The strength of the sliding zone soil determines the stability of reservoir landslides.Fluctuations in water levels cause a change in the seepage field,which serves as both the external hydrogeological environment and...The strength of the sliding zone soil determines the stability of reservoir landslides.Fluctuations in water levels cause a change in the seepage field,which serves as both the external hydrogeological environment and the internal component of a landslide.Therefore,considering the strength changes of the sliding zone with seepage effects,they correspond with the actual hydrogeological circumstances.To investigate the shear behavior of sliding zone soil under various seepage pressures,24 samples were conducted by a self-developed apparatus to observe the shear strength and measure the permeability coefficients at different deformation stages.After seepage-shear tests,the composition of clay minerals and microscopic structure on the shear surface were analyzed through X-ray and scanning electron microscope(SEM)to understand the coupling effects of seepage on strength.The results revealed that the sliding zone soil exhibited strain-hardening without seepage pressure.However,the introduction of seepage caused a significant reduction in shear strength,resulting in strain-softening characterized by a three-stage process.Long-term seepage action softened clay particles and transported broken particles into effective seepage channels,causing continuous damage to the interior structure and reducing the permeability coefficient.Increased seepage pressure decreased the peak strength by disrupting occlusal and frictional forces between sliding zone soil particles,which carried away more clay particles,contributing to an overhead structure in the soil that raised the permeability coefficient and decreased residual strength.The internal friction angle was less sensitive to variations in seepage pressure than cohesion.展开更多
The Qinghai-Xizang Plateau of China faces challenges like thaw slumping,threatening slope stability and infrastructure.Understanding the mechanical properties of the roots of the dominant herbaceous plant species in t...The Qinghai-Xizang Plateau of China faces challenges like thaw slumping,threatening slope stability and infrastructure.Understanding the mechanical properties of the roots of the dominant herbaceous plant species in the alpine meadow layer of the permafrost regions on the Qinghai-Xizang Plateau is essential for evaluating their role in enhancing soil shear strength and mitigating slope deformation in these fragile environments.In this study,the roots of four dominant herbaceous plant species—Kobresia pygmaea,Kobresia humilis,Carex moorcroftii,and Leontopodium pusillum—that are widely distributed in the permafrost regions of the Qinghai-Xizang Plateau were explored to determine their mechanical properties and effects in enhancing soil shear strength.Through indoor single root tensile and root group tensile tests,we determined the root diameter,tensile force,tensile strength,tensile ratio,and strength frequency distributions.We also evaluated their contributions to inhibiting slope deformation and failure during the formation and development of thermal thaw slumps in the alpine meadow.The results showed that the distribution of the root diameter of the dominant plant species is mostly normal,while the tensile strength tends to be logarithmically normally distributed.The relationship between the root diameter and root tensile strength conforms to a power function.The theoretical tensile strength of the root group was calculated using the Wu-Waldron Model(WWM)and the Fiber Bundle Model(FBM)under the assumption that the cumulative single tensile strength of the root bundle is identical to the tensile strength of the root group in the WWM.The FBM considers three fracture modes:FBM-D(the tensile force on each single root is proportional to its diameter relative to the total sum of all the root diameters),FBM-S(the cross-sectional stress in the root bundle is uniform),and FBM-N(each tensile strength test of individual roots experiences an equal load).It was found that the model-calculated tensile strength of the root group was 162.60%higher than the test value.The model-derived tensile force of the root group from the FBM-D,FBM-S,and FBM-N was 73.10%,28.91%,and 13.47%higher than the test values,respectively.The additional cohesion of the soil provided by the roots was calculated to be 25.90-45.06 kPa using the modified WWM,67.05-38.15 kPa using the FBM-S,and 57.24-32.74 kPa using the FBM-N.These results not only provide a theoretical basis for further quantitative evaluation of the mechanical effects of the root systems of herbaceous plant species in reinforcing the surface soil but also have practical significance for the effective prevention and control of thermal thaw slumping disasters in the permafrost regions containing native alpine meadows on the Qinghai-Xizang Plateau using flexible plant protection measures.展开更多
The primary objective of this work is to improve our understanding of the mechanical involvements of two-order roughness in shear.First,wavelet analysis is used to separate the waviness(first-order)and unevenness(seco...The primary objective of this work is to improve our understanding of the mechanical involvements of two-order roughness in shear.First,wavelet analysis is used to separate the waviness(first-order)and unevenness(second-order)from four granite joint surfaces,with roughness characterized using Grasselli’s 3D morphology parameters.The results reveal that first-order roughness is more pronounced than second-order roughness,highlighting the dominant role of waviness in joint surface roughness.Additionally,the variation in first-order roughness with strike direction corresponds to the total roughness,while second-order roughness remains largely constant,indicating that roughness anisotropy is primarily driven by waviness.Then,direct shear tests on joint replicas are performed to investigate the contributions of both roughness orders to peak shear strength.The results show that the peak dilation angle is closely related to first-order roughness,while the shear component angle is closely associated with second-order roughness,both exhibiting a linear correlation.Based on these findings,relationships are established between the angles and their respective roughness orders.Finally,a joint shear strength criterion based on two-order roughness is proposed.A comparative analysis of prediction accuracy reveals that the average relative error for the proposed criterion is 13.79%,while the errors for Xia's,Yang's,and Ban's criteria are 15.19%,16.29%,and 13.87%,respectively.It demonstrates the proposed criterion can predict the peak shear strength of rock joints.展开更多
Shear strength of hydrate-bearing sediment is an essential parameter for assessing landslide potential ofhydrate reservoirs under exploration conditions. However, the characteristics and simulation of thisshear streng...Shear strength of hydrate-bearing sediment is an essential parameter for assessing landslide potential ofhydrate reservoirs under exploration conditions. However, the characteristics and simulation of thisshear strength under varying dissociation conditions have not been thoroughly investigated. To this end,a series of triaxial compression tests were first carried out on sediments with varying initial hydratesaturations along dissociation pathways. Combining measured data with microscale analysis, the underlyingmechanism for the evolution of shear strength in hydrate-bearing sediment was studied undervarying partial dissociation pathways. Moreover, a shear strength model for hydrate-bearing sedimentwas proposed, taking into account the hydrate saturation and the unhydrated water content. Apart fromthe parameters derived from the hydrate characteristic curve, only one additional model parameter isrequired. The proposed model was validated using measured data on hydrate sediments. The resultsindicate that the proposed model can effectively capture the shear strength behavior of hydrate-bearingsediment under varying dissociation paths. Finally, a sensitivity analysis of the model parameters wasconducted to characterize the proposed model.展开更多
This study presents an experimental investigation into the effects of microbial induced carbonate precipitation(MICP)treatment factors on the shear behavior of MICP-treated loess soil.Several groups of loess samples w...This study presents an experimental investigation into the effects of microbial induced carbonate precipitation(MICP)treatment factors on the shear behavior of MICP-treated loess soil.Several groups of loess samples were prepared and subjected to MICP treatment with varying cementation reagent concentration,calcium source,and curing duration across three levels.The results indicate that the shear strength of MICP-treated loess achieves optimal performance when treated with the cementation reagent concentration of 1.0 M,cured for 14 days,and using calcium chloride as the calcium source.Compared to untreated loess,the cohesion and internal friction angle increased by approximately 77%and 26%,respectively.To evaluate the influence of these treatment variables,orthogonal analysis was performed on the obtained shear strength parameters.The analysis indicates that the cementation reagent concentration is the primary factor influencing shear strength,followed by the calcium source and curing duration.Additionally,scanning electron microscopy(SEM)tests were performed to investigate the microstructure of the MICP-treated samples.The results reveal that calcium carbonate significantly enhances the loess structure by creating large effective bonding areas,which in turn increases the bridging force.As a result,the overall shear strength of the treated loess shows a marked improvement compared to the untreated samples.展开更多
Comprehensive investigations have been conducted to study the structure and overconsolidation of upper Shanghai clays, i.e. Layers 2–6 clays, typically located at depths of 30–40 m. However, limited information is a...Comprehensive investigations have been conducted to study the structure and overconsolidation of upper Shanghai clays, i.e. Layers 2–6 clays, typically located at depths of 30–40 m. However, limited information is available on their anisotropy, and even less is known about the correlation between structure, overconsolidation, and anisotropy. In this study, the undrained anisotropy characteristics of shear strength and small-strain shear stiffness in upper Shanghai Layers 2–6 clays were thoroughly assessed using a series of K0-consolidated undrained triaxial compression (TC) and triaxial extension (TE) tests (K0 is the coefficient of lateral earth pressure at rest). The effective stress paths, shear strength, and small-strain shear stiffness from the undrained TC and TE tests demonstrate the anisotropic behaviors in upper Shanghai clays. Analyses of data from upper Shanghai clays and other clays worldwide indicate that the shear strength anisotropy ratio (Ks) converges at 0.8 as the overconsolidation ratio (OCR) and plasticity index (Ip) increase, while the small-strain shear stiffness anisotropy ratio (Re) converges at 1.0. The influence of OCR on Ks and Re is more pronounced than that of Ip and sensitivity (St). Nevertheless, no clear correlation between Ks and Re is observed in upper Shanghai clays.展开更多
Joint surface roughness is comprised of two components:large-scale(waviness,first-order)and smallscale(unevenness,second-order).To investigate the scale effect of two-order roughness,a surface area method is used to s...Joint surface roughness is comprised of two components:large-scale(waviness,first-order)and smallscale(unevenness,second-order).To investigate the scale effect of two-order roughness,a surface area method is used to separate the waviness and unevenness from the whole joint surface of each size.The results show that as the joint size increases,the first-order roughness exhibits a decrease,whereas the second-order roughness remains constant.This indicates that the roughness scale effect is primarily attributed to the waviness.Then,the influence of two-order roughness on the scale effect of joint shear strength is examined through a series of laboratory tests.The test results reveal a negative scale effect on the peak friction angle,which exhibits a strong correlation with the first-order roughness.The decrease in the peak friction angle with normal stress can be attributed primarily to the degradation of second-order roughness,which shows a negative exponential trend.Finally,an empirical model for estimating the peak friction angle of rock joints at different scales is proposed based on two-order roughness.The prediction accuracy of this model is validated by comparing the existing criteria.展开更多
The interpretation and application of CPT(cone penetration test)results is characterized by considerable variability of data,either in measured or correlated parameters.According to the requirements of Eurocode 7 the ...The interpretation and application of CPT(cone penetration test)results is characterized by considerable variability of data,either in measured or correlated parameters.According to the requirements of Eurocode 7 the existing variability in soil properties has to be taken into account statistically during the determination of the characteristic values of each parameter.This should be done by selecting a cautious estimate of the value affecting the limit state.Obtaining the characteristic values of CPT measurements is not an easy task and on this aspect nor clear neither unified guidelines exist.This paper focuses in several approaches to characterize the cone resistance and the sleeve friction using simple statistical analysis,in order for these parameters to be applicable in design.Similar procedures are then applied to determine the characteristic values of correlated parameters from CPT such as the effective friction angle for sands and the undrained shear strength for clays.The resulting characteristic values of the considered parameters emphasize the fact that the prediction and the interpretation of characteristic values of soil properties is a complicated and biased procedure.展开更多
This opinion article discusses the original research work of Yünkül et al.(the Authors)published in the Journal of Mountain Science 21(9):3108–3122.Employing non-linear regression,fuzzy logic and artificial...This opinion article discusses the original research work of Yünkül et al.(the Authors)published in the Journal of Mountain Science 21(9):3108–3122.Employing non-linear regression,fuzzy logic and artificial neural network modeling techniques,the Authors interrogated a large database assembled from the existing research literature to assess the performance of twelve equation rules in predicting the undrained shear strength(s_(u))mobilized for remolded fine-grained soils at different values of liquidity index(I_(L))and water content ratio.Based on their analyses,the Authors proposed a simple and reportedly reliable correlation(i.e.,Eq.9 in their paper)for predicting s_(u) over the I_(L) range of 0.15 to 3.00.This article describes various shortcomings in the Authors’assembled database(including potentially anomalous data and covering an excessively wide I_(L) range in relation to routine geotechnical and transportation engineering applications)and their proposed s_(u)=f(I_(L))correlation.Contrary to the Authors’assertions,their proposed correlation is not reliable for fine-grained soils with consistencies in the general firm to stiff range(i.e.,for 0.15<I_(L)<0.40),increasingly overestimating s_(u) for reducing I_(L),and eventually predicting s_(u)→+∞for I_(L)→0.15+(while producing mathematically undefined s_(u) for I_(L)<0.15),thus rendering their correlation unconservative and potentially leading to unsafe geotechnical designs.Exponential or regular-power type s_(u)=f(I_(L))models are more s_(u)itable when developing correlations that are applicable over the full plastic range(of 0<I_(L)<1),thereby providing reasonably conservative s_(u) predictions for use in the preliminary design for routine geotechnical engineering applications.展开更多
This study presents an in-depth investigation into the shear strength characteristics of unsaturated soils,focusing on the influenceof shear rate and initial water saturation(S_(r0)).Utilizing the drained-vented(DV)tr...This study presents an in-depth investigation into the shear strength characteristics of unsaturated soils,focusing on the influenceof shear rate and initial water saturation(S_(r0)).Utilizing the drained-vented(DV)triaxial test method,the present study investigated the shear strength behavior of silty clay under various shear rates and water saturation levels,and compared the outcomes with traditional suction-controlled(SC)and constant water content(CW)tests.The findingshighlight the pivotal role of excess pore water pressure dissipation during shearing,which significantlyaffects the shear strength of both saturated and unsaturated soils.Notably,for soils with high initial water saturation,a decrease in shear strength is observed with an increase in shear rate,which is attributed to the rise in pore water pressure.Conversely,for soils with low initial water saturation,the shear rate exhibits minimal impact on shear strength due to negligible water drainage.The research delineates the optimal shear rates for DV tests based on the initial water saturation:a slower rate of 0.0028 mm/min for samples with high water saturation(S_(r0)>66%)and a faster rate of 0.8 mm/min for samples with low water saturation(S_(r0)≤66%).A novel testing methodology for determining unsaturated soil shear strength under DV conditions is introduced,streamlining the measurement process and significantly reducing testing time.This method not only promises substantial cost savings but also aligns closely with natural engineering conditions,offering valuable guidance for geotechnical applications.展开更多
The scale effect on shear strength of rock joints is well-documented.However,whether scale effects are negative,positive,or even exist or not is still controversial.Joint roughness significantly influences the shear s...The scale effect on shear strength of rock joints is well-documented.However,whether scale effects are negative,positive,or even exist or not is still controversial.Joint roughness significantly influences the shear strength of rock joints.Compared to the shear tests,using the joint roughness coefficient(JRC)and its roughness parameters offers a more convenient method for describing the scale effect on shear strength.However,it is crucial to understand that the scale effect mechanisms of JRC are distinct from those of shear strength.Therefore,this paper aims to clarify these distinct mechanisms.By digitally extracting roughness parameters from granite samples,it is found that the scale effect of roughness parameters mainly comes from the sampling methods and the geometric characteristics of parameters.Furthermore,a full data sampling method considering heterogeneity is proposed to obtain more representative roughness parameters.To reveal the scale effect mechanisms of shear strength,Gaussian filtering is firstly used to separate the waviness and unevenness components of roughness,facilitating a deeper understanding of the geometric characteristics of roughness.It is suggested that the wavelength of the waviness component can reflect the scale effect on shear strength.Secondly,numerical simulations of ideal artificial joint models are conducted to validate that the wavelength of the waviness component serves as the dividing point between positive and negative scale effects.The mechanical mechanisms of positive and negative scale effects are also interpreted.Finally,these mechanisms successfully elucidate the occurrence patterns of the scale effect on natural joint profiles.展开更多
In response to the effectiveness of reforestation in controlling soil erosion,there has been a dramatic increase in forest coverage in the hilly red soil region of southern China.Aggregate stability and soil shear str...In response to the effectiveness of reforestation in controlling soil erosion,there has been a dramatic increase in forest coverage in the hilly red soil region of southern China.Aggregate stability and soil shear strength are indicators that reflect soil resistance to erosion and its ability to prevent shallow landslides,respectively.However,limited research has focused on the response of soil aggregate stability and shear strength to reforestation.We selected three types of reforestations(Phyllostachys edulis forest,Cunninghamia lanceolata(Lamb.)Hook.forest,Citrus sinensis(L.)Osbeck.orchard),a natural forest(mixed coniferous and broadleaf forests),and a fallow land as study plots,and measured root traits,and soil physicochemical traits,i.e.,pH,soil organic matter(SOC),Soil water content(SWC),soil bulk density(BD),soil cohesion(c),soil internal friction angle(φ)and analyzed their multiple interactions.The soil aggregate stability traits,refer to the mean weight diameter(MWD)and geometric mean diameter(GMD),exhibited a significant increase in reforested plots,approximately 200%compared to fallow land and 50%compared to natural forests.For soil shear strength the values were approximately 20%higher than in fallow land and approximately 10%lower than in natural forests.Soil aggregate stability and soil shear strength did not exhibit a significant positive correlation across all plots,and the underlying drivers of these traits were variable.For instance,in natural forest and timber stands,soil aggregate stability was mainly influenced by soil organic carbon,while soil shear strength was primarily affected by root length density.In economic forest,aggregate stability and shear strength are mainly affected by organic carbon.Overall,we found that vegetation restoration enhances soil erosion resistance,however,the primary drivers for the improvement of aggregate stability(soil organic carbon)and shear strength(root length density)are different.Therefore,in future benefit assessments of vegetation restoration projects aimed at soil erosion control,different indicators should be considered based on specific conditions.展开更多
Conventional empirical equations for estimating undrained shear strength(s_(u))from piezocone penetration test(CPTu)data,without incorporating soil physical properties,often lack the accuracy and robustness required f...Conventional empirical equations for estimating undrained shear strength(s_(u))from piezocone penetration test(CPTu)data,without incorporating soil physical properties,often lack the accuracy and robustness required for geotechnical site investigations.This study introduces a hybrid virus colony search(VCS)algorithm that integrates the standard VCS algorithm with a mutation-based search mechanism to develop high-performance XGBoost learning models to address this limitation.A dataset of 372 seismic CPTu and corresponding soil physical properties data from 26 geotechnical projects in Jiangs_(u)Province,China,was collected for model development.Comparative evaluations demonstrate that the proposed hybrid VCS-XGBoost model exhibits s_(u)perior performance compared to standard meta-heuristic algorithm-based XGBoost models.The res_(u)lts highlight that the consideration of soil physical properties significantly improves the predictive accuracy of s_(u),emphasizing the importance of considering additional soil information beyond CPTu data for accurate s_(u)estimation.展开更多
Machine learning(ML)models are widely used for predicting undrained shear strength(USS),but interpretability has been a limitation in various studies.Therefore,this study introduced shapley additive explanations(SHAP)...Machine learning(ML)models are widely used for predicting undrained shear strength(USS),but interpretability has been a limitation in various studies.Therefore,this study introduced shapley additive explanations(SHAP)to clarify the contribution of each input feature in USS prediction.Three ML models,artificial neural network(ANN),extreme gradient boosting(XGBoost),and random forest(RF),were employed,with accuracy evaluated using mean squared error,mean absolute error,and coefficient of determination(R^(2)).The RF achieved the highest performance with an R^(2) of 0.82.SHAP analysis identified pre-consolidation stress as a key contributor to USS prediction.SHAP dependence plots reveal that the ANN captures smoother,linear feature-output relationships,while the RF handles complex,non-linear interactions more effectively.This suggests a non-linear relationship between USS and input features,with RF outperforming ANN.These findings highlight SHAP’s role in enhancing interpretability and promoting transparency and reliability in ML predictions for geotechnical applications.展开更多
Reactive brazing of TiAl-based intermetallics and Ni-based alloy with Ti foil as interlayer was investigated. The interfacial microstructure and shear strength of the joints were studied. According to the experimental...Reactive brazing of TiAl-based intermetallics and Ni-based alloy with Ti foil as interlayer was investigated. The interfacial microstructure and shear strength of the joints were studied. According to the experimental observations, the molten interlayer reacts vigorously with base metals, forming several continuous reaction layers. The typical interfacial microstructure of the joint can be expressed as GH99/（Ni,Cr）ss（γ）/TiNi（β2）＋TiNi2Al（τ4）＋Ti2Ni（δ）/δ＋Ti3Al（α2）＋Al3NiTi2（τ3）/α2＋τ3/TiAl. The maximum shear strength is 258 MPa for the specimen brazed at 1000°C for 10 min. Higher brazing temperature or longer brazing time causes coarsening of the phases in the brazing seam and formation of brittle intermetallic layer, which greatly depresses the shear strength of the joints.展开更多
The bonding interface characteristic and shear strength of diffusion bonded Ti-17 titanium alloy at different bonding time were investigated. The results show that the average size of voids decreases while the amount ...The bonding interface characteristic and shear strength of diffusion bonded Ti-17 titanium alloy at different bonding time were investigated. The results show that the average size of voids decreases while the amount of voids decreases after increasing to the maximum value with the increasing bonding time. The irregular void with a scraggly edge tends to an ellipse void with smooth surface and then changes to a tiny void with round shape. The grains across bonding interface occur at bonding time of 60 min. The shear strength of bond increases with increasing bonding time, and the highest shear strength of bond is 887.4 MPa at 60 min. The contribution of plastic deformation on the void closure and the increase of shear strength is significant even though the action time of plastic deformation is short.展开更多
In order to obtain the reasonable undrained shear strength Su for geotechnical analyses of bridge foundations in Yangtze River floodplain clayey soils, a site-specific study is conducted using the imported piezocone p...In order to obtain the reasonable undrained shear strength Su for geotechnical analyses of bridge foundations in Yangtze River floodplain clayey soils, a site-specific study is conducted using the imported piezocone penetration test （CPTu） with dissipation phases at the Fourth Nanjing Yangtze River Bridge construction sites. Taking the values of Su from laboratory tests as references, several existing Su-predicted methods based on CPTu are compared and evaluated. To verify the presented cone factor Nk, additional test sites are selected and examined. The results show that the values of cone factors such as Nkt, Nke, and Nau, depend on the shear test mode and disturbance. Generally, the values of Nke show more scattering than those of Nkt and N△u. For the stratified and layered sediments of the Yangtze River floodplain, it is recommended using the net cone resistance qT to estimate Su and the preliminary cone factor values Nkt are from 7 to 16, with an average of 11. It is also confirmed that the CPTu test, as a new technique in site characterization, can present reasonable parameters for bridge foundations.展开更多
A rigorous back analysis of shear strength parameters of landslide slip was presented. Kinematical element method was adopted to determine factor of safety and critical failure surface, which overcomes the disadvantag...A rigorous back analysis of shear strength parameters of landslide slip was presented. Kinematical element method was adopted to determine factor of safety and critical failure surface, which overcomes the disadvantage of limit equilibrium method. The theoretical relationship between the combination of shear strength parameters and stability state was studied. The results show that the location of critical slip surface, F/tan f and F/c depend only on the value of c/tan f. The failure surface moves towards the inside of slope as c/tan f increases. According to the information involving factor of safety and critical failure surface in a specific cross-section, strength parameters can be back calculated based on the above findings. Three examples were given for demonstrating the validity of the present method. The shear strength parameters obtained by back analysis are almost consistent with their correct solutions or test results.展开更多
For sites susceptible to liquefaction induced lateral spreading during a probable earthquake, geotechnical engineers often need to know the undrained residual shear strength of the liquefied soil deposit to estimate l...For sites susceptible to liquefaction induced lateral spreading during a probable earthquake, geotechnical engineers often need to know the undrained residual shear strength of the liquefied soil deposit to estimate lateral spreading displacements, and the forces acting on the piles from the liquefied soils in order to perform post liquefaction stability analyses. The most commonly used methods to estimate the undrained residual shear strength （Su~） of liquefied sand deposits are based on the correlations determined from liquefaction induced flow failures with SPT and CPT data. In this study, 44 lateral spread case histories are analyzed and a new relationship based on only lateral spread case histories is recommended, which estimates the residual shear strength ratio of the liquefiable soil layer from normalized shear wave velocity. The new proposed method is also utilized to estimate the residual lateral displacement of an example bridge problem in an area susceptible to lateral spreading in order to provide insight into how the proposed relationship can be used in geotechnical engineering practice.展开更多
Accurate assessment of undrained shear strength(USS)for soft sensitive clays is a great concern in geotechnical engineering practice.This study applies novel data-driven extreme gradient boosting(XGBoost)and random fo...Accurate assessment of undrained shear strength(USS)for soft sensitive clays is a great concern in geotechnical engineering practice.This study applies novel data-driven extreme gradient boosting(XGBoost)and random forest(RF)ensemble learning methods for capturing the relationships between the USS and various basic soil parameters.Based on the soil data sets from TC304 database,a general approach is developed to predict the USS of soft clays using the two machine learning methods above,where five feature variables including the preconsolidation stress(PS),vertical effective stress(VES),liquid limit(LL),plastic limit(PL)and natural water content(W)are adopted.To reduce the dependence on the rule of thumb and inefficient brute-force search,the Bayesian optimization method is applied to determine the appropriate model hyper-parameters of both XGBoost and RF.The developed models are comprehensively compared with three comparison machine learning methods and two transformation models with respect to predictive accuracy and robustness under 5-fold cross-validation(CV).It is shown that XGBoost-based and RF-based methods outperform these approaches.Besides,the XGBoostbased model provides feature importance ranks,which makes it a promising tool in the prediction of geotechnical parameters and enhances the interpretability of model.展开更多
基金supported by the Major Program of the National Natural Science Foundation of China (Grant No.42090055)the National Major Scientific Instruments and Equipment Development Projects of China (Grant No.41827808)the National Nature Science Foundation of China (Grant No.42207216).
文摘The strength of the sliding zone soil determines the stability of reservoir landslides.Fluctuations in water levels cause a change in the seepage field,which serves as both the external hydrogeological environment and the internal component of a landslide.Therefore,considering the strength changes of the sliding zone with seepage effects,they correspond with the actual hydrogeological circumstances.To investigate the shear behavior of sliding zone soil under various seepage pressures,24 samples were conducted by a self-developed apparatus to observe the shear strength and measure the permeability coefficients at different deformation stages.After seepage-shear tests,the composition of clay minerals and microscopic structure on the shear surface were analyzed through X-ray and scanning electron microscope(SEM)to understand the coupling effects of seepage on strength.The results revealed that the sliding zone soil exhibited strain-hardening without seepage pressure.However,the introduction of seepage caused a significant reduction in shear strength,resulting in strain-softening characterized by a three-stage process.Long-term seepage action softened clay particles and transported broken particles into effective seepage channels,causing continuous damage to the interior structure and reducing the permeability coefficient.Increased seepage pressure decreased the peak strength by disrupting occlusal and frictional forces between sliding zone soil particles,which carried away more clay particles,contributing to an overhead structure in the soil that raised the permeability coefficient and decreased residual strength.The internal friction angle was less sensitive to variations in seepage pressure than cohesion.
基金supported by the Qinghai Science and Technology Department Project(2025-QY-225)the National Natural Science Foundation of China(42267024)the Second Comprehensive Scientific Investigation and Research Project of the Qinghai-Xizang Plateau(2019QZKK0905).
文摘The Qinghai-Xizang Plateau of China faces challenges like thaw slumping,threatening slope stability and infrastructure.Understanding the mechanical properties of the roots of the dominant herbaceous plant species in the alpine meadow layer of the permafrost regions on the Qinghai-Xizang Plateau is essential for evaluating their role in enhancing soil shear strength and mitigating slope deformation in these fragile environments.In this study,the roots of four dominant herbaceous plant species—Kobresia pygmaea,Kobresia humilis,Carex moorcroftii,and Leontopodium pusillum—that are widely distributed in the permafrost regions of the Qinghai-Xizang Plateau were explored to determine their mechanical properties and effects in enhancing soil shear strength.Through indoor single root tensile and root group tensile tests,we determined the root diameter,tensile force,tensile strength,tensile ratio,and strength frequency distributions.We also evaluated their contributions to inhibiting slope deformation and failure during the formation and development of thermal thaw slumps in the alpine meadow.The results showed that the distribution of the root diameter of the dominant plant species is mostly normal,while the tensile strength tends to be logarithmically normally distributed.The relationship between the root diameter and root tensile strength conforms to a power function.The theoretical tensile strength of the root group was calculated using the Wu-Waldron Model(WWM)and the Fiber Bundle Model(FBM)under the assumption that the cumulative single tensile strength of the root bundle is identical to the tensile strength of the root group in the WWM.The FBM considers three fracture modes:FBM-D(the tensile force on each single root is proportional to its diameter relative to the total sum of all the root diameters),FBM-S(the cross-sectional stress in the root bundle is uniform),and FBM-N(each tensile strength test of individual roots experiences an equal load).It was found that the model-calculated tensile strength of the root group was 162.60%higher than the test value.The model-derived tensile force of the root group from the FBM-D,FBM-S,and FBM-N was 73.10%,28.91%,and 13.47%higher than the test values,respectively.The additional cohesion of the soil provided by the roots was calculated to be 25.90-45.06 kPa using the modified WWM,67.05-38.15 kPa using the FBM-S,and 57.24-32.74 kPa using the FBM-N.These results not only provide a theoretical basis for further quantitative evaluation of the mechanical effects of the root systems of herbaceous plant species in reinforcing the surface soil but also have practical significance for the effective prevention and control of thermal thaw slumping disasters in the permafrost regions containing native alpine meadows on the Qinghai-Xizang Plateau using flexible plant protection measures.
基金funded by the National Natural Science Foundation of China (Grant nos. 42272333 and 42377154)the China Association for Science and Technology Youth Talent Support Program for PhD Students.
文摘The primary objective of this work is to improve our understanding of the mechanical involvements of two-order roughness in shear.First,wavelet analysis is used to separate the waviness(first-order)and unevenness(second-order)from four granite joint surfaces,with roughness characterized using Grasselli’s 3D morphology parameters.The results reveal that first-order roughness is more pronounced than second-order roughness,highlighting the dominant role of waviness in joint surface roughness.Additionally,the variation in first-order roughness with strike direction corresponds to the total roughness,while second-order roughness remains largely constant,indicating that roughness anisotropy is primarily driven by waviness.Then,direct shear tests on joint replicas are performed to investigate the contributions of both roughness orders to peak shear strength.The results show that the peak dilation angle is closely related to first-order roughness,while the shear component angle is closely associated with second-order roughness,both exhibiting a linear correlation.Based on these findings,relationships are established between the angles and their respective roughness orders.Finally,a joint shear strength criterion based on two-order roughness is proposed.A comparative analysis of prediction accuracy reveals that the average relative error for the proposed criterion is 13.79%,while the errors for Xia's,Yang's,and Ban's criteria are 15.19%,16.29%,and 13.87%,respectively.It demonstrates the proposed criterion can predict the peak shear strength of rock joints.
基金supported by the National Natural Science Foundation of China(Grant No.51939011)the Science and Technology Program of CNOOC Research Institute(Grant No.2023OTKK03)supported by the program of the Youth Innovation Promotion Association,Chinese Academy of Sciences(Grant No.2020326).
文摘Shear strength of hydrate-bearing sediment is an essential parameter for assessing landslide potential ofhydrate reservoirs under exploration conditions. However, the characteristics and simulation of thisshear strength under varying dissociation conditions have not been thoroughly investigated. To this end,a series of triaxial compression tests were first carried out on sediments with varying initial hydratesaturations along dissociation pathways. Combining measured data with microscale analysis, the underlyingmechanism for the evolution of shear strength in hydrate-bearing sediment was studied undervarying partial dissociation pathways. Moreover, a shear strength model for hydrate-bearing sedimentwas proposed, taking into account the hydrate saturation and the unhydrated water content. Apart fromthe parameters derived from the hydrate characteristic curve, only one additional model parameter isrequired. The proposed model was validated using measured data on hydrate sediments. The resultsindicate that the proposed model can effectively capture the shear strength behavior of hydrate-bearingsediment under varying dissociation paths. Finally, a sensitivity analysis of the model parameters wasconducted to characterize the proposed model.
基金This work is supported by the Young Talent Fund of Association for Science and Technology in Shaanxi,China(20240722)the Shaanxi Province Postdoctoral Research Project(2023BSHYDZZ138)+2 种基金the Open Research Fund of Key Laboratory of Construction and Safety of Water Engineering of the Ministry of Water Resources,China Institute of Water Resources and Hydropower Research(NO.IWHR-ENGI-202305)Open project of Engineering Research Center of Concrete Technology under Marine Environment,Ministry of Education(2024KFKT-YB12)Shandong Youth Innovation Team(No.2023KJ324).
文摘This study presents an experimental investigation into the effects of microbial induced carbonate precipitation(MICP)treatment factors on the shear behavior of MICP-treated loess soil.Several groups of loess samples were prepared and subjected to MICP treatment with varying cementation reagent concentration,calcium source,and curing duration across three levels.The results indicate that the shear strength of MICP-treated loess achieves optimal performance when treated with the cementation reagent concentration of 1.0 M,cured for 14 days,and using calcium chloride as the calcium source.Compared to untreated loess,the cohesion and internal friction angle increased by approximately 77%and 26%,respectively.To evaluate the influence of these treatment variables,orthogonal analysis was performed on the obtained shear strength parameters.The analysis indicates that the cementation reagent concentration is the primary factor influencing shear strength,followed by the calcium source and curing duration.Additionally,scanning electron microscopy(SEM)tests were performed to investigate the microstructure of the MICP-treated samples.The results reveal that calcium carbonate significantly enhances the loess structure by creating large effective bonding areas,which in turn increases the bridging force.As a result,the overall shear strength of the treated loess shows a marked improvement compared to the untreated samples.
基金the financial support received from the National Natural Science Foundation of China(Grant Nos.42072317 and 41727802)the Science and Technology Commission of Shanghai Municipality(Funding No.21DZ1204300).
文摘Comprehensive investigations have been conducted to study the structure and overconsolidation of upper Shanghai clays, i.e. Layers 2–6 clays, typically located at depths of 30–40 m. However, limited information is available on their anisotropy, and even less is known about the correlation between structure, overconsolidation, and anisotropy. In this study, the undrained anisotropy characteristics of shear strength and small-strain shear stiffness in upper Shanghai Layers 2–6 clays were thoroughly assessed using a series of K0-consolidated undrained triaxial compression (TC) and triaxial extension (TE) tests (K0 is the coefficient of lateral earth pressure at rest). The effective stress paths, shear strength, and small-strain shear stiffness from the undrained TC and TE tests demonstrate the anisotropic behaviors in upper Shanghai clays. Analyses of data from upper Shanghai clays and other clays worldwide indicate that the shear strength anisotropy ratio (Ks) converges at 0.8 as the overconsolidation ratio (OCR) and plasticity index (Ip) increase, while the small-strain shear stiffness anisotropy ratio (Re) converges at 1.0. The influence of OCR on Ks and Re is more pronounced than that of Ip and sensitivity (St). Nevertheless, no clear correlation between Ks and Re is observed in upper Shanghai clays.
基金supported by the National Natural Science Foundation of China(Grant Nos.42272333 and 42377154)the Ningbo Natural Science Foundation(Grant No.2023J085).
文摘Joint surface roughness is comprised of two components:large-scale(waviness,first-order)and smallscale(unevenness,second-order).To investigate the scale effect of two-order roughness,a surface area method is used to separate the waviness and unevenness from the whole joint surface of each size.The results show that as the joint size increases,the first-order roughness exhibits a decrease,whereas the second-order roughness remains constant.This indicates that the roughness scale effect is primarily attributed to the waviness.Then,the influence of two-order roughness on the scale effect of joint shear strength is examined through a series of laboratory tests.The test results reveal a negative scale effect on the peak friction angle,which exhibits a strong correlation with the first-order roughness.The decrease in the peak friction angle with normal stress can be attributed primarily to the degradation of second-order roughness,which shows a negative exponential trend.Finally,an empirical model for estimating the peak friction angle of rock joints at different scales is proposed based on two-order roughness.The prediction accuracy of this model is validated by comparing the existing criteria.
文摘The interpretation and application of CPT(cone penetration test)results is characterized by considerable variability of data,either in measured or correlated parameters.According to the requirements of Eurocode 7 the existing variability in soil properties has to be taken into account statistically during the determination of the characteristic values of each parameter.This should be done by selecting a cautious estimate of the value affecting the limit state.Obtaining the characteristic values of CPT measurements is not an easy task and on this aspect nor clear neither unified guidelines exist.This paper focuses in several approaches to characterize the cone resistance and the sleeve friction using simple statistical analysis,in order for these parameters to be applicable in design.Similar procedures are then applied to determine the characteristic values of correlated parameters from CPT such as the effective friction angle for sands and the undrained shear strength for clays.The resulting characteristic values of the considered parameters emphasize the fact that the prediction and the interpretation of characteristic values of soil properties is a complicated and biased procedure.
文摘This opinion article discusses the original research work of Yünkül et al.(the Authors)published in the Journal of Mountain Science 21(9):3108–3122.Employing non-linear regression,fuzzy logic and artificial neural network modeling techniques,the Authors interrogated a large database assembled from the existing research literature to assess the performance of twelve equation rules in predicting the undrained shear strength(s_(u))mobilized for remolded fine-grained soils at different values of liquidity index(I_(L))and water content ratio.Based on their analyses,the Authors proposed a simple and reportedly reliable correlation(i.e.,Eq.9 in their paper)for predicting s_(u) over the I_(L) range of 0.15 to 3.00.This article describes various shortcomings in the Authors’assembled database(including potentially anomalous data and covering an excessively wide I_(L) range in relation to routine geotechnical and transportation engineering applications)and their proposed s_(u)=f(I_(L))correlation.Contrary to the Authors’assertions,their proposed correlation is not reliable for fine-grained soils with consistencies in the general firm to stiff range(i.e.,for 0.15<I_(L)<0.40),increasingly overestimating s_(u) for reducing I_(L),and eventually predicting s_(u)→+∞for I_(L)→0.15+(while producing mathematically undefined s_(u) for I_(L)<0.15),thus rendering their correlation unconservative and potentially leading to unsafe geotechnical designs.Exponential or regular-power type s_(u)=f(I_(L))models are more s_(u)itable when developing correlations that are applicable over the full plastic range(of 0<I_(L)<1),thereby providing reasonably conservative s_(u) predictions for use in the preliminary design for routine geotechnical engineering applications.
基金The authors are grateful for the Beijing Natural Science Foundation(Grant No.8242017)。
文摘This study presents an in-depth investigation into the shear strength characteristics of unsaturated soils,focusing on the influenceof shear rate and initial water saturation(S_(r0)).Utilizing the drained-vented(DV)triaxial test method,the present study investigated the shear strength behavior of silty clay under various shear rates and water saturation levels,and compared the outcomes with traditional suction-controlled(SC)and constant water content(CW)tests.The findingshighlight the pivotal role of excess pore water pressure dissipation during shearing,which significantlyaffects the shear strength of both saturated and unsaturated soils.Notably,for soils with high initial water saturation,a decrease in shear strength is observed with an increase in shear rate,which is attributed to the rise in pore water pressure.Conversely,for soils with low initial water saturation,the shear rate exhibits minimal impact on shear strength due to negligible water drainage.The research delineates the optimal shear rates for DV tests based on the initial water saturation:a slower rate of 0.0028 mm/min for samples with high water saturation(S_(r0)>66%)and a faster rate of 0.8 mm/min for samples with low water saturation(S_(r0)≤66%).A novel testing methodology for determining unsaturated soil shear strength under DV conditions is introduced,streamlining the measurement process and significantly reducing testing time.This method not only promises substantial cost savings but also aligns closely with natural engineering conditions,offering valuable guidance for geotechnical applications.
基金funded by the National Natural Science Foundation Projects(Grant Nos.41772287 and 42277132)the Key R&D Project of Zhejiang Province(Grant No.2021C03159).
文摘The scale effect on shear strength of rock joints is well-documented.However,whether scale effects are negative,positive,or even exist or not is still controversial.Joint roughness significantly influences the shear strength of rock joints.Compared to the shear tests,using the joint roughness coefficient(JRC)and its roughness parameters offers a more convenient method for describing the scale effect on shear strength.However,it is crucial to understand that the scale effect mechanisms of JRC are distinct from those of shear strength.Therefore,this paper aims to clarify these distinct mechanisms.By digitally extracting roughness parameters from granite samples,it is found that the scale effect of roughness parameters mainly comes from the sampling methods and the geometric characteristics of parameters.Furthermore,a full data sampling method considering heterogeneity is proposed to obtain more representative roughness parameters.To reveal the scale effect mechanisms of shear strength,Gaussian filtering is firstly used to separate the waviness and unevenness components of roughness,facilitating a deeper understanding of the geometric characteristics of roughness.It is suggested that the wavelength of the waviness component can reflect the scale effect on shear strength.Secondly,numerical simulations of ideal artificial joint models are conducted to validate that the wavelength of the waviness component serves as the dividing point between positive and negative scale effects.The mechanical mechanisms of positive and negative scale effects are also interpreted.Finally,these mechanisms successfully elucidate the occurrence patterns of the scale effect on natural joint profiles.
基金supported by the National Natural Science Foundation of China(NO.32201626)the Key Research and Development Program of Jiangxi Province(20223BBG74S01,20223BBG71013).
文摘In response to the effectiveness of reforestation in controlling soil erosion,there has been a dramatic increase in forest coverage in the hilly red soil region of southern China.Aggregate stability and soil shear strength are indicators that reflect soil resistance to erosion and its ability to prevent shallow landslides,respectively.However,limited research has focused on the response of soil aggregate stability and shear strength to reforestation.We selected three types of reforestations(Phyllostachys edulis forest,Cunninghamia lanceolata(Lamb.)Hook.forest,Citrus sinensis(L.)Osbeck.orchard),a natural forest(mixed coniferous and broadleaf forests),and a fallow land as study plots,and measured root traits,and soil physicochemical traits,i.e.,pH,soil organic matter(SOC),Soil water content(SWC),soil bulk density(BD),soil cohesion(c),soil internal friction angle(φ)and analyzed their multiple interactions.The soil aggregate stability traits,refer to the mean weight diameter(MWD)and geometric mean diameter(GMD),exhibited a significant increase in reforested plots,approximately 200%compared to fallow land and 50%compared to natural forests.For soil shear strength the values were approximately 20%higher than in fallow land and approximately 10%lower than in natural forests.Soil aggregate stability and soil shear strength did not exhibit a significant positive correlation across all plots,and the underlying drivers of these traits were variable.For instance,in natural forest and timber stands,soil aggregate stability was mainly influenced by soil organic carbon,while soil shear strength was primarily affected by root length density.In economic forest,aggregate stability and shear strength are mainly affected by organic carbon.Overall,we found that vegetation restoration enhances soil erosion resistance,however,the primary drivers for the improvement of aggregate stability(soil organic carbon)and shear strength(root length density)are different.Therefore,in future benefit assessments of vegetation restoration projects aimed at soil erosion control,different indicators should be considered based on specific conditions.
基金funded by the National Science Fund for Distinguished Young Scholars(Grant No.42225206)the National Key R&D Program of China(Grant No.2020YFC1807200)the National Natural Science Foundation of China(Grant No.42072299).
文摘Conventional empirical equations for estimating undrained shear strength(s_(u))from piezocone penetration test(CPTu)data,without incorporating soil physical properties,often lack the accuracy and robustness required for geotechnical site investigations.This study introduces a hybrid virus colony search(VCS)algorithm that integrates the standard VCS algorithm with a mutation-based search mechanism to develop high-performance XGBoost learning models to address this limitation.A dataset of 372 seismic CPTu and corresponding soil physical properties data from 26 geotechnical projects in Jiangs_(u)Province,China,was collected for model development.Comparative evaluations demonstrate that the proposed hybrid VCS-XGBoost model exhibits s_(u)perior performance compared to standard meta-heuristic algorithm-based XGBoost models.The res_(u)lts highlight that the consideration of soil physical properties significantly improves the predictive accuracy of s_(u),emphasizing the importance of considering additional soil information beyond CPTu data for accurate s_(u)estimation.
基金Ho Chi Minh City University of Technology (HCMUT), VNU-HCM for supporting this study
文摘Machine learning(ML)models are widely used for predicting undrained shear strength(USS),but interpretability has been a limitation in various studies.Therefore,this study introduced shapley additive explanations(SHAP)to clarify the contribution of each input feature in USS prediction.Three ML models,artificial neural network(ANN),extreme gradient boosting(XGBoost),and random forest(RF),were employed,with accuracy evaluated using mean squared error,mean absolute error,and coefficient of determination(R^(2)).The RF achieved the highest performance with an R^(2) of 0.82.SHAP analysis identified pre-consolidation stress as a key contributor to USS prediction.SHAP dependence plots reveal that the ANN captures smoother,linear feature-output relationships,while the RF handles complex,non-linear interactions more effectively.This suggests a non-linear relationship between USS and input features,with RF outperforming ANN.These findings highlight SHAP’s role in enhancing interpretability and promoting transparency and reliability in ML predictions for geotechnical applications.
基金Projects (50975062, 51105107, 51021002) supported by the National Natural Science Foundation of ChinaProjects (QC2011C044) supported by the Natural Science Foundation of Heilongjiang Province,China+1 种基金Project (20112302130005) supported by the Specialized Research Fund for the Doctoral Program of Higher Education,ChinaProject (CUGA4130902510) supported by the China Postdoctoral Science Foundation Funded
文摘Reactive brazing of TiAl-based intermetallics and Ni-based alloy with Ti foil as interlayer was investigated. The interfacial microstructure and shear strength of the joints were studied. According to the experimental observations, the molten interlayer reacts vigorously with base metals, forming several continuous reaction layers. The typical interfacial microstructure of the joint can be expressed as GH99/（Ni,Cr）ss（γ）/TiNi（β2）＋TiNi2Al（τ4）＋Ti2Ni（δ）/δ＋Ti3Al（α2）＋Al3NiTi2（τ3）/α2＋τ3/TiAl. The maximum shear strength is 258 MPa for the specimen brazed at 1000°C for 10 min. Higher brazing temperature or longer brazing time causes coarsening of the phases in the brazing seam and formation of brittle intermetallic layer, which greatly depresses the shear strength of the joints.
基金Project(51275416)supported by the National Natural Science Foundation of China
文摘The bonding interface characteristic and shear strength of diffusion bonded Ti-17 titanium alloy at different bonding time were investigated. The results show that the average size of voids decreases while the amount of voids decreases after increasing to the maximum value with the increasing bonding time. The irregular void with a scraggly edge tends to an ellipse void with smooth surface and then changes to a tiny void with round shape. The grains across bonding interface occur at bonding time of 60 min. The shear strength of bond increases with increasing bonding time, and the highest shear strength of bond is 887.4 MPa at 60 min. The contribution of plastic deformation on the void closure and the increase of shear strength is significant even though the action time of plastic deformation is short.
基金The National Natural Science Foundation of China(No.40702047)
文摘In order to obtain the reasonable undrained shear strength Su for geotechnical analyses of bridge foundations in Yangtze River floodplain clayey soils, a site-specific study is conducted using the imported piezocone penetration test （CPTu） with dissipation phases at the Fourth Nanjing Yangtze River Bridge construction sites. Taking the values of Su from laboratory tests as references, several existing Su-predicted methods based on CPTu are compared and evaluated. To verify the presented cone factor Nk, additional test sites are selected and examined. The results show that the values of cone factors such as Nkt, Nke, and Nau, depend on the shear test mode and disturbance. Generally, the values of Nke show more scattering than those of Nkt and N△u. For the stratified and layered sediments of the Yangtze River floodplain, it is recommended using the net cone resistance qT to estimate Su and the preliminary cone factor values Nkt are from 7 to 16, with an average of 11. It is also confirmed that the CPTu test, as a new technique in site characterization, can present reasonable parameters for bridge foundations.
基金Project(51174228)supported by the National Natural Science Foundation of ChinaProject(CX2012B069)supported by Hunan Provincial Innovation Foundation for PostgraduateProject(201003)supported by Transportation Science and Technology Projects of Hunan Province,China
文摘A rigorous back analysis of shear strength parameters of landslide slip was presented. Kinematical element method was adopted to determine factor of safety and critical failure surface, which overcomes the disadvantage of limit equilibrium method. The theoretical relationship between the combination of shear strength parameters and stability state was studied. The results show that the location of critical slip surface, F/tan f and F/c depend only on the value of c/tan f. The failure surface moves towards the inside of slope as c/tan f increases. According to the information involving factor of safety and critical failure surface in a specific cross-section, strength parameters can be back calculated based on the above findings. Three examples were given for demonstrating the validity of the present method. The shear strength parameters obtained by back analysis are almost consistent with their correct solutions or test results.
文摘For sites susceptible to liquefaction induced lateral spreading during a probable earthquake, geotechnical engineers often need to know the undrained residual shear strength of the liquefied soil deposit to estimate lateral spreading displacements, and the forces acting on the piles from the liquefied soils in order to perform post liquefaction stability analyses. The most commonly used methods to estimate the undrained residual shear strength （Su~） of liquefied sand deposits are based on the correlations determined from liquefaction induced flow failures with SPT and CPT data. In this study, 44 lateral spread case histories are analyzed and a new relationship based on only lateral spread case histories is recommended, which estimates the residual shear strength ratio of the liquefiable soil layer from normalized shear wave velocity. The new proposed method is also utilized to estimate the residual lateral displacement of an example bridge problem in an area susceptible to lateral spreading in order to provide insight into how the proposed relationship can be used in geotechnical engineering practice.
基金financial support from High-end Foreign Expert Introduction program(No.G20190022002)Chongqing Construction Science and Technology Plan Project(2019-0045)as well as Chongqing Engineering Research Center of Disaster Prevention&Control for Banks and Structures in Three Gorges Reservoir Area(Nos.SXAPGC18ZD01 and SXAPGC18YB03)。
文摘Accurate assessment of undrained shear strength(USS)for soft sensitive clays is a great concern in geotechnical engineering practice.This study applies novel data-driven extreme gradient boosting(XGBoost)and random forest(RF)ensemble learning methods for capturing the relationships between the USS and various basic soil parameters.Based on the soil data sets from TC304 database,a general approach is developed to predict the USS of soft clays using the two machine learning methods above,where five feature variables including the preconsolidation stress(PS),vertical effective stress(VES),liquid limit(LL),plastic limit(PL)and natural water content(W)are adopted.To reduce the dependence on the rule of thumb and inefficient brute-force search,the Bayesian optimization method is applied to determine the appropriate model hyper-parameters of both XGBoost and RF.The developed models are comprehensively compared with three comparison machine learning methods and two transformation models with respect to predictive accuracy and robustness under 5-fold cross-validation(CV).It is shown that XGBoost-based and RF-based methods outperform these approaches.Besides,the XGBoostbased model provides feature importance ranks,which makes it a promising tool in the prediction of geotechnical parameters and enhances the interpretability of model.
