Application of machine learning to the Vs-based soil liquefaction potential assessment

Earthquakes can cause violent liquefaction of the soil, resulting in unstable foundations that can cause serious damage to facilities such as buildings, roads, and dikes. This is a primary cause of major earthquake disasters. Therefore, the discrimination and prediction of earthquake-induced soil liquefaction has been a hot issue in geohazard research. The soil liquefaction assessment is an integral part of engineering practice. This paper evaluated a dataset of 435 seismic sand liquefaction events using machine learning algorithms. The dataset was analyzed using seven potential assessment parameters. Ten machine learning algorithms are evaluated for their ability to assess seismic sand liquefaction potential, including Linear Discriminant Analysis(LDA), Quadratic Discriminant Analysis(QDA), Naive Bayes(NB), KNearest Neighbor(KNN), Artificial Neural Network(ANN), Classification Tree(CT), Support Vector Machine(SVM), Random Forest(RF), e Xtreme Gradient Boosting(XGBoost), Light Gradient Boosting Machine(Light GBM). A 10-fold cross-validation(CV) method was used in the modeling process to verify the predictive performance of the machine learning models. The final percentages of significant parameters that influenced the prediction results were obtained as Cyclic Stress Ratio(CSR) and Shear-Wave Velocity( VS1) with 56% and 38%, respectively. The final machine learning algorithms identified as suitable for seismic sand liquefaction assessment were the CT, RF, XGBoost algorithms, with the RF algorithm performing best.

Adaptive sampling strategy for characterizing spatial distribution of soil liquefaction potential using cone penetration test

Characterizing spatial distribution of soil liquefaction potential is critical for assessing liquefactionrelated hazards(e.g.building damages caused by liquefaction-induced differential settlement).However,in engineering practice,soil liquefaction potential is usually measured at limited locations in a specific site using in situ tests,e.g.cone penetration tests(CPTs),due to the restrictions of time,cost and access to subsurface space.In these cases,liquefaction potential of soil at untested locations requires to be interpreted from limited measured data points using proper interpolation method,leading to remarkable statistical uncertainty in liquefaction assessment.This underlines an important question of how to optimize the locations of CPT soundings and determine the minimum number of CPTs for achieving a target reliability level of liquefaction assessment.To tackle this issue,this study proposes a smart sampling strategy for determining the minimum number of CPTs and their optimal locations in a selfadaptive and data-driven manner.The proposed sampling strategy leverages on information entropy and Bayesian compressive sampling(BCS).Both simulated and real CPT data are used to demonstrate the proposed method.Illustrative examples indicate that the proposed method can adaptively and sequentially select the required number and optimal locations of CPTs.

An improved method of evaluating liquefaction potential with the velocity of shear-waves

According to the results of cyclic triaxial tests, a linear correlation is presented between liquefaction resistance and elastic shear modulus, which shows the relation of Gmax (kPa) with (σd/2)1/2(kPa)1/2. When applied to soils from different sites, the correlation can be normalized in reference to its minimum void ratio (emin). Accordingly, an improved method is established to evaluate the liquefaction potential with shear-wave velocity. The critical shear-wave velocity of liquefaction is in linear relation with 1/4 power of depth and the maximum acceleration during earthquakes, which can be used to explain the phenomenon that the possibility of liquefaction decreases with the increment of the depth. Compared with previous methods this method turns out simple and effective, which is also verified by the results of cyclic triaxial tests.

The Hidden Earthquake Induced Liquefaction Risks in the Rohingya Refugee Camp Hills & Surrounding Areas of Ukhiya, Cox’s Bazar, Bangladesh—A Geotechnical Engineering Approach

Liquefaction is one of the major catastrophic geohazards which usually occurs in saturated or partially saturated sandy or silty soils during a seismic event. Evaluating the potential liquefaction risks of a seismically prone area can significantly reduce the loss of lives and damage to civil infrastructures. This research is mainly focused on the earthquake-induced liquefaction risk assessment based on Liquefaction Potential Index (LPI) values at different earthquake magnitudes (M = 5.0, 7.0 and 8.0) with a peak ground acceleration (amax) of 0.28 g in the Rohingya Refugee camp and surrounding areas of Ukhiya, Cox’s Bazar, Bangladesh. Standard Penetration Test (SPT) results have been evaluated for potential liquefaction assessment. The soils are mainly composed of very loose to loose sands with some silts and clays. Geotechnical properties of these very loose sandy soils are very much consistent with the criteria of liquefiable soil. It is established from the grain size analysis results;the soil of the study area is mainly sand dominated (SP) with some silty clay (SC) which consists of 93.68% to 99.48% sand, 0.06% to 4.71% gravel and 0% to 6.26% silt and clay. Some Clayey Sand (SC) is also present. The silty clay can be characterized as medium (CI) to high plasticity (CH) inorganic clay soil. LPI values have been calculated to identify risk zones and to prepare risk maps of the investigated area. Based on these obtained LPI values, four (4) susceptible liquefaction risk zones are identified as low, medium, high and very high. The established “Risk Maps” can be used for future geological engineering works as well as for sustainable planning, design and construction purposes relating to adaptation and mitigation of seismic hazards in the investigated area.

Dynamic properties and liquefaction behaviour of cohesive soil in northeast India under staged cyclic loading

Estimation of strain-dependent dynamic soil properties, e.g. the shear modulus and damping ratio, along with the liquefaction potential parameters, is extremely important for the assessment and analysis of almost all geotechnical problems involving dynamic loading. This paper presents the dynamic properties and liquefaction behaviour of cohesive soil subjected to staged cyclic loading, which may be caused by main shocks of earthquakes preceded or followed by minor foreshocks or aftershocks, respectively. Cyclic triaxial tests were conducted on the specimens prepared at different dry densities （1.5 g/cm3 and 1.75 g/cm3） and different water contents ranging from 8% to 25%. The results indicated that the shear modulus reduction （G/Gmax） and damping ratio of the specimen remain unaffected due to the changes in the initial dry density and water content. Damping ratio is significantly affected by confining pressure, whereas G/Gmax is affected marginally. It was seen that the liquefaction criterion of cohesive soils based on single-amplitude shear strain （3.75% or the strain at which excess pore water pressure ratio becomes equal to 1, whichever is lower） depends on the initial state of soils and applied stresses. The dynamic model of the regional soil, obtained as an outcome of the cyclic triaxial tests, can be successfully used for ground response analysis of the region.

Intensity measures for seismic liquefaction hazard evaluation of sloping site

This work investigates the correlation between a large number of widely used ground motion intensity measures(IMs) and the corresponding liquefaction potential of a soil deposit during earthquake loading. In order to accomplish this purpose the seismic responses of 32 sloping liquefiable site models consisting of layered cohesionless soil were subjected to 139 earthquake ground motions. Two sets of ground motions, consisting of 80 ordinary records and 59 pulse-like near-fault records are used in the dynamic analyses. The liquefaction potential of the site is expressed in terms of the the mean pore pressure ratio, the maximum ground settlement, the maximum ground horizontal displacement and the maximum ground horizontal acceleration. For each individual accelerogram, the values of the aforementioned liquefaction potential measures are determined. Then, the correlation between the liquefaction potential measures and the IMs is evaluated. The results reveal that the velocity spectrum intensity(VSI) shows the strongest correlation with the liquefaction potential of sloping site. VSI is also proven to be a sufficient intensity measure with respect to earthquake magnitude and source-to-site distance, and has a good predictability, thus making it a prime candidate for the seismic liquefaction hazard evaluation.

Liquefaction Assessment in Golem Area, Albania, by Means of In-situ Tests

The liquefaction is a very significant phenomenon in clayey silty soils, silty sands and also sands. The high potential of liquefaction is generally recognized when these types of soils are laid under the hydrostatic water table. Low plasticity silts, silty sands and sands are found as recent alluvial deposits in the western coastal part of Albania, especially in the sandy beaches of Adriatic Sea near Durres City. The aim of this study is to evaluate the soil liquefaction potential in the area of Golem. Ten CPTUs （cone penetration test with pore pressure measurements） are carried out for the site investigation of soils. In this paper, results of the CPTU based liquefaction analysis are presented. The data of two CPTUs （10 in total） are analyzed and factor of safety was found by considering different levels of hazard and ground water. The results of liquefaction potential analysis show that the soils in the area of Golem have a high risk of liquefaction.

Soil Liquefaction Hazard Evaluation by the New Seismic Dilatometer Marchetti Test （SDMT）

Soil liquefaction, and the permanent deformations that frequently result from it, have caused significant damage in past earthquakes. The procedures used in contemporary geotechnical earthquake engineering practice are generally based on simplified procedures for the evaluation of the liquefaction potential. The work describes a framework for performance-based earthquake engineering and its use in the development of a performance-based procedure for liquefaction hazard evaluation. The performance-based procedure will be used to show how consistent application of conventional procedures for evaluation of liquefaction potential can influence performance prediction. Implications for liquefaction-resistant design will also be discussed. The purpose is to summarize current procedures for practical prediction of liquefaction behavior, to describe recent advances in the understanding of liquefaction behavior, and to describe the incorporation of this improved understanding into new solutions for detailed modeling of soil liquefaction, Simplified procedures for evaluation of liquefaction hazards will be reviewed relatively briefly, with more details devoted to emerging knowledge about the mechanics of liquefiable soil behavior, and methods for incorporating those mechanics into improved models for performance prediction. In particular it focuses about the influence on the evaluation of Cyclic Resistance Ratio （CRR） by different in-situ tests （Cone Penetration Test （CPT）. Standard Penetration Test （SPT） and Seismic Dilatometer Marchetti Test （SDMT）） and by different shear waves velocity measurements （Down Hole D-H. Cross Hole C-H, Seismic Dilatometer Marchetti Test SDMT）.

中国模式下砾性土液化指数评价新方法

砾性土地震液化是我国土木工程震害防治技术体系建设面临的新问题,砾性土液化指数作为砾性土液化判别的最终表征量及抗液化对策的实际控制指标,是当前体系发展中亟待突破的瓶颈。该文继承中国模式,基于2008年我国汶川地震实测数据,提出以超重型圆锥动力触探击数比余量为基本变量的砾性土液化指数评价新方法。采用我国独有资料,构建砾性土的理论计算液化指数与宏观调查液化指数间函数关系,分析我国规范中既有液化指数与液化等级对应关系的可行性,建立适于砾性土液化指数与液化等级的对等关系,以实测资料多方论证新方法的合理性。提出的新方法,保持了中国模式的优势,可为实际工程应用和我国规范修订提供直接的技术支持。

黄土液化作用及其次生灾害风险评估方法初探——以积石山M_(S)6.2地震为例

2023年12月18日甘肃积石山6.2级地震引发的黄土液化泥石流地质灾害,造成了大量人员伤亡和财产损失。以此为背景,系统梳理了液化调查、风险评估和风险区划的工作思路:选定地区液化的内在和诱发因子,收集并分析资料,拟定调查工作底图,以调查问卷的形式开展液化影响因素摸底工作,开展针对性的外业调查或辅以物探和钻探工作,得到土壤液化发育潜势实际材料图;其后借用常规地质灾害的评估方法,开展调查区域的易发性、易损性评价和风险评价,得到调查区风险区划图。利用此方法对积石山地震诱发的泥石流灾害进行验证,分析其液化灾害的地质环境背景、形成机理和危害特征,并对其发生的风险进行评价。文章首次提出土的液化作用及其次生灾害区域调查因子的选择和风险评估的流程,可为开展土层液化作用及其次生灾害隐患区排查、国土空间规划,特别是高烈度山区集中居住点地质安全评估和选址建议提供参考。

地下水位上升对北京土层地震液化的影响

地下水位是砂土地震液化的重要影响因素,南水北调中线工程全线贯通后北京市地下水位持续上升。收集99个钻孔资料和122个地下水位监测站数据,采用基于最佳拟合分布的克里金插值方法,构建北京研究区域地下水位埋深和地层空间分布模型,分析地下水位变化对北京土层地震液化的影响。取地震加速度为0.2和0.4g,采用两种基于标准贯入试验的液化判别方法,在北京市平均水位上升1m、3m、5m的情况下,计算液化指数的空间分布和区域液化等级划分。结果表明:随着北京市地下水位的上升,研究区域液化面积增大,液化等级整体有所提升。地震加速度为0.4g时,规范法计算得到地下水位升高5m,液化面积占比从30.8%提升至70.7%。该液化危险性评价为北京市岩土工程建设提供了有力指导,也为全球类似地下水位上升地区提供了计算方法。

饱和黄土隧道围岩地震液化特征的振动台试验研究

以兰合铁路某在建隧道为工程背景,对其黄土段开展了1∶20大比例尺饱和黄土围岩地震液化的系列振动台试验,揭示了土-结构相互作用对隧道衬砌周围土体液化势的影响规律。研究结果表明:土-结构相互作用显著提高了隧道衬砌周围土体的液化势,当遭遇地震荷载时,围岩会先于同一水平地层土体而产生液化;激励地震加速度a_(max)的较小时(a_(max)=1.0 m/s^(2)),同一水平高度靠近衬砌结构土体的动剪应力峰值ι_(d,maxt)、动孔隙水压力峰值d,U_(max)、和加速度峰值a_(p,max)都要比远离衬砌结构的大,具有明显的放大效应;随着激励地震加速度a_(max)的增大至接近围岩液化阶段时,动孔隙水压力U_(d)的放大效应明显,而剪应力ι_(d,maxt)和加速度a_(p,max)的放大效应减小或者消失。同时上述结果中围岩土体加速度的变化特征,揭示了前期相关数值分析研究存在的不足。

基于液化势指标的砂土抗液化评价方法及应用

在散粒体最大最小孔隙比理论的基础上,采用一种新的液化势指标对不同颗粒级配的二元砂粉混合物进行抗液化能力预测。通过对已有文献数据中砂粉混合物的体积应变势进行计算,分析混合物的细粒含量、粒径比与循环抗力比CRR的关系,结合混合物所处实际孔隙比提出了一个新的液化势指标ε′v。与砂粉混合物的CRR试验数据对比分析显示:各类砂粉混合物的CRR均随ε′v的增大而单调地降低,且两者呈现良好的幂函数关系,并进一步建立了CRR-ε′v关系曲线最佳拟合参数与粒径比的关系。对蒙特利砂混合耶茨维尔粉土试样进行的CRR预测结果表明,ε′v综合反映了土粒径分布、密实状态和颗粒形状的综合影响,是表征砂粉混合物CRR的一个有效指标。实际工程中可代替传统的现场原位试验,仅通过简单的室内试验预测场地土层抗液化能力。

日本地震砂土液化破坏案例及抗液化处置措施

砂土液化是地震灾害的主要形式之一。在振动作用下,饱和的松散粉、细砂土中孔隙水压力急剧增大,当有效应力降为零时砂土颗粒处于悬浮状态,即发生砂土液化。砂土液化可引起地基强度丧失,从而加重建筑物破坏程度,使重要运输生命线瘫痪,造成巨大经济损失和人员伤亡。本文梳理了砂土液化的历史文献记录,阐述了砂土液化产生的机制,概述了砂土液化造成的各种破坏,并介绍了日本判别砂土液化的最新方法、抗液化处置措施及一些成功案例,以期为砂土液化研究及抗液化处置措施的改进提供有益参考。

Dynamic characterization of tailing dam using fully coupled dynamic analysis with different boundary conditions-a case study

The seismic response analysis of a tailing dam is studied using a fully coupled effective stress approach in conjunction with an advanced multi yield surface plastic constitutive model for tailing material.Strain controlled static and cyclic triaxial tests were carried out to obtain the constitutive model for the tailing material.The tailing materials were collected from the Rampura Agucha tailing dam(Rajasthan State,India).A 2D nonlinear finite element(FE)model was then developed using different boundary conditions from the tailing embankment constructed using the downstream and upstream method of rising using OpenSees software.In first case,the model boundary was fixed in both the X and Y directions,and in the second case,viscous dashpots were introduced for both side and horizontal boundaries.The model was validated with experimental results on tailing material.Analyses were carried out considering five different earthquake motions,which were applied at the base.Comparisons of the different boundary conditions in terms of displacement flow vectors,pore pressure and stress-strain curves during shaking are presented.From the analysis,it was observed that the viscous boundary condition replicates the actual field conditions more accurately than the fixed boundary condition.In addition,it was found that the tailing embankment constructed by the downstream and upstream method of rising is not susceptible to liquefaction and lateral spreading for earthquake motions,even for a magnitude>5.5.

场地液化势评价概率模型

本文提出了场地液化势评价概率模型。基于世界范围内多次震后场地液化性状实测资料 ,采用人工神经网络 (ANN)来搜索标贯击数Ncr与抗液化能力之间的非线性关系 ,进而导出安全系数与液化概率之间的关系。建立的修正液化指数公式可对可液化场地进行液化概率评价。修正液化指数有明确的概率意义 ,适用于饱和砂土等可液化场地液化势评价的风险决策。

瞬态瑞利面波勘探技术在公路工程勘察中的应用

结合工程实例介绍了一种新的勘探技术———瞬态瑞利面波法 ,并总结了测试技术要点 ,包括测试分析系统、测试条件等 ,最后分析了应用效果及应用中的一些问题。

上海苏州河故道砂质粉土液化综合判别

本文根据工程实际的大量现场测试和室内试验 ,提出了上海苏州河故道砂质粉土综合运用宏观评判和微观定量判定进行液化评价的原则和方法。对上海地区地基土液化判别有一定的实际意义。

淮安市砂土地震液化势的综合评估

建立了BP神经网络模型LM算法的砂土地震液化判别方法,采用LVQ模式分类网络对数据进行分类,选取唐山地震时76个场地、320组现场液化勘察资料为研究对象,训练和检验网络模型的数据各160组,结果表明,该方法的砂土液化预测准确度为96.8%。根据淮安市典型场地土的钻孔资料,采用《建筑抗震设计规范法》液化判别法(规范法)、美国国家地震研究中心建议的液化判别简化方法(NCEER法)和谢君斐-陈国兴液化判别方法(谢-陈法)等3种液化判别方法及本文提出的BP神经网络模型液化判别方法对22个钻孔、120标贯点进行了液化判别,结果表明:14种地震液化判别方法的计算结果基本一致;2NCEER法比谢-陈法的计算结果要保守;3谢-陈法与规范法的计算结果很吻合,预测正确率相差不大;4规范法相对BP神经网络模型法的计算结果要保守;5BP神经网络模型法比其他3种方法的预测结果更符合实际的液化规律。最后,综合4种方法的液化判别结果给出了该地区地震液化势的评估结果,并给出了部分典型地质剖面土层地震液化势分布图。

地基砂土液化判别方法探讨

本文以我国《水利水电工程地质勘察规范》(GB 50287-2008)中的标准贯入锤击数液化判别法为基础,以该法进行应力修正的方法为依据,提出了采用附加应力等效埋深来考虑有限面积基础建筑物附加应力随深度扩散影响的液化判别新思路,以工程设计中参数取值常用平均值和小值平均值来考虑标贯击数和黏粒含量的离散性,同时借用有限元网格离散求解思路确定砂层可能液化的区域。以宁朗水电站闸坝工程为例进行了案例分析,结果表明:在地震设防烈度VII度时,闸基下中细砂层近震时不发生液化,远震时在闸基4个角点下的局部小区域内发生液化。