Red clay, inherently high in liquid limit and plasticity, is unsuitable for direct subgrade construction, making green soil stabilization a critical demand for sustainable transportation infrastructure. Lignosulfonate (LS), a by-product of the papermaking industry, exhibits broad prospects for red clay stabilization. This study investigated Fujian high liquid limit red clay, conducting Atterberg limit tests under varying LS contents, mixing methods, and curing ages, combined with a series of micro-tests. SPSS multiple linear regression was employed to quantify factor influence, and high-precision prediction models were established via nonlinear fitting. Results showed that ≥ 1% LS removed soil from the high liquid limit category; optimal content was 3% for calcium lignosulfonate (CLS), 3% for sodium lignosulfonate (SLS) plasticity improvement, and 7% for SLS liquid limit reduction. No new chemical bonds are formed and no significant mineral dissolution occurs during CLS stabilization; plasticity improves via Ca2+ cation exchange and sulfonate-hydroxyl hydrogen bonding, which compresses the electric double layer and forms dense aggregates. All variables had a variance inflation factor (VIF) of 1.000, with LS content as the absolute dominant factor. The models achieved a maximum R2 of 0.997, providing a potential reference for the quantitative design of LS-stabilized red clay subgrades.
Traditional kinematic analysis methods are limited in representing fragmented discontinuities, variable slope geometries, and spatial relationships between joint surfaces. This study presents a point-cloud-based kinematic analysis framework integrating a shallow artificial neural network, DBSCAN clustering, PCA-based orientation calculation, co-planar reconstruction, and spatial proximity analysis. These components enable direct, integrated processing from point-cloud acquisition to kinematic assessment. Rock joints are identified from normal vectors and curvatures, and fragmented patches are connected using distance-based clustering. Equivalent trace lengths visualize the reconstructed structures. The framework evaluates planar, wedge, and toppling failure criteria while considering actual spatial intersections, and iteratively determines the maximum kinematically admissible excavation angle. Applications to three field datasets show that calculated joint orientations agree with manual and field measurements within 5°. The proposed framework reduces processing time by 63.5% compared with the conventional workflow. Comparison with Dips produces the same primary kinematic conclusion, while spatial filtering reduces theoretical intersection lines from 12,403 to 36 in Case C. The framework identifies joints and locations potentially associated with structurally controlled failure, but direct localization of complete unstable rock bodies requires further integration of linear rock joints, hidden internal joint networks, site-specific physical parameters, and field-based evaluation.
This work explores the phenomena of instability in low stress tilt-testing of dry sands using novel imbedded shear stress and strain sensors to capture quantitative data of constitutive behaviour leading up to and during instability. A unique strategy of testing dry soil enabled the investigation of instability in isolation from pore pressure consequences. Six shear stress and shear distortion sensors were distributed along the transparent sidewall and centre of the tilt-table, which was then rotated until instability and failure. Shear strain measured next to the transparent sidewall were 50% of those measured along the table centerline. Friction reduction strategies were successful in significantly reducing, but not eliminating, this discrepancy. Soil layers tilted until reaching instability, unloaded, and subjected to a second tilt cycle only exhibit instability at angles in excess of the first instability event. This phenomenon, analogous to the Kaiser effect, reveals the importance of void ratio with respect to instability. Tilt-table testing, when equipped with internal shear distortion sensors, was shown to be able to capture the modulus degradation curve over the strain range of 10 −5 to 10 −1 providing a promising new experimental technique to explore the stiffness degradation curve at very low confining pressures.
Field observations at the toes of loess landslides show seepage outlets, preferential flow paths, salt efflorescence, soils with high water content, and indurated crusts or blocks formed by salt precipitation, which may locally increase resistance to movement. To examine how salinity affects loess shear strength at different shear rates, ARS ring shear tests were conducted on Heifangtai loess with different salt concentrations and shear rates, combined with scanning electron microscopy (SEM) characterization. The results show that (1) shear strength first increases and then decreases with salinity, reaching a maximum at 8%; above this concentration, strength decreases and brittle behavior becomes more pronounced. Vertical settlement indicates shear-induced contraction. (2) Residual strength decreases with increasing shear rate, indicating a negative rate effect. The absolute value of the rate effect coefficient first decreases and then increases, with the strongest rate dependence at 20% salinity. (3) Low to moderate salinity promotes salt bridge cementation, flocculation, and dense fine aggregates, whereas high salinity induces salt coatings, particle breakage, poor contact quality, and connected pore networks, thereby reducing strength. (4) Higher salinity decreases zeta potential and electrokinetic mobility, compressing the electrical double layer and enhancing particle flocculation. These findings clarify the coupled salinity and rate effects on loess strength.
The in-situ stress field is a key factor influencing reservoir geological stability. High-efficiency, high-accuracy in-situ stress inversion is crucial for reservoir stability assessment and exploitation strategy optimization. To improve efficiency and accuracy, this study proposes an in-situ stress inversion method integrating numerical manifold models, efficient surrogate models, and deep reinforcement learning. It addresses traditional numerical simulation limitations in handling discontinuous interfaces and overcomes high computational costs and strong sensitivity to input conditions. The method's effectiveness is validated through an engineering application for a shale gas reservoir in the Sichuan Basin. Results show that the surrogate model constructed from extensive simulation data significantly accelerates the inversion process. While maintaining accuracy, its computational efficiency is 14 times higher than the numerical simulation model. Compared to the single-step scheme, the multi-step inversion optimization scheme, which dynamically adjusts the inversion path through staged, iterative steps, provides more accurate results under the same parameters. The maximum error rates of normal stress and shear stress are reduced by 6.78% and 198.23%, respectively, while corresponding average error rates decrease by 1.72% and 23.53%. The proposed method provides an effective tool for mitigating geological hazards and optimizing exploitation, demonstrating strong potential for practical engineering applications.
Catastrophic failures in deep rock engineering are closely associated with the hydromechanical deterioration of rock masses, yet the lithology-dependent mechanisms of water-induced softening remain insufficiently understood. This study comparatively investigated the mechanical degradation, acoustic emission characteristics, and microstructural mechanisms of granite and slate under different water-content conditions. The results reveal a pronounced lithological contrast in water sensitivity. Under saturated conditions, slate underwent severe degradation, with marked reductions in tensile strength, compressive strength, and elastic modulus, whereas granite showed only moderate weakening. Acoustic emission analysis further indicated distinct damage evolution and failure processes between the two lithologies. Kernel smoothing analysis showed a more concentrated and stage-dependent damage evolution in granite, while slate exhibited a right-skewed frequency distribution, suggesting a more abrupt and unstable fracture process associated with high-frequency microcracking. Microstructural observations indicate that slate deterioration is mainly governed by clay-mineral hydration, swelling, and interlayer weakening, whereas granite weakening is primarily related to pore-water effects, stress corrosion, and microcrack propagation. These findings clarify lithology-dependent water-softening pathways and provide a mechanistic basis for understanding seepage-driven instability in deep rock engineering.
The application of flocculants is a critical step in mud-water separation for slurry shield tunneling, where the selection of dosing parameters depends on a scientifically rigorous flocculation evaluation system. However, current evaluation systems suffer from ambiguous frameworks, redundant indicators, and a lack of quantitative analysis. To address these bottlenecks limiting the efficiency of mud-water separation, this study proposes the Complete Destabilization of Shield Slurry (CDSS) theory based on typical organic flocculants. Multiscale mud-water separation experiments and multistage particle edge image recognition technology (EMW) were employed to establish macroscopic mud-water criteria and deduce the morphological critical thresholds of slurry particles. Key findings include: (1) CDSS reactions achieve rapid flocculation, producing sludge with a solid content exceeding 40%, wastewater turbidity below 150 NTU, and negligible post-separation volume changes. (2) CDSS reaction probability exceeds 95% when the average particle size ≥ 0.22 mm and density distribution ≥ 5550 counts/dm². (3) CPAM demonstrates superior triggering efficiency (92%) compared to NPAM (66%) and APAM (18%), attributed to synergistic effects between electric double layer interactions (energy barrier < 185×10⁻⁹ J·m⁻¹) and bridging mechanisms. This study's flocculation framework optimizes mud-water separation and sets a new standard for slurry efficiency in engineering and environmental applications.
Due to the sensitivity of compression waves to the presence of water in porous media, P-wave velocity has regularly been proposed as a method for estimating the water content of soils. However, various studies have shown that compression wave velocity depends not only on the degree of water saturation of the soil, but also on pore fluid distribution. Meanwhile, in the field of rock physics, several physical phenomena have been shown to induce dispersion of compression waves in partially saturated rocks, meaning that wave velocity is also a function of its frequency. Through an experimental study on Fontainebleau sand, we show that similar observations can be made in quasi-saturated sands: compression wave velocity is highly dependent upon probing frequency. These findings are then analysed through the lens of several existing compression wave dispersion mechanisms. Finally, in situ cross-hole tests recorded in quasi-saturated sands and previously published in the literature are reanalysed. The conclusion of the study is that reliable estimates of the degree of water saturation from compression wave velocity are difficult to make unless: (i) great care is paid to characterising the frequency of the emitted wave and (ii) potential wave dispersion mechanisms are investigated.
The clogging-column effect severely impedes the efficiency of vacuum preloading with prefabricated vertical drains (PVDs) in treating ultra-soft dredged slurries. Existing equivalent smear-based models are useful for overall prediction, but they generally do not explicity distinguish the different physical processes involved in clogging development. In this study, a large-strain radial vertical consolidation model is developed in which the clogging effect is represented by two components: (1) permeability deterioration associated with soil fabric reorganization, described by the parameter α, and (2) rapid local densification near the drain under high hydraulic gradients, described by the parameter η. The model also considers time-dependent decay of drain discharge capacity through β, together with spatially varying radial permeability. An implicit finite difference scheme is used to solve the governing equations. Parametric analyses indicate that η mainly affects the early stage development of clogging, α is more closely related to the long-term low-permeability state, and β mainly influences drainage efficiency during the middle and late stages of consolidation. Comparison with a field case in Wenzhou, China shows that the proposed model gives reasonable agreement with the measured settlement. For the case considered, the MAE and RMSE are reduced by 15.5% and 18.5%, respectively, relative to an existing large-strain solution with time-dependent well resistance, while the coefficient of determination R 2 increases from 0.881 to 0.921. The proposed formulation may provide a useful basis for interpreting clogging affected vacuum consolidation of dredged slurry.
Prolonged axial cyclic loading leads to degradation in both the strength and stiffness of the pile –soil interface. Therefore, developing computational models that can reasonably predict the axial cyclic behavior of pile foundations is essential. This paper presents a novel cyclic t–z model formulated within a single bounding surface plasticity framework. The model incorporates a newly developed elastoplastic interface stiffness interpolation function to capture the evolution of interface stiffness under cyclic loading. By integrating an interface strength degradation function correlated with accumulated vertical plastic displacement, the model effectively captures the progressive degradation of interfacial strength during cyclic loading. The proposed t–z model is numerically implemented in MATLAB and validated through simulations of published interface shear tests. It is subsequently embedded into ABAQUS via a user-defined element subroutine to create a nonlinear spring element. The model demonstrates strong predictive capability for both monotonic and cyclic axial responses of pile foundations. With a relatively small set of parameters, it effectively simulates the evolution of axial force along the pile shaft and captures key characteristics of the pile head response, including hysteresis, nonlinearity, and accumulated displacement.
Loess subgrades are highly susceptible to hydro-mechanical degradation under the combined effects of rainfall and temperature variations, leading to reduced structural stability and threatening highway operational safety. This paper investigates a cut-fill loess subgrade along the G85 highway in Guyuan City, monitoring temperature–moisture variations and settlement characteristics from the completion of construction through 2 years of operation. The results indicate that seasonal temperature fluctuations are more pronounced in fill sections than in cut sections. The left highway shoulder and slope toe shows the highest sensitivity to environmental changes. The volumetric moisture content in the fill section exhibits seasonal variations, with sensitive areas concentrated near the top layer of the subgrade and the slope toe. Influenced by rainfall, snowmelt, and freeze–thaw cycles in winter and evaporation in summer, the volumetric moisture content fluctuates significantly. Infiltration and temperature changes trigger localized cracking, resulting in differential settlement and compromising slope stability. Subgrade settlement displays layered and differential characteristics. During construction, the settlement rate reached 0.56 mm/day in fill sections and 0.13 mm/day in cut sections. After March 2024, settlement rates across the entire section stabilized below 0.1 mm/day. This research provides academic guidance for the construction and operation of subgrades in loess areas.
To promote the large-scale resource utilization of industrial solid wastes such as magnesium slag and realize sustainable development, this paper investigates the deformation, failure, and energy evolution laws of magnesium slag–based fully solid waste cemented tailings backfill (MFG-CTB) under different magnesium slag dosages (0%, 20%, 30%, and 40%) and confining pressures (0.5 and 2.0 MPa) through conventional triaxial compression tests. The internal mechanism of magnesium slag dosage affecting the mechanical properties of MFG-CTB is clarified using mercury intrusion porosimetry and scanning electron microscopy. On this basis, a damage constitutive model was established based on energy dissipation. The results show that the optimal dosage of magnesium slag as an alkali activator is 30%. At this dosage, the total porosity and macropore proportion of MFG-CTB are the lowest, and its triaxial compressive strength is increased by 112.7% and 130.5%, respectively, compared with the backfill without magnesium slag. However, the triaxial compressive strength of MFG-CTB first increases and then decreases with the increase of magnesium slag dosage. In addition, the addition of magnesium slag reduces the ductility, increases the stiffness, and enhances the energy storage capacity of MFG-CTB. With the increase of magnesium slag dosage, the failure mode of MFG-CTB under low confining pressure changes from tensile-shear composite failure to single shear failure, while that under high confining pressure transforms from local compression-shear failure to overall compression-shear failure, with the main crack gradually approaching the middle of the specimen. Finally, the MFG-CTB damage constitutive model established based on dissipated strain energy and considering residual strength is in high agreement with the experimental curves, which can reasonably and authentically describe the deformation, instability, and failure process of the backfill.
Offshore wind is a rapidly expanding source of renewable energy, with new developments planned in seismically active regions such as the West coast of North America and South East Asia. These areas often feature loose, liquefiable sands, which pose challenges to the stability of both monopile foundations and their scour protection during an earthquake. Scour protection, typically consisting of graded rock berms placed around the foundation, is essential for long-term turbine performance, preventing localized soil erosion and minimizing changes to foundation behaviour. However, the behaviour of rock-scour protection under seismic liquefaction remains poorly understood. Large settlements of the rock berm have been observed in previous studies, yet best practice for designing scour protection in liquefiable soils is still unclear. A deeper understanding of this phenomenon will be key to ensuring the safe and efficient construction and operation of future offshore wind farms in earthquake-prone regions. This paper presents the results of several saturated, dynamic centrifuge tests, comparing the behaviour of scour protection rock berms. The effect of rock size, rock density, foundation soil density, and earthquake intensity is explored. Finally, nondimensional plots are provided to aid both the design of future rock berms and further research in this area.
With tunnel boring machines (TBMs) advancing into deeper and more complex ground conditions, hyperbaric intervention has become an indispensable but high-risk operation. The success of such interventions relies heavily on the formation and durability of an airtight filter cake at the tunnel face. However, under sustained air pressure, the filter cake undergoes compression and consolidation, which may eventually lead to perforation failure. The time-dependent nature of this process introduces significant uncertainty in predicting the duration of airtightness, posing operational and safety risks. To address this issue, this paper develops a consolidation–perforation model that captures both the increase in effective stress from consolidation and the failure of the filter cake. An airtightness test was conducted to validate the proposed model, and the results demonstrated good agreement with theoretical predictions. The findings provide quantitative insight into the failure evolution of filter cakes under air pressure and enable prediction of airtightness duration. This study offers guidance for safe hyperbaric interventions in TBM tunneling.
The simplified physically-based dam breach models are effective tools for predicting outburst flood hydrographs of landslide dams. However, their predictive reliability is severely constrained by parameter uncertainties, particularly regarding in soil erosion. To address this, this study applies an established Bayesian multilevel framework to develop a probabilistic modeling approach for landslide dam breaches. A highly computationally efficient simplified model is developed and subsequently embedded into a Bayesian multilevel framework to systematically quantify the uncertainties in the erosion parameters. Using observational data from ten documented landslide dam failure cases, model inversion is executed via a Markov chain Monte Carlo simulation combining Gibbs and Metropolis-Hastings sampling. As a primary contribution, this study quantifies the uncertainty of the erosion parameter specifically for landslide dams for the first time. Following inversion, parameters with non-informative priors are updated to well-defined posterior distributions with distinct peaks. Furthermore, the results reveal that approximately two-thirds of the uncertainty in the predicted peak discharge stems from the epistemic uncertainty of key parameters, with the remainder attributed to residual error. This framework significantly improves the reliability of outburst flood predictions and substitutes subjective empirical assumptions with data-driven probabilistic inference, providing highly valuable insights for downstream hazard mitigation.
Field tests were conducted to evaluate the penetration resistance of steel plates in scour protection systems, usually employed on bottom-fixed offshore foundations. Scour protection typically consists of loose rock, whose particle size is comparable to the thickness of the skirts or piles, introducing a grain-size effect. To investigate this effect, plates under plane–strain conditions with four different thicknesses were tested in limestone gravel. In addition, interaction between closely spaced skirts was examined through double-plate configurations. The results indicate that penetration resistance increases with both penetration depth and plate thickness, and the response followed four phases: (i) an initial sharp increase in resistance, (ii) a transient phase while the failure mechanism develops around the tip, (iii) a linear increase once the mechanism is fully mobilized, and (iv) a plateau at a critical depth where resistance becomes nearly constant. Interaction effects between double-plate skirts were negligible at shallow penetration depths. A comparison with previous tests in high-density gravel of similar mean particle size showed lower penetration resistance for the limestone material. Analytical predictions based on bearing capacity theory agreed with the data when grain-size effects were included using an equivalent plate thickness equal to the actual thickness plus the mean particle diameter.
Slope deformation is a critical indicator of pit wall stability in open-pit mining, yet its prediction remains challenging due to the combined effects of mining operations and environmental conditions. This study investigates the use of machine learning to predict deformation trends in a large open-pit mine in British Columbia, using multi-source field-monitoring data. Slope displacement was continuously measured using ground-based radar, and associated datasets, including bench excavation records, blasting vibration, precipitation, pore water pressure, and temperature, were integrated into the analysis. Deformation analysis indicates that bench excavations near fault-controlled unstable zones significantly accelerate slope movement, whereas blasting-induced vibrations have a limited direct influence. Three machine learning algorithms, linear regression, regression trees, and support vector machines, were evaluated under both random and chronological data-splitting strategies. While all models reproduced overall deformation trends under random data splits, regression tree and support vector machine models demonstrated superior performance in forward prediction scenarios. Short-term forward predictions accurately captured deformation velocities and peak movement periods, with regression trees providing the best predictive accuracy. The results demonstrate that machine learning, when trained on comprehensive field datasets, can provide reliable short-term forecasts of slope deformation, thereby supporting operational decision-making and risk mitigation in open-pit mining.
