Modeling of SPT Seismic Liquefaction Data Using Minimax Probability Machine

The determination of seismic liquefaction potential of soil is an important task in geotechnical engineering. This article uses Minimax Probability Machine (MPM) for determination of seismic liquefaction potential of soil based on Standard Penetration Test value (N). MPM is developed based on the use of hyperplanes. It is a discriminant classifier. This study uses MPM as a classification tool. MPM uses the database collected from Chi–Chi earthquake. Two models (MODEL I and MODEL II) have been developed. MODEL I uses Cyclic Stress Ratio and N as input variables. Peck Ground Acceleration and N have been adopted as inputs for MODEL II. The performance of MODEL I and MODEL II are 97.67 and 96.51 % respectively. The performance of MODEL II is 94.11 % for the global data. The developed MPM shows good generalization capability. The results show that the developed MPM has ability for determination of seismic liquefaction potential of soil.     

Earthquake-induced liquefaction hazard mapping at national-scale in Australia using deep learning techniques

Australia is a relatively stable continental region but not tectonically inert, having geological conditions that are susceptible to liquefaction when subjected to earthquake ground motion. Liquefaction hazard assessment for Australia was conducted because no Australian liquefaction maps that are based on modern AI techniques are currently available. In this study, several conditioning factors including Shear wave velocity (Vs30), clay content, soil water content, soil bulk density, soil thickness, soil pH, distance from river, slope and elevation were considered to estimate the liquefaction potential index (LPI). By considering the Probabilistic Seismic Hazard Assessment (PSHA) technique, peak ground acceleration (PGA) was derived for 50 yrs period (500 and 2500 yrs return period) in Australia. Firstly, liquefaction hazard index (LHI) (effects based on the size and depth of the liquefiable areas) was estimated by considering the LPI along with the 2% and 10% exceedance probability of earthquake hazard. Secondly, ground acceleration data from the Geoscience Australia projecting 2% and 10% exceedance rate of PGA for 50 yrs were used in this study to produce earthquake induced soil liquefaction hazard maps. Thirdly, deep neural networks (DNNs) were also exerted to estimate liquefaction hazard that can be reported as liquefaction hazard base maps for Australia with an accuracy of 94% and 93%, respectively. As per the results, very-high liquefaction hazard can be observed in Western and Southern Australia including some parts of Victoria. This research is the first ever country-scale study to be considered for soil liquefaction hazard in Australia using geospatial information in association with PSHA and deep learning techniques. This study used an earthquake design magnitude threshold of M_w 6 using the source model characterization. The resulting maps present the earthquake-triggered liquefaction hazard and are intending to establish a conceptual structure to guide more detailed investigations as may be required in the future. The limitations of deep learning models are complex and require huge data, knowledge on topology, parameters, and training method whereas PSHA follows few assumptions. The advantages deal with the reusability of model codes and its transferability to other similar study areas. This research aims to support stakeholders’ on decision making for infrastructure investment, emergency planning and prioritisation of post-earthquake reconstruction projects.     

Least square support vector machine and relevance vector machine for evaluating seismic liquefaction potential using SPT

The determination of liquefaction potential of soil is an imperative task in earthquake geotechnical engineering. The current research aims at proposing least square support vector machine (LSSVM) and relevance vector machine (RVM) as novel classification techniques for the determination of liquefaction potential of soil from actual standard penetration test (SPT) data. The LSSVM is a statistical learning method that has a self-contained basis of statistical learning theory and excellent learning performance. RVM is based on a Bayesian formulation. It can generalize well and provide inferences at low computational cost. Both models give probabilistic output. A comparative study has been also done between developed two models and artificial neural network model. The study shows that RVM is the best model for the prediction of liquefaction potential of soil is based on SPT data.     

Elementary empirical model to assess seismic soil liquefaction

Liquefaction potential (LP) assessment plays a significant role in damages due to earthquake. The spirit underlying the present work is the evaluation of LP by correlating most significant parameters reflecting the dynamic response of soil with actual field behavior wherein an attempt of integrating the effect of dynamic soil properties and ground motion parameters simulating the actual site conditions is being made. Accordingly, a dynamic response–based Elementary Empirical Liquefaction Model (EELM) is proposed by processing a total of 314 reported case records covering a wide range of parameters demarcating “yes” and “no” zones of liquefaction. The method to develop the EELM essentially consists of evaluation of liquefaction potential, defining functional form of EELM representing dynamic response of soil to earthquake shaking, collection of data, computation of model parameters and formulation followed by validation of the model. The proposed empirical model though in fundamental form is found to perform fairly well resulting into an overall success rate of 86 % for both liquefaction and non-liquefaction points with significantly high success rate of 98 % for liquefied cases. Comparison of predictive performance of the proposed EELM with other approaches shows higher efficiency and thus signifies the theme of employing integrated approach.     

Assessment and Prediction of Liquefaction Potential Using Different Artificial Neural Network Models: A Case Study

Soil liquefaction as a transformation of granular material from solid to liquid state is a type of ground failure commonly associated with moderate to large earthquakes and refers to the loss of strength in saturated, cohesionless soils due to the build-up of pore water pressures and reduction of the effective stress during dynamic loading. In this paper, assessment and prediction of liquefaction potential of soils subjected to earthquake using two different artificial neural network models based on mechanical and geotechnical related parameters (model A) and earthquake related parameters (model B) have been proposed. In model A the depth, unit weight, SPT-N value, shear wave velocity, soil type and fine contents and in model B the depth, stress reduction factor, cyclic stress ratio, cyclic resistance ratio, pore pressure, total and effective vertical stress were considered as network inputs. Among the numerous tested models, the 6-4-4-2-1 structure correspond to model A and 7-5-4-6-1 for model B due to minimum network root mean square errors were selected as optimized network architecture models in this study. The performance of the network models were controlled approved and evaluated using several statistical criteria, regression analysis as well as detailed comparison with known accepted procedures. The results represented that the model A satisfied almost all the employed criteria and showed better performance than model B. The sensitivity analysis in this study showed that depth, shear wave velocity and SPT-N value for model A and cyclic resistance ratio, cyclic stress ratio and effective vertical stress for model B are the three most effective parameters on liquefaction potential analysis. Moreover, the calculated absolute error for model A represented better performance than model B. The reasonable agreement of network output in comparison with the results from previously accepted methods indicate satisfactory network performance for prediction of liquefaction potential analysis.     

Evaluation of liquefaction potential based on CPT results using evolutionary polynomial regression

In this paper a new approach is presented, based on evolutionary polynomial regression (EPR), for determination of liquefaction potential of sands. EPR models are developed and validated using a database of 170 liquefaction and non-liquefaction field case histories for sandy soils based on CPT results. Three models are presented to relate liquefaction potential to soil geometric and geotechnical parameters as well as earthquake characteristics. It is shown that the EPR model is able to learn, with a very high accuracy, the complex relationship between liquefaction and its contributing factors in the form of a function. The attained function can then be used to generalize the learning to predict liquefaction potential for new cases not used in the construction of the model. The results of the developed EPR models are compared with a conventional model as well as a number of neural network-based models. It is shown that the proposed EPR model provides more accurate results than the conventional model and the accuracy of the EPR results is better than or at least comparable to that of the neural network-based models proposed in the literature. The advantages of the proposed EPR model over the conventional and neural network-based models are highlighted.     

Support vector machines‐based modelling of seismic liquefaction potential

This paper investigates the potential of support vector machines (SVM)‐based classification approach to assess the liquefaction potential from actual standard penetration test (SPT) and cone penetration test (CPT) field data. SVMs are based on statistical learning theory and found to work well in comparison to neural networks in several other applications. Both CPT and SPT field data sets is used with SVMs for predicting the occurrence and non‐occurrence of liquefaction based on different input parameter combination. With SPT and CPT test data sets, highest accuracy of 96 and 97%, respectively, was achieved with SVMs. This suggests that SVMs can effectively be used to model the complex relationship between different soil parameter and the liquefaction potential. Several other combinations of input variable were used to assess the influence of different input parameters on liquefaction potential. Proposed approach suggest that neither normalized cone resistance value with CPT data nor the calculation of standardized SPT value is required with SPT data. Further, SVMs required few user‐defined parameters and provide better performance in comparison to neural network approach. Copyright © 2006 John Wiley & Sons, Ltd.     

Application of the adaptive neuro-fuzzy inference system for prediction of soil liquefaction

The determination of liquefaction potential of soils induced by earthquake is a major concern and an essential criterion in the design process of the civil engineering structures. A purely empirical interpretation of the filed case histories relating to liquefaction potential is often not well constrained due to the complication associated with this problem. In this study, an integrated fuzzy neural network model, called Adaptive Neuro-Fuzzy Inference System (ANFIS), is developed for the assessment of liquefaction potential. The model is trained with large databases of liquefaction case histories. Nine parameters such as earthquake magnitude, the water table, the total vertical stress, the effective vertical stress, the depth, the peak acceleration at the ground surface, the cyclic stress ratio, the mean grain size, and the measured cone penetration test tip resistance were used as input parameters. The results revealed that the ANFIS model is a fairly promising approach for the prediction of the soil liquefaction potential and capable of representing the complex relationship between seismic properties of soils and their liquefaction potential.     

1989～2011期间8次强地震中抗液化地基处理成功案例的回顾与启示

回顾了1989年美国Mw6.9级Loma Prieta地震、1993年日本Ms7.8级Kushiro-Oki地震、1994年日本Mw8.2级Hokkaido Toho-Oki地震、1995年日本Ms7.2级阪神地震、1999年台湾集集地震、1999年土耳其Mw7.4级Kocaeli地震、2001年美国Mw6.8级Nisqually地震以及2011年Mw9.0级东日本地震中场地抗液化工程措施的成功案例,初步分析了各种抗液化工程措施的有效性与优劣性,可以给出如下工程场地抗液化处理的经验：（1）对于易液化的沿海及填海造陆场地,采用适宜的抗液化工程措施应成为地基处理不可缺少的环节;（2）应基于场地条件、经济条件及环境要求,综合考虑场地抗液化地基处理措施的选择;（3）挤密砂桩法和碎石桩法运用广泛、技术成熟且比较经济,宜优先选择作为抗震设防烈度Ⅷ度及以下地区的场地抗液化地基处理措施;（4）强夯法使用机具简单、费用低廉,适宜选择作为抗震设防烈度Ⅷ度及以下地区大面积场地的抗液化地基处理措施;（5）注浆法、深层搅拌法、旋喷法作为抗震设防烈度Ⅸ度及以下地区的场地抗液化地基处理措施是有效的;（6）多种抗液化地基处理措施联合使用的处理效果往往优于单一措施单独使用的处理效果,在条件许可的情况下,宜选择多种抗液化地基处理措施联合使用,以期达到更好的处理效果。     

Role of the confined aquifer in the mechanism of soil liquefaction due to the 7.5 Mw earthquake in Palu (Indonesia) on 28 September 2018

Soil liquefaction on 28 September 2018 in Palu, Indonesia, included one of the largest soil movements ever, where objects on the ground surface moved hundreds of meters away and settlements sank into the mud. Some preliminary studies show that in addition to a strong earthquake, there are strong indications that a confined aquifer in the Palu valley worsened the liquefaction. The role of the confined aquifer can be recognized early on from one of various signs, namely the presence of massive surface inundations suspected due to groundwater expulsion which is thought to originate mostly from the confined aquifer. This paper describes the mechanism of the soil liquefaction in Palu from the perspective of earthquake hydrogeology, focusing on the groundwater expelled from an unconfined aquifer and especially from the underlying confined aquifer through hydraulic inter-connection between the two, which is possible due to simultaneous interaction of excess pore pressure dissipation and enhanced permeability driven by an earthquake in the near field. If this hypothesis proves to be strong, there are implications for engineering practices because the evaluation of potential soil liquefaction carried out currently in the geotechnical engineering field generally only involves the role of shallow groundwater and/or the unconfined aquifer and the role of soil layers not deeper than 30 m from the ground surface. It may be necessary to complement current evaluation practice with an evaluation of the deep groundwater response to earthquakes, especially if the deep groundwater is artesian and productive, with a relatively thin confining layer.

     

Test Investigation on Liquefied Deformation Structure in Saturated Lime-Mud Composites Triggered by Strong Earthquakes

Three identical model boxes were made from transparent plexiglass and angle iron. Using the method of sinking water and according to the sedimentary rhythm of saturated calcium carbonate(lime-mud) intercalated with cohesive soil,calcites with particle sizes diameters of ≤ 5 μm,10–15 μm and 23–30 μm as well as cohesive soil were sunk alternatively in water of three boxes to build three test models,each of which has a specific size of calcite. Pore water pressure gauges were buried in lime-mud layers at different depths in each model,and connected with a computer system to collect pore water pressures. By means of soil tests,physical property parameters and plasticity indices(Ip) were obtained for various grain-sized saturated lime-muds. The lime-muds with Ip ranging from 6.3 to 8.5(lower than 10) are similar to liquid saturated silt in the physical nature,indicating that saturated silt can be liquefied once induced by a strong earthquake. One model cart was pushed quickly along the length direction of the model so that its rigid wheels collided violently with the stone stair,thus generating an artificial earthquake with seismic wave magnitude greater than VI degree. When unidirectional cyclic seismic load of horizontal compression-tension-shear was imposed on the soil layers in the model,enough great pore water pressure has been accumulated within pores of lime-mud,resulting in liquefaction of lime-mud layers. Meanwhile,micro-fractures formed in each soil layer provided channels for liquefaction dewatering,resulting in formation of macroscopic liquefaction deformation,such as liquefied lime-mud volcanoes,liquefied diapir structures,vein-like liquefied structures and liquefied curls,etc. Splendid liquefied lime-mud eruption lasted for two to three hours,which is similar to the sand volcano eruption induced by strong earthquake. However,under the same artificial seismic conditions,development of macroscopic liquefied structures in three experimental models varied in shape,depth and quantity,indicating that excess pore water pressure ratios at initial liquefaction stage and complete liquefaction varied with depth. With size increasing of calcite particle in lime-mud,liquefied depth and deformation extent increase accordingly. The simulation test verifies for the first time that strong earthquakes may cause violent liquefaction of saturated lime-mud composed of micron-size calcite particles,uncovering the puzzled issue whether seafloor lime-mud can be liquefied under strong earthquake. This study not only provides the latest simulation data for explaining the earthquake-induced liquefied deformations of saturated lime-mud and seismic sedimentary events,but also is of great significance for analysis of foundation stability in marine engineering built on the soft calcium carbonate layers in neritic environment.     

Geotechnical reconnaissance and liquefaction analyses of a liquefaction site with silty fine sand in Southern Taiwan

Geotechnical reconnaissance of a recurrent liquefaction site at a Quaternary alluvial deposit in southern Taiwan was conducted to establish a comprehensive case history for liquefaction on silty fine sand with high fines content. The liquefaction occurred at a silty fine sand layer with D₅₀ = 0.09 mm and fines content greater than 35% and was triggered by a M_w = 6.4 earthquake on March 4, 2010, which induced maximum horizontal acceleration up to 0.189 g at the site. In situ subsurface characterizations, including standard penetration test, cone penetration test, and shear wave velocity measurement, were performed as well as cyclic simple shear tests on undisturbed specimens retrieved by a modified hydraulic piston sampler. Comparisons of cyclic resistance ratios (CRRs) indicate that CPT sounding with standard penetration rate could overestimate the resistance ratio and drainage conditions during penetration should be considered for high fines content soil in the liquefaction analysis. Additionally, variations of CRRs from different in situ tests indicate that correlations among in situ tests and CRR could be soil specific and precautions should be taken when using these curves on silty fine sands.     

Simplified DMT-based methods for evaluating liquefaction resistance of soils

This paper presents simplified dilatometer test (DMT)-based methods for evaluation of liquefaction resistance of soils, which is expressed in terms of cyclic resistance ratio (CRR). Two DMT parameters, horizontal stress index (K_D) and dilatometer modulus (E_D), are used as an index for assessing liquefaction resistance of soils. Specifically, CRR–K_D and CRR–E_D boundary curves are established based on the existing boundary curves that have already been developed based on standard penetration test (SPT) and cone penetration test (CPT). One key element in the development of CRR–K_D and CRR–E_D boundary curves is the correlations between K_D (or E_D) and the blow count (N) in the SPT or cone tip resistance (q_c) from the CPT. In this study, these correlations are established through regression analysis of the test results of SPT, CPT, and DMT conducted side-by-side at each of five sites selected. The validity of the developed CRR–K_D and CRR–E_D curves for evaluating liquefaction resistance is examined with published liquefaction case histories. The results of the study show that the developed DMT-based models are quite promising as a tool for evaluating liquefaction resistance of soils.     

Effects of principal stress rotation on wave-induced soil response in a poro-elastoplastic sandy seabed

In this study, a constitutive model is developed in order to investigate wave–seabed interactions. This model takes into account the impact of principal stress rotation (PSR) and is based on the generalized plasticity theory, in which plastic strain generated by PSR is considered an additional item in the constitutive relationship of soil. The normalized loading direction and plastic flow direction were determined based on the stress tensor invariant. Comparisons between the present model and previous Hollow Cylinder Apparatus tests and geotechnical centrifugal wave tests show good agreement. Numerical results show the effects of PSR on predictions of liquefaction potential due to: (a) the cumulative impact of plastic strain in the seafloor and (b) the buildup of pore pressure. Parametric study shows that the model parameters, including the wave and seabed parameters, have significant effects on the wave-induced soil liquefaction.

     

Nonlinear Site Response and Liquefaction Analysis in the New Madrid Seismic Zone

Many existing highway bridges in the New Madrid Seismic Zone are located in the Mississippi Embayment, consisting of deep soil deposits and liquefaction susceptible near surface soils. It is important to understand the comprehensive impact of deep soil deposits and liquefaction on the response of the bridge foundations under seismic loading. A nonlinear soil model is then presented to study the impacts of the deep soil deposit and liquefaction on response analysis. The soil model has the advantage of using input parameters that can be obtained from conventional field and laboratory testing methods, which makes it attractive to engineering practice. The model calibrations used field recorded motions and laboratory test data, which indicate that the model provides an acceptable outcome based on simple input parameters. The model is implemented into the site response analysis for a typical Missouri highway bridge site in this seismic zone. The effect of the deep soil deposit and liquefaction on the site response analyses is discussed.     

Explanation of liquefaction in after shock of the 2011 great east Japan earthquake using numerical analysis

During the 2011 Great East Japan Earthquake, severe liquefaction occurred in reclaimed ground in Urayasu city, Chiba prefecture. This liquefaction provided important lessons for us to re-recognize the liquefaction mechanism. A distinct feature of the liquefaction in this earthquake is that severe liquefaction happened not only in the main shock but also in an aftershock with a maximum acceleration of 25 gal. In some areas, liquefaction happened in the aftershock is even more serious than that happened in the main shock. In this paper, focus is placed on the characteristic features in the occurrence of liquefaction and consequent ground settlement. Based on the observed data, a series of dynamic–static analyses, considering not only the earthquake loading but also static loading during the consolidation after the earthquake shocks, are conducted in a sequential way just the same as the scenario in the earthquake. The calculation is conducted with 3D soil–water coupling finite element–finite difference analyses based on a cyclic elasto-plastic constitutive model. From the results of analyses, it is recognized that small sequential earthquakes, which cannot cause liquefaction of a ground in an independent earthquake vibration, cannot be neglected when the ground has already experienced liquefaction after a major vibration. In addition, the aftershock has great influence on the long-term settlement of low permeability soil layer. The observed and predicted liquefaction and settlements are compared and discussed carefully. It is confirmed that the numerical method used in this study can describe the ground behavior correctly under repeated earthquake shocks.     

砂土地震液化预测的GA_SVM_Adaboost模型

为快速准确地对砂土液化情况作出预测,选取地震烈度、地下水位、覆盖厚度、标贯击数、平均粒径、地貌单元、土质及不均匀系数为主要影响因素,运用相关性分析和因子分析模型对其进行分析和属性约减,采用遗传算法（GA）对支持向量机（SVM）的参数寻优,结合Adaboost迭代算法,建立预测砂土地震液化的GA_SVM_Adaboost模型。选用唐山地震砂土液化现场勘察资料中的329组数据对模型进行训练,利用该模型对剩余68组砂土液化数据进行预测。最后,将预测结果与GA_SVM和SVM模型预测结果进行比较。结果表明,3个模型的平均预测准确率分别为100%、98.04%、89.71%,基于因子分析的GA_SVM_Adaboost模型的预测准确性优于GA_SVM模型和SVM模型,是一种解决砂土地震液化预测问题的有效方法,具有一定的应用参考价值。      

Entropy analysis for identifying significant parameters for seismic soil liquefaction

A novel application of multi-criteria decision making (MCDM) technique to seismic soil liquefaction, a complex problem in earthquake geotechnical engineering, is presented. Seismic soil liquefaction depends on a diversified set of physical parameters with highly non-linear interconnections. Factors governing liquefaction may broadly be grouped as seismic parameters, site conditions and primarily dynamic soil properties, as the stimulus itself is manifestly dynamic. Each of these factors incorporates a wide range of variety of parameters that characterize liquefaction, to a varying degree of significance, such as: the magnitude, effective overburden pressure, shear modulus, normalized standard penetration blow count [N₁]₆₀, etc. Estimating rapid, yet accurate and reliable liquefaction susceptibility requires identification of the most significant factors controlling liquefaction. Thus a new concept of extracting significant parameters and gauging their importance is carried out by assigning them weights by applying MCDM introduced herein, whose evaluation is accomplished by means of an ‘entropy method’. In line with this, a relative reliability risk index (R³I) is computed indicating the ranking that directly reflects the severity of risk for liquefaction. Although the entropy analysis is carried out separately for the three multivariate criteria, it is remarkable that the R³I evaluated for each of these gives consistent ranking.     

Multilinear Regression Analysis for Seismic Response and Engineering Properties of Liquefiable Coromandel Coastal Soil Deposits

This paper investigates the seismic characteristics and geotechnical properties with respect to the liquefaction potential of the deposits in the Coromandel coastal line of Nagapattinam town, Tamilnadu, India. A series of field tests were conducted using standard penetration test, cone penetration test and plate load test. Laboratory tests were conducted on the collected samples. From the results, a microzonation map was developed for the liquefaction potential and settlements. Some multilinear regression models between permeability, fines content, relative density, coefficient of curvature, coefficient of uniformity, mean particle size, factor of safety, settlement, standard penetration test values and cone penetration test values were developed. The shear wave velocity and shear modulus were calculated from the field penetration tests and correlations between the normalized values of peak ground accelerations, velocities and displacements, which were obtained from the equivalent linear ground motion analysis using SHAKE software, with other parameters of soil. From the results, it was found that at some of the areas are vulnerable to high amplification of waves even for small earthquake.     

Evaluation of generalized linear models for soil liquefaction probability prediction

A soil deposit subjected to seismic loading can be viewed as a binary system: it will either liquefy or not liquefy. Generalized linear models are versatile tools for predicting the response of a binary system and hence potentially applicable to liquefaction prediction. In this study, the applicability of four generalized linear models (i.e., logistic, probit, log–log, and c-log–log) for liquefaction potential evaluation is assessed and compared. Eight liquefaction models based on the four generalized linear models and two sets of explanatory variables are evaluated. These models are first calibrated with past liquefaction performance data. A weighted-likelihood function method is used to consider the sampling bias in the calibration database. The predicted liquefaction probabilities from various models are then compared. When liquefaction probability is small, the predicted liquefaction probability is sensitive to the regression models used. The effect of sampling bias is more marked in the high cyclic stress ratio region. The eight models are finally ranked using a Bayesian model comparison method. For the generalized linear models examined, the logistic and c-log–log regression models are most supported by the past performance data. On the other hand, the probit and c-log–log regression models are much less applicable to liquefaction prediction.