Computers and Geotechnics最新文献

{"title":"Joint spatial variability of depth to bedrock, hydraulic, and shear strength parameters in 3D landslide prediction","authors":"David Sebastian Calpa ,&nbsp;Guilherme J.C. Gomes ,&nbsp;Euripedes A. Vargas Jr. ,&nbsp;Raquel Q. Velloso ,&nbsp;Fabricio Fernández ,&nbsp;Marcelo Miqueletto","doi":"10.1016/j.compgeo.2025.107662","DOIUrl":"10.1016/j.compgeo.2025.107662","url":null,"abstract":"<div><div>Landslides in unsaturated slopes result from complex and uncertain interactions among rainfall infiltration, soil hydraulic and mechanical parameters, and subsurface geometry. While these factors have been widely studied in isolation, their combined influence under fully three-dimensional and transient conditions remains insufficiently explored. This study develops and applies an integrated probabilistic framework for slope stability analysis that explicitly incorporates the spatial variability of shear strength, hydraulic parameters, and bedrock surface geometry. The framework combines the Stepwise Covariance Matrix Decomposition technique for random field generation, transient pore-pressure simulation via Richards equation, and Numerical Limit Analysis to evaluate failure mechanisms. It is applied to a tropical hillslope in Rio de Janeiro, Brazil, using site-specific geotechnical and hydrological data. Slope stability is evaluated at multiple time steps during a 22-day extreme rainfall event, following a 180-day spin-up period to establish realistic initial conditions. Results reveal that bedrock variability exerts dominant control on pore-pressure evolution, while spatial fluctuations in cohesion primarily govern failure extent and timing. Initial failures occur in zones characterized by low cohesion, low hydraulic conductivity, and thin soil cover. These findings underscore the importance of jointly modeling multiple sources of spatial uncertainty to improve the reliability of landslide hazard assessments in tropical environments.</div></div>","PeriodicalId":55217,"journal":{"name":"Computers and Geotechnics","volume":"189 ","pages":"Article 107662"},"PeriodicalIF":6.2,"publicationDate":"2025-09-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145158738","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":"Improved micro-elastoplastic constitutive model based on peridynamics and its numerical verification","authors":"Zixiang Yao ,&nbsp;Zhiliang Wang ,&nbsp;Jianguo Wang ,&nbsp;Zhitang Lu","doi":"10.1016/j.compgeo.2025.107664","DOIUrl":"10.1016/j.compgeo.2025.107664","url":null,"abstract":"<div><div>This study proposes an improved micro-elastoplastic (IME) constitutive model within the bond-based peridynamic framework to tackle the challenges in characterizing the nonlinear mechanical behavior and fracture patterns of rock materials. Firstly, a statistical damage model based on the Weibull distribution was integrated to simulate the spatial distribution of natural microcracks. Moreover, a dual-parameter bond stiffness formulation was introduced to overcome the inherent limitation of the fixed Poisson’s ratio in conventional bond-based peridynamics. The validity and robustness of the proposed IME model were subsequently demonstrated through three representative numerical simulations: uniaxial compression of intact sandstone, cyclic loading–unloading of marble, and uniaxial compression of pre-fissured sandstone. The results show that the IME model can accurately reproduce the complete stress–strain responses of sandstone under uniaxial compression, including the elastic stage, strain hardening, and post-peak softening, while also capturing the characteristic hysteresis effect during cyclic loading and unloading of marble. Furthermore, the model effectively describes damage distribution in fractured rock under uniaxial compression and highlights the significant influence of fissure angle on peak strength. The simulation outcomes exhibit strong agreement with experimental results in terms of stress–strain relationships, crack propagation paths, and final failure patterns. Overall, the proposed IME model provides a solid theoretical framework and a reliable numerical tool for simulating the multi-scale mechanical behavior of rock materials.</div></div>","PeriodicalId":55217,"journal":{"name":"Computers and Geotechnics","volume":"189 ","pages":"Article 107664"},"PeriodicalIF":6.2,"publicationDate":"2025-09-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145158740","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":"Quasi-static modeling with explicit numerical manifold method via timestep scaling technique and velocity-based viscous damping model","authors":"Junfeng Li ,&nbsp;Yongtao Yang ,&nbsp;Xuhai Tang ,&nbsp;Yun Zheng ,&nbsp;Shuilin Wang","doi":"10.1016/j.compgeo.2025.107663","DOIUrl":"10.1016/j.compgeo.2025.107663","url":null,"abstract":"<div><div>An explicit numerical manifold method (NMM) is developed to solve quasi-static continuous and discontinuous geotechnical engineering problems. However, due to the small timestep size for explicit time integration scheme, the number of calculation steps is usually very large, and the analysis process may be very time-consuming, especially for large-scale rock and soil masses. As the extension to the mass-scaling technique, a timestep-based mass-scaling technique (timestep scaling technique) is introduced to explicit NMM in this study to model quasi-static geotechnical problems more effectively. This technique scales the lumped mass matrix through the formula of critical timestep size <span><math><mrow><mi>Δ</mi><msub><mi>t</mi><mi>c</mi></msub></mrow></math></span> directly, and <span><math><mrow><mi>Δ</mi><msub><mi>t</mi><mi>c</mi></msub></mrow></math></span> can be increased to be larger than the required timestep size <span><math><mrow><mi>Δ</mi><mi>t</mi></mrow></math></span>. In addition, to dissipate the kinetic energy of the system, a velocity-based viscous damping model is used to weaken the dynamic response and accelerate the convergence speed to reach the final state of the quasi-static problem. Through a series of verification and application examples, it has been proven that the timestep scaling technique and velocity-based viscous damping model in the context of the explicit NMM can efficiently solve quasi-static problems.</div></div>","PeriodicalId":55217,"journal":{"name":"Computers and Geotechnics","volume":"189 ","pages":"Article 107663"},"PeriodicalIF":6.2,"publicationDate":"2025-09-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145158739","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 semi-analytical solution for the load-transfer behaviors of precast concrete pile reinforced by cemented soil with long-core in layer soils","authors":"Chaozhe Zhang ,&nbsp;Roberto Cudmani ,&nbsp;Songyu Liu ,&nbsp;Dingwen Zhang ,&nbsp;Zhen Wang ,&nbsp;Pan Zhou","doi":"10.1016/j.compgeo.2025.107665","DOIUrl":"10.1016/j.compgeo.2025.107665","url":null,"abstract":"<div><div>Precast Concrete Piles Reinforced by Cemented Soil (PCCS), a novel composite pile, has been widely employed for soft ground treatment. In this technique, the precast concrete (PC) pile is installed in the center of a deep mixing (DM) column, thereby effectively harnessing the benefits of both technologies. However, modelling the load transfer of the composite pile is quite complex due to the different materials (soft soil, cemented soil and concrete) and interfaces (soft soil-cemented soil, soft soil-concrete, cemented soil-pile) involved. Particularly, studies on the axial load-transfer behavior of the PCCS with long-core (inner core), in which the PC piles is significant longer than the DM columns, remain limited. This contribution proposes an 1D semi-analytical load transfer model for the PCCS with long-core in layered soils, which takes into account the shear force transfer at the different interfaces and the mechanical behavior of the natural soil, cemented soil, the pile and the interface. By comparing and analyzing the mechanical responses of PCCSs under loads applied to the inner core head (loading mode I) and the entire cross-section of the PCCSs (loading mode II), the ability of the model to describe the experimental behavior is investigated. The results indicate that the distribution of the axial force along the pile shaft and the skin friction resistance at the interfaces of the PCCS with long-core can be realistically reproduced by the model. It is observed that the ultimate bearing capacity of the PCCS under loading mode II is about 15% higher than that under loading mode I.</div></div>","PeriodicalId":55217,"journal":{"name":"Computers and Geotechnics","volume":"189 ","pages":"Article 107665"},"PeriodicalIF":6.2,"publicationDate":"2025-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145158743","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 closed-form model of saline soil freezing curve developed using symbolic regression and its numerical implementation","authors":"Yandi Wu ,&nbsp;Yijie Wang ,&nbsp;Liming Hu","doi":"10.1016/j.compgeo.2025.107659","DOIUrl":"10.1016/j.compgeo.2025.107659","url":null,"abstract":"<div><div>The soil freezing characteristic curve (SFCC) plays a crucial role in modeling coupled water-heat-salt transport in saline frozen soils. However, existing SFCC models are either empirical, lacking the ability to capture dynamic salt transport, or theoretically based but too complex for numerical implementation due to intricate formulations or the requirement for iterative algorithms and many parameters. To overcome these limitations, this study employs symbolic regression to derive a closed-form SFCC model, trained on synthetic data generated by a theoretical model grounded in the pore water freezing theory. The proposed model performs well in reflecting theoretical results across various soil types and initial water contents, providing a concise yet physically meaningful representation of the pore solution freezing mechanisms. The reliability of the model is demonstrated by comparison with 26 sets of experimental SFCC data. Furthermore, the model is simple in form and is successfully applied to simulate the coupled water-heat-salt transport during one-dimensional freezing. The simulations reasonably capture frost heave, temperature evolution, and water content variation under different initial and boundary conditions, in agreement with experimental results. The findings indicate that increasing boundary salinity elevates internal salt concentration and depresses the local freezing point, leading to an upward shift (closer to the cold end) of the freezing front. Meanwhile, ice formation causes salt accumulation at the freezing front. The proposed closed-form SFCC model allows for time-varying salt concentration in soils to be incorporated into fully coupled moisture and temperature fields, offering both strong theoretical grounding and promising potential for numerical applications.</div></div>","PeriodicalId":55217,"journal":{"name":"Computers and Geotechnics","volume":"189 ","pages":"Article 107659"},"PeriodicalIF":6.2,"publicationDate":"2025-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145158742","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":"Three–dimensional continuous–discontinuous pipe network method for simulating fractured and porous seepage","authors":"Hanxun Wang ,&nbsp;Zhe Sun ,&nbsp;Bin Zhang","doi":"10.1016/j.compgeo.2025.107656","DOIUrl":"10.1016/j.compgeo.2025.107656","url":null,"abstract":"<div><div>Fractured seepage and porous seepage are critical factors affecting the safety of geotechnical structures in tunneling, slope engineering, and underground energy storage projects. On the basis of the pipe network method (PNM), the three–dimensional continuous–discontinuous pipe network method (C–DPNM^3D) with high grid adaptability is developed to address the problem of insufficient simulation efficiency of fractured seepage and porous seepage. C–DPNM^3D uses Darcy’s law and the mass conservation law as the control equations and combines the first and second boundary conditions. The simulation results of this method are compared with those of the typical finite element method (FEM) and discrete element method (DEM), and its feasibility is confirmed. This method is applied to underground water–sealed oil storage in China. The results reveal that compared with those of the FEM and DEM, the error rate of C–DPNM^3D in calculating three–dimensional large–scale porous seepage and fractured seepage is less than 4.34%, and the computational efficiency increases by more than 99.90%. The large–scale three–dimensional fractured seepage and porous seepage is simulated efficiently, and the application effect in the field of underground water–sealed oil storage is good. C–DPNM^3D is a universal, reliable and efficient method for seepage calculation of underground rock mass structures.</div></div>","PeriodicalId":55217,"journal":{"name":"Computers and Geotechnics","volume":"189 ","pages":"Article 107656"},"PeriodicalIF":6.2,"publicationDate":"2025-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145121221","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 flexible membrane boundary method based on color function in DEM simulations for triaxial tests of granular materials","authors":"Hongze Li ,&nbsp;Chun Feng ,&nbsp;J.L. Feng","doi":"10.1016/j.compgeo.2025.107657","DOIUrl":"10.1016/j.compgeo.2025.107657","url":null,"abstract":"<div><div>This technical note presents a novel flexible membrane algorithm for DEM-based triaxial test simulations. Using color functions to calculate the normal vectors of membrane particles, the algorithm can model mechanical response of flexible membrane under large deformations and curvatures. It also uses imaginary particles to improve calculation accuracy of normal vector nearby high curvature area. After validation based on lab test result, the algorithm effectively reproduces the stress–strain behavior of specimens under triaxial compression, offering a more precise and reliable tool for studying the mechanical response of granular materials.</div></div>","PeriodicalId":55217,"journal":{"name":"Computers and Geotechnics","volume":"189 ","pages":"Article 107657"},"PeriodicalIF":6.2,"publicationDate":"2025-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145121220","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":"Physics-informed neural network modeling of one-dimensional wave propagation under broadband and long-duration incident waves","authors":"Hamid Taghavi Ganji,&nbsp;Elnaz Seylabi","doi":"10.1016/j.compgeo.2025.107652","DOIUrl":"10.1016/j.compgeo.2025.107652","url":null,"abstract":"<div><div>This paper presents a physics-informed neural network (PINN) to model one-dimensional wave propagation in visco-elastic media for seismic site response analysis. Dimensionless relations are introduced to effectively scale the loss terms associated with the wave equation, as well as initial and boundary conditions, during PINN training. Several problem settings with varying complexities demonstrate the effectiveness of the proposed approach compared to the standard (vanilla) PINN and those enhanced with self-adaptive weighting and Fourier feature strategies for handling broadband frequency wave propagation problems with real-world mechanical properties. The results show the dimensionless PINN has less than 9% normalized error in all cases, while the other variants had at least 40% error. Additionally, we present how sequential transfer learning over short intervals can be used to reduce the number of iterations required for the same problem by half, where new boundary conditions are applied to the system. This approach, combined with spatial domain decomposition, can enhance the prediction accuracy of wave responses in layered media subjected to long-duration incident waves, such as earthquake ground motions.</div></div>","PeriodicalId":55217,"journal":{"name":"Computers and Geotechnics","volume":"189 ","pages":"Article 107652"},"PeriodicalIF":6.2,"publicationDate":"2025-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145121222","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":"Modelling the lateral behavior of trunk-root-soil systems with representative branching patterns using a combined p-y framework","authors":"Fan He ,&nbsp;Teng Liang ,&nbsp;Rui Zhao ,&nbsp;Tianfeng Zhu ,&nbsp;Yu Zhao ,&nbsp;Anthony Kwan Leung ,&nbsp;Liangtong Zhan","doi":"10.1016/j.compgeo.2025.107654","DOIUrl":"10.1016/j.compgeo.2025.107654","url":null,"abstract":"<div><div>The Beam-on-Non-linear-Winkler-foundation (BNWF) model using <em>p</em>-<em>y</em> formulation is widely used on laterally loaded structures (e.g., piles) due to its computational efficiency. However, its applicability to branched structural systems (e.g., trunk-root-soil systems) remains highly uncertain given their intricate branching geometries and proclivity for substantial deflections, which contrast with conventional pile structures. This work tries to address this issue through developing a combined <em>p</em>-<em>y</em> framework. Initially, physical model tests were conducted using four 3D-printed representative branched root architectures—tap, herringbone, dichotomous, and T-shape patterns—under dry and saturated sand conditions, thereby providing a comprehensive dataset for numerical model development. Subsequently, the <em>p</em>-<em>y</em> framework was established by considering the large deflection response of parent roots and the push-in/uplift behavior of branch roots during tree overturning. For modeling large deflection response, two types of <em>p</em>-<em>y</em> formulations were critically evaluated: the commonly used Reese <em>p</em>-<em>y</em> curve within previous studies and the hyperbolic <em>p</em>-<em>y</em> curve. For push-in/uplift behavior, the deep flow mechanism and a bilinear softening trend were considered, respectively. The proposed <em>p-y</em> framework was integrated into an efficient hybrid beam-FE model to predict the lateral response of the trunk-root-soil systems. Model validation demonstrated that the Reese <em>p</em>-<em>y</em> curve inadequately captures large deflection responses due to its inherent assumptions. Conversely, the hyperbolic <em>p</em>-<em>y</em> curve effectively predicts lateral resistance and root behavior in branched root systems. Compared to other analytical models, the proposed model enhances the capability to predict the overturning behavior of tree structures and enables the modeling of complex 3D root system geometries and root material variability, offering a potentially effective method for evaluating tree stability in practical applications.</div></div>","PeriodicalId":55217,"journal":{"name":"Computers and Geotechnics","volume":"189 ","pages":"Article 107654"},"PeriodicalIF":6.2,"publicationDate":"2025-09-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145109585","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":"High-Fidelity Morphological and Mineralogical Reconstruction of Chang’E-5 Lunar Regolith Grains","authors":"Quan Zheng ,&nbsp;Ziwei Tian ,&nbsp;Songzheng Yu ,&nbsp;Ronghua Pang ,&nbsp;Guang Zhang ,&nbsp;Guanghui Liu ,&nbsp;Yiwei Liu ,&nbsp;Yang Li ,&nbsp;Xin Liu ,&nbsp;Shijing He ,&nbsp;Ran Niu ,&nbsp;Peng Zhang","doi":"10.1016/j.compgeo.2025.107661","DOIUrl":"10.1016/j.compgeo.2025.107661","url":null,"abstract":"<div><div>The complex structure and physicochemical properties of lunar soil particles form a critical foundation for future lunar exploration and in-situ resource utilization. To address the distortion introduced by oversimplified particle morphology and composition assumptions in conventional modeling approaches, we propose a high-fidelity morphological and mineralogical reconstruction method of Chang’E-5 lunar regolith grains using micro-CT characterization and advanced image processing. Twenty-one representative lunar particles were selected and subjected to high-resolution X-ray computed tomography (X-CT) scanning for comprehensive characterization of their 3D internal structures. A segmentation workflow was developed to achieve precise particle identification and mesh model generation, integrating unsupervised clustering, morphological optimization, and voxel-to-mesh conversion. Scanning electron microscopy and energy-dispersive spectroscopy were employed to identify mineral phases and reconstruct their spatial distribution within the particles. Based on these characterizations, models applicable to the discrete element method and finite element method were constructed. This work forms a comprehensive digital lunar particle dataset encompassing optical images, X-CT 3D images, mesh models, discrete element and finite element models. This integrated approach enables high-fidelity representation of morphology, internal structure, and mineralogical composition of particles, significantly enhancing modeling accuracy and simulation scalability compared to previous methods. The digital model repository established in this study provides standardized and extensible data resources to support mechanical simulations, in-situ resource evaluation, and the design of lunar detection devices.</div></div>","PeriodicalId":55217,"journal":{"name":"Computers and Geotechnics","volume":"189 ","pages":"Article 107661"},"PeriodicalIF":6.2,"publicationDate":"2025-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145107014","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}