55217,"journal":{"name":"Computers and Geotechnics","volume":"181 ","pages":"Article 107111"},"PeriodicalIF":5.3,"publicationDate":"2025-02-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143349932","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Xiaoming Liu , Qinji Jia , R.A. Galindo , Fengshan Mao
{"title":"Probabilistic bearing capacity analysis of square and rectangular footings on cohesive soil slopes considering three-dimensional rotational anisotropy","authors":"Xiaoming Liu , Qinji Jia , R.A. Galindo , Fengshan Mao","doi":"10.1016/j.compgeo.2025.107117","DOIUrl":"10.1016/j.compgeo.2025.107117","url":null,"abstract":"<div><div>The influence of soil variability on the probabilistic bearing capacity of strip footings near slopes has been extensively studied, particularly under short-term undrained conditions. However, these investigations, predominantly based on the plane-strain assumption, fall short in accurately estimating the bearing capacity of square and rectangular footings and in capturing the spatial variability of soils. This study focuses on short-term undrained conditions and employs the random finite element method (RFEM) and Monte Carlo simulation (MCS) techniques to explore the effect of rotational anisotropy on the bearing capacity response and failure probability of a square and rectangular footing-cohesive slope system under a three-dimensional (3D) framework. The findings reveal that the rotation angles of soil strata significantly impact both the mean and coefficient of variation of the bearing capacity, with distinct variation patterns emerging for different footing orientations and aspect ratios. Typical failure patterns are identified, illustrating the correlation between the bearing capacity response, the footing orientations and aspect ratios, and the extension direction of plasticity. The probabilistic results are presented as probability density functions (PDF) and cumulative distribution functions (CDF) for various rotation angles around the <em>x</em>-axis and <em>y</em>-axis and for different <em>L</em>/B ratios of the footings. Additionally, detailed design tables, including failure probability results and corresponding safety factors for specific target failure probabilities, are provided to guide engineering applications.</div></div>","PeriodicalId":55217,"journal":{"name":"Computers and Geotechnics","volume":"181 ","pages":"Article 107117"},"PeriodicalIF":5.3,"publicationDate":"2025-02-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143324760","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Numerical investigations into the influence of drainage on the uplift behaviour of plate anchors in clay","authors":"Maozhu Peng , Zhen-Yu Yin , Zefeng Zhou","doi":"10.1016/j.compgeo.2025.107128","DOIUrl":"10.1016/j.compgeo.2025.107128","url":null,"abstract":"<div><div>This paper presents one of the first systematic finite element investigations into the rate effect during plate anchor uplift in clay, with a focus on the influence of soil drainage. Some numerical difficulties in simulating partially/fully drained anchor uplift are first discussed, and a method to address these issues is proposed. The most striking benefit of the proposed method is that it allows the hydro-mechanical coupled anchor-soil interaction to be efficiently simulated, with improved numerical stability, by simply using a non-detachable anchor-soil interface. After validation, the proposed method is used to simulate anchor uplift at different embedment and uplift rates, spanning from drained to fully undrained conditions. The influence of uplift rate on the force–displacement relations and failure mechanisms are discussed. The threshold rates to differentiate drained, partially drained, and undrained conditions are quantified. Anchor penetrations are also simulated under the same conditions. Comparisons with uplift cases indicate that the force–displacement curves in uplift and penetration are only identical in undrained conditions.</div></div>","PeriodicalId":55217,"journal":{"name":"Computers and Geotechnics","volume":"181 ","pages":"Article 107128"},"PeriodicalIF":5.3,"publicationDate":"2025-02-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143324761","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"A novel fluid-solid coupling method for fractured reservoirs: 3D DDM-EDFM integration with proppant mechanics","authors":"Luoyi Huang , Wentao Zhan , Hui Zhao , Guanglong Sheng","doi":"10.1016/j.compgeo.2025.107127","DOIUrl":"10.1016/j.compgeo.2025.107127","url":null,"abstract":"<div><div>This paper proposes a novel numerical simulation approach for fluid–solid coupling in fractured reservoirs, incorporating the mechanical properties of proppants into the coupling framework. By integrating the Displacement Discontinuity Method (DDM) with a three-dimensional Embedded Discrete Fracture Model (3D EDFM), the proposed method enables coupled simulations of geomechanics and fluid flow. Unlike conventional approaches, this method considers the time-dependent evolution of fracture aperture and accurately captures the mechanical effects of proppants during fracture closure. A comprehensive analysis is conducted to investigate the impacts on fracture dynamics and fluid flow behavior. The model’s reliability is validated through comparisons with analytical solutions and the Extended Finite Element Method (XFEM). Results demonstrate that the time-dependent evolution of fracture apertures significantly influences fluid flow in fractured reservoirs. Specifically, during the production stage, fracture closure results in a sharp decline in oil production rates. Larger proppant-supported apertures effectively mitigate fracture closure, sustain high fracture conductivity, and prolong the production period. Additionally, low-porosity reservoirs are shown to be more sensitive to rock deformation, leading to substantial changes in flow parameters during production. Thus, incorporating fluid–solid coupling is crucial for accurate modeling in low-porosity reservoirs. This study provides valuable theoretical insights for the development of unconventional resources.</div></div>","PeriodicalId":55217,"journal":{"name":"Computers and Geotechnics","volume":"181 ","pages":"Article 107127"},"PeriodicalIF":5.3,"publicationDate":"2025-02-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143324444","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Weiwei Niu , Yuan-Yuan Zheng , Zhen-Yu Yin , Chi Yao , Pengchang Wei
{"title":"Multiscale mechanical behavior of hydrated expansive soil: Insights from experimental and MD study","authors":"Weiwei Niu , Yuan-Yuan Zheng , Zhen-Yu Yin , Chi Yao , Pengchang Wei","doi":"10.1016/j.compgeo.2025.107129","DOIUrl":"10.1016/j.compgeo.2025.107129","url":null,"abstract":"<div><div>The mechanical behavior of expansive soil in geotechnical engineering is significantly sensitive to loading rates, hydration, confining pressure, etc., where most engineering problems are attributed to the existence of montmorillonite in expansive soil. Here, the hydration, confining pressure, and loading rate effect on the mechanical behavior of montmorillonite were investigated through the triaxial tests and molecular dynamics (MD) simulation method, revealing their fundamental mechanism between the microscale and macroscale. The average basal spacing of hydrated montmorillonite system, the diffusion coefficient and density distribution of interlayer water molecules were calculated for the verification of MD model. The experimental results indicated that the stress–strain relationship of montmorillonite was the strain-hardening type. The failure stress did not increase monotonously with the increase in loading rate, and there were two obvious critical points. The failure stress of the soil sample increased with the increase of the confining pressure, and the decrease of the water content, where their fundamental mechanism between microscale and macroscale were adequately discussed. Furthermore, the stress–strain response, total energy evolution, deformation evolution of atomistic structure, and broken bonds evolution were analyzed to deeply understand the fundamental deformation mechanism at the microscale. The multi-scale studies could effectively examine the macroscopic mechanical behavior of expansive soil and elucidate its microscopic mechanisms.</div></div>","PeriodicalId":55217,"journal":{"name":"Computers and Geotechnics","volume":"181 ","pages":"Article 107129"},"PeriodicalIF":5.3,"publicationDate":"2025-02-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143325092","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"A Comprehensive Investigation of Physics-Informed Learning in Forward and Inverse Analysis of Elastic and Elastoplastic Footing","authors":"Xiao-Xuan Chen , Pin Zhang , Zhen-Yu Yin","doi":"10.1016/j.compgeo.2025.107110","DOIUrl":"10.1016/j.compgeo.2025.107110","url":null,"abstract":"<div><div>Physics-informed learning has emerged as a promising approach for solving forward and inverse partial differential equations in engineering practice, but selecting an optimal loss function remains unclear and parameter identification for inverse analysis lacks efficiency. Meanwhile, their values for engineering-scale elastoplastic problems have not been deeply investigated. In this research, a comprehensive comparison between the strong-form collocation point method (CPM) and the deep Ritz method (DRM) based loss functions for both forward and inverse analysis is conducted, and a novel exponential acceleration method is proposed to enlarge the search space of unknown parameters for inverse analysis. By applying these methods to linear elasticity and elastoplasticity footing cases, we found that physics-informed learning equipped with DRM-based loss functions shows more excellent accuracy in forwardly computing displacement but poor accuracy in predicting strain and stress. Physics-informed learning with CPM-based loss functions shows more excellent performance in inverse analysis than their forward-solving ability. The exponential acceleration method largely enhances the efficiency of inverse analysis without sacrificing accuracy. These new findings inspire the future application of physics-informed learning to engineering-scale elastoplastic problems.</div></div>","PeriodicalId":55217,"journal":{"name":"Computers and Geotechnics","volume":"181 ","pages":"Article 107110"},"PeriodicalIF":5.3,"publicationDate":"2025-02-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143324679","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}