Follower-embedded plate anchors are advantageous relative to drag-embedded anchors as installation depth is controlled, reducing uncertainty on anchor capacity. After installation the plate is vertical in the seabed, but orientates towards the loading direction as mooring line tension develops. Plate anchor translations and rotations during this ‘keying’ process and during subsequent ‘post-keyed’ environmental loading causes excess pore pressure accumulation in the soil around the anchor. In soils that contract during shearing, this pore pressure generation will reduce soil strength, although over time consolidation leads to subsequent strength increases that increase anchor capacity.This paper introduces an effective stress macro-element model that calculates changes in plate anchor capacity due to excess pore pressure generation and dissipation. The model works iteratively between two calculation modules: (1) a macro-element module that generates the anchor kinematics and anchor resistance, and (2) a critical-state based effective stress framework that calculates accumulated plastic shear strain, from which excess pore pressure, and in turn, effective stress and undrained shear strength is calculated.The merit of this effective-stress macro-element model is demonstrated through comparisons with results from finite element simulations and experiments of plate anchor keying in clay that included pauses during keying to allow for consolidation. The model is shown to capture both the degradation in anchor resistance associated with excess pore-pressure accumulation during anchor keying and also the subsequent recovery in anchor resistance due to consolidation. As also demonstrated by the experiments, the model predicts greater anchor capacities for the longer consolidation durations, highlighting how time-dependent strength evolution can be incorporated to more reliably calculate plate-anchor capacity over the design life of an asset.
The effective thermal conductivity of sand–fines mixtures is controlled by fines content, packing state, pore-fluid conductivity, stress, and the conductivity contrast between particle fractions, yet the particle-scale mechanisms linking these factors to the macroscopic response remain unclear. This study develops a generalized-contact thermal DEM framework for gap-graded mixtures, in which the conductive network includes both real contacts and near-contact particle pairs connected through narrow pore-fluid gaps. Heat-transfer contributions are decomposed into coarse–coarse, coarse–fine, and fine–fine pathways. A generalized-contact structural potential is introduced to characterize the combined effects of particle-pair size, separation, orientation, and network connectivity. At the composition-specific densest states, effective thermal conductivity first increases and then decreases with fines content. The initial increase is driven by improved packing and the development of coarse–fine GC-SFS thermal bridges, whereas the subsequent decrease is associated with the loss of coarse-skeleton continuity and the reduced efficiency of the fines-dominated network. Under low pressure, GC-SFS conduction dominates over real solid–solid contact conduction. Moreover, the thermal network evolves more gradually than the force-chain network because near-contact particle pairs can transfer heat without carrying substantial force. By combining the structural potential with corrections for pore-fluid conductivity, pressure-enhanced contact conduction, and fine-particle conductivity, an explicit predictive model is established. After limited calibration, the model reproduces two published sand–fines datasets with mean absolute percentage errors of 4.66% and 4.68%, demonstrating its potential for mechanism-informed prediction across different mixture conditions.
A nonlinear shaft-plate coupled load transfer model is proposed for axially compressed helical piles. The novelty of the model lies in treating the helix plate as a discrete internal bearing interface embedded within a continuous shaft load transfer system, so that the axial-force jump, displacement compatibility, and nonlinear resistance mobilization are solved simultaneously within a single boundary value problem. Hyperbolic load–displacement relations are adopted for both the distributed shaft resistance and the localized plate bearing, with the initial stiffness parameters ks0 and Kb0 derived from the same elastic continuum solution to ensure internal consistency within a unified framework. Displacement compatibility and axial force jump at each plate location complete the nonlinear boundary value problem, which is solved through an iterative finite-difference scheme updating equivalent secant stiffness until convergence on both plate load and displacement field. The pile head response, axial force distribution, shaft friction mobilization, plate bearing ratio, and interface stiffness degradation are obtained. Parametric studies reveal that pile geometry, helix embedment depth, and soil stiffness gradient collectively govern the nonlinear load transfer characteristics. Within the investigated parameter range, increasing load promotes a progressive redistribution of axial resistance from the shaft toward the helix plate. The extent of this transition depends on the relative shaft-plate strength and stiffness, rather than being a fixed response mode. Both shaft friction mobilization and interface stiffness degradation exhibit a spatially non-monotonic distribution, peaking in a primary transfer zone above the helix plate and diminishing below. This depth-zoning pattern is induced by plate-mediated load diversion and is absent in conventional straight piles.
The shear behavior of rock joints is governed by joint roughness, normal stress, and specimen size. Prior studies characterize roughness with a single index, obscuring the distinct mechanical contributions of waviness and unevenness across varying normal stresses and sizes. This study conducted a systematic program of numerical direct shear simulations to quantify differentiated contributions of waviness and unevenness to shear strength. Equivalent specimens of multiple sizes were generated using a statistical equivalence method constrained by power spectral density, and joint morphology was decomposed into waviness and unevenness through wavelet transform. Results show that the unevenness-related strength component increases linearly with normal stress, whereas the waviness-related strength component shows a decline at normal stresses of 50% and 80% UCS. With coupled increases in normal stress and specimen size, the relative contribution of unevenness to peak shear strength rises progressively and exceeds 94% at 0.3 m under normal stress of 80% UCS, while the relative contribution of waviness correspondingly decreases. The roughness angle associated with waviness exhibits negative size dependence, whereas the unevenness becomes nearly size-invariant at 0.2 m. Microcrack statistics reveal a transition from tensile-dominated cracking at low normal stress to increasingly shear-dominated failure at higher normal stress, whereas energy dissipation remains consistently dominated by shear. These findings provide a controlled numerical basis for interpreting the differentiated contributions of waviness and unevenness.
In this study, a coupled numerical approach combining the discrete element method (DEM) and the finite difference method (FDM) was developed to simulate the cone penetration test (CPT) in a virtual calibration chamber (VCC), with the objective of investigating the factors influencing cone tip resistance qc, sleeve friction fs, and friction ratio Rf. The microscopic parameters of Fontainebleau sand were calibrated by matching the simulated macroscopic stress–strain behavior with experimental results from drained triaxial tests. Based on the calibrated model, the effects of VCC dimensions, boundary conditions and stress state on qc and fs were systematically evaluated. Additionally, the influence of horizontal confining pressure σ3 on Rf was analyzed. An artificial neural network (ANN) model was trained to predict the qc, fs, and Rf as functions of relative density Dr, p, particle Young’s modulus E and peak internal friction angle ϕpeak. The results demonstrate that the computational time of DEM-FDM simulation is significantly lower than that of DEM. The size effect in the coupled DEM-FDM simulations becomes negligible when the DEM and FDM radial dimensions are set to 8 and 12 times the cone diameter, respectively. The contours of displacement and velocity in the simulation show good continuity. The reliability of the coupled method is validated by the close agreement between the simulated radial stress σr and predictions from cavity expansion method. Finally, the trained ANN model exhibits a satisfactory predictive capability, showing reasonable agreement between the predicted and measured values of qc, fs, and Rf.
Global urbanization accelerates the expansion of highway and railway networks into climate-sensitive permafrost regions, where freeze–thaw-related processes induce road distresses and impair critical infrastructure. Such damage primarily arises from asymmetric temperature and deformation due to differential radiation absorption across embankment surfaces. However, the physical mechanisms linking solar radiation, atmospheric radiation and turbulent heat to subsurface freeze–thaw processes remain a challenging problem, thus hindering accurate assessment and effective mitigation of infrastructure instability in these regions. To address this scientific issue, a novel thermo-hydro-mechanical framework coupling land–atmosphere interactions, abbreviated as PerLA, is developed for rationally evaluating the performance of permafrost highway and railway embankments. This model implicitly integrates subsurface processes with land surface balance through two-way coupling, enabling the simulation of the multi-dimensional thermo-hydro-mechanical dynamics under sub-daily resolution meteorological forcing in frozen soil engineering. Both the radiative and turbulent heat fluxes are taken into account, coupled with the ground conductive heat, and treated as an upper boundary condition for the subsurface system. The subsurface model involving the ground heat transfer, water–ice phase transition, moisture migration, ice segregation, frost heave and thaw settlement is constructed based on energy, mass and momentum conservation equations. A novel soil freezing characteristic curve considering ice lens growth is adopted. Both modules are validated against field and laboratory observations. The shady-sunny slope effect of a permafrost embankment is also reproduced based on meteorological and geotechnical inputs using the proposed model. Numerical results reveal that climate warming drives the permafrost table decline and progressive subsidence of the embankment. Besides, the sunny slope exhibits more pronounced settlement than the shady slope, leading to marked asymmetric deformation of the embankment.
A phase field model based on an orthogonality-based energy split and incorporating the effect of in-situ stress is proposed to simulate hydraulic fracturing in transversely isotropic rocks under complex stress conditions. The driving force is reconstructed through volumetric–deviatoric decomposition in a transformed strain space, thereby rigorously satisfying the energy orthogonality condition required for tension–compression asymmetry in anisotropic materials. Meanwhile, the influence of in-situ stress field is incorporated into the energy functional. Within the framework of Biot poroelasticity and phase field fracture theory, the proposed PFM is formulated and validated through a series of numerical examples and experimental comparisons. Then the fracture propagation mechanisms under bedding anisotropy, perforation angle, and in-situ stress difference are then systematically discussed. The anisotropy of the critical energy release rate exerts the strongest control on the fracture trajectory, followed by elastic modulus anisotropy, whereas permeability anisotropy primarily governs pore-pressure redistribution and peak pressure response despite its weaker influence on the final fracture path.
In modern tunneling engineering, accurately predicting the surrounding rock temperature field is important for mitigating freeze–thaw hazards in cold-region tunnels and optimizing heat exchange efficiency in energy tunnels. This paper derives two analytical methods based on the separation of variables and the Laplace integral transform to determine the temperature field of tunnel surrounding rock in a two-phase medium. To simplify the two-phase heat transfer problem, three established equivalent models are systematically incorporated. Analytical solutions are then derived under both the first and third type boundary conditions. The derived solutions are validated against a stochastic porous medium finite element model constructed via a COMSOL-MATLAB linkage, as well as field measurements. Parametric analyses are finally conducted. Numerical results indicate that while the Laplace transform method is slightly less computationally efficient than the separation of variables, it offers greater versatility for complex boundary conditions. Furthermore, the Wiener or Maxwell-Eucken models are preferred for their clear physical insights under low porosity conditions, whereas a collective evaluation of multiple equivalent models is advised for high porosity. Porosity is found to primarily govern the temperature field within the near to mid field zone. The main contributions in this study lie in incorporating established equivalent models into transient tunnel analytical solutions while systematically evaluating their applicability, and developing a stochastic porous medium finite element benchmark. What’s more, the validity of the assumed degeneration from the third type to the first type boundary condition is examined, and three Laplace inversion methods are also compared for computational performance.
