Soil Dynamics and Earthquake Engineering最新文献_第7页

Convolutional neural network-based seismic response prediction method using spectral acceleration of earthquakes and conditional vector of structural property 基于卷积神经网络的地震响应预测方法（利用地震频谱加速度和结构属性条件向量

IF 4.2 2区工程技术 Q1 ENGINEERING, GEOLOGICAL

Soil Dynamics and Earthquake Engineering

Pub Date : 2024-10-11 DOI: 10.1016/j.soildyn.2024.109021

Insub Choi , Han Yong Lee , Byung Kwan Oh

This study proposes a method for predicting the seismic response of structures using seismic information and structural properties. In the proposed method, the relationship between seismic and structural characteristics and seismic responses was investigated using a convolutional neural network (CNN) to predict the seismic response. Spectral acceleration (Sa) calculated from the seismic wave was selected as seismic information used in CNN-based seismic response prediction techniques. The study introduced seismic information and structural properties, which correspond to the parameters to express the structure's unique characteristics or nonlinear hysteretic behaviors that determine the response characteristics of the structure subjected to seismic waves. Meanwhile, Sa and structural properties were utilized to constitute the input map of CNN and predict the maximum inter-story drift ratio that corresponds to the output of CNN. As data corresponding to the period range of interest rather than a scalar value for a specific period, Sa is rearranged in matrix form to constitute the input map of CNN. Structural properties are also placed in the input map of CNN as scalar values are converted into conditional vectors. To confirm the validity of the proposed method, multiple CNN-based models with changes in the information of the input map are presented, and their prediction performances are compared. Furthermore, CNN-based models that additionally consider seismic intensity measures are presented, and their influences on seismic response prediction performance are analyzed. In addition, a vast number of linear and nonlinear structures were generated, and seismic responses extracted via seismic analysis of multiple earthquakes were used to create datasets for training the presented models. The prediction performance of the presented models trained using the datasets was compared. The validity of the simultaneous use of structural properties with Sa, the introduction of conditional vectors, the additional use of seismic intensity measures, and their contributions to improving prediction performance were also examined.

本研究提出了一种利用地震信息和结构特性预测结构地震响应的方法。在该方法中，利用卷积神经网络（CNN）研究了地震和结构特性与地震反应之间的关系，从而预测地震反应。基于卷积神经网络的地震反应预测技术选择地震波计算出的谱加速度（Sa）作为地震信息。研究引入了地震信息和结构属性，它们对应于表达结构独特特性或非线性滞回行为的参数，这些参数决定了结构在地震波作用下的响应特性。同时，利用 Sa 和结构特性构成 CNN 的输入图，并预测 CNN 输出所对应的最大层间漂移比。作为与所关注的周期范围相对应的数据，而不是特定周期的标量值，Sa 以矩阵形式重新排列，构成 CNN 的输入图。由于标量值被转换为条件向量，结构属性也被置于 CNN 的输入图中。为了证实所提方法的有效性，我们介绍了输入图信息发生变化的多个基于 CNN 的模型，并比较了它们的预测性能。此外，还介绍了额外考虑地震烈度度量的基于 CNN 的模型，并分析了它们对地震响应预测性能的影响。此外，还生成了大量线性和非线性结构，并使用通过多次地震的地震分析提取的地震响应来创建数据集，用于训练所提出的模型。比较了使用数据集训练的模型的预测性能。此外，还考察了同时使用结构属性和 Sa、引入条件向量、额外使用地震烈度测量的有效性，以及它们对提高预测性能的贡献。

{"title":"Convolutional neural network-based seismic response prediction method using spectral acceleration of earthquakes and conditional vector of structural property","authors":"Insub Choi , Han Yong Lee , Byung Kwan Oh","doi":"10.1016/j.soildyn.2024.109021","DOIUrl":"10.1016/j.soildyn.2024.109021","url":null,"abstract":"<div><div>This study proposes a method for predicting the seismic response of structures using seismic information and structural properties. In the proposed method, the relationship between seismic and structural characteristics and seismic responses was investigated using a convolutional neural network (CNN) to predict the seismic response. Spectral acceleration (Sa) calculated from the seismic wave was selected as seismic information used in CNN-based seismic response prediction techniques. The study introduced seismic information and structural properties, which correspond to the parameters to express the structure's unique characteristics or nonlinear hysteretic behaviors that determine the response characteristics of the structure subjected to seismic waves. Meanwhile, Sa and structural properties were utilized to constitute the input map of CNN and predict the maximum inter-story drift ratio that corresponds to the output of CNN. As data corresponding to the period range of interest rather than a scalar value for a specific period, Sa is rearranged in matrix form to constitute the input map of CNN. Structural properties are also placed in the input map of CNN as scalar values are converted into conditional vectors. To confirm the validity of the proposed method, multiple CNN-based models with changes in the information of the input map are presented, and their prediction performances are compared. Furthermore, CNN-based models that additionally consider seismic intensity measures are presented, and their influences on seismic response prediction performance are analyzed. In addition, a vast number of linear and nonlinear structures were generated, and seismic responses extracted via seismic analysis of multiple earthquakes were used to create datasets for training the presented models. The prediction performance of the presented models trained using the datasets was compared. The validity of the simultaneous use of structural properties with Sa, the introduction of conditional vectors, the additional use of seismic intensity measures, and their contributions to improving prediction performance were also examined.</div></div>","PeriodicalId":49502,"journal":{"name":"Soil Dynamics and Earthquake Engineering","volume":"187 ","pages":"Article 109021"},"PeriodicalIF":4.2,"publicationDate":"2024-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142419971","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

AI-driven predictions of the dynamic properties of fine-grained soils in South Italy based on laboratory testing 根据实验室测试对南意大利细粒土的动态特性进行人工智能预测

IF 4.2 2区工程技术 Q1 ENGINEERING, GEOLOGICAL

Soil Dynamics and Earthquake Engineering

Pub Date : 2024-10-10 DOI: 10.1016/j.soildyn.2024.109009

Michele Placido Antonio Gatto , Francesco Castelli , Valentina Lentini , Lorella Montrasio

This study presents the use of Artificial Intelligence (AI) to predict the dynamic behaviour of fine-grained soils of South Italy based on a detailed laboratory investigation. The investigation consists of Resonant Column (RC), Cyclic Torsional Shear (CTS), and Cyclic Triaxial (CTx) tests performed on 25 specimens of fine-grained soils retrieved from 11 sites in Sicily (South Italy). To develop accurate predictive models of soil dynamic properties, essential for site response analyses and dynamic soil-structure interaction, various regression techniques were applied. These techniques range from Multiple Linear Regression (MLR) to more complex AI methods, specifically Machine Learning (ML) and Deep Learning (DL) based on FeedForward Neural networks (FFN). Three predictive models were developed to derive strain-dependent shear modulus (G), damping ratio (D), and normalized shear modulus (G/G₀), using four inputs: shear strain (γ), plasticity index (PI), confining pressure (p’₀), and the Over Consolidation Ratio (OCR). To determine the optimal FFN topology, 1350 networks were developed by varying hidden layers (1–3), hidden neurons (1–50 per layer), and activation functions (ReLU, sigmoid and hyperbolic tangent). Hybrid FFN optimised through Genetic Algorithm and Particle Swarm Optimization techniques were also investigated. Single-hidden layer networks with fewer than 15 neurons provided acceptable predictions (R²_test of 0.97 for G-γ, 0.93 for G/G₀-γ, and 0.85 for D-γ models). Multiple-hidden layer networks yielded higher accuracy for G and D models but are more complex for practical use. The FFN models outperformed MLR and other established empirical formulations, highlighting the site-specificity of the modelling parameters of the latter.

本研究以详细的实验室调查为基础，介绍了使用人工智能（AI）预测南意大利细粒土动态行为的方法。调查包括共振柱 (RC)、循环扭转剪切 (CTS) 和循环三轴 (CTx) 测试，测试对象是从西西里岛（南意大利）的 11 个地点取回的 25 个细粒土试样。为了建立对场地响应分析和土壤与结构动态相互作用至关重要的土壤动态特性的精确预测模型，采用了各种回归技术。这些技术包括从多元线性回归（MLR）到更复杂的人工智能方法，特别是基于前馈神经网络（FFN）的机器学习（ML）和深度学习（DL）。我们开发了三种预测模型，利用剪切应变 (γ)、塑性指数 (PI)、约束压力 (p'0) 和过固结比 (OCR) 这四项输入，得出随应变变化的剪切模量 (G)、阻尼比 (D) 和归一化剪切模量 (G/G0)。为确定最佳 FFN 拓扑，通过改变隐藏层（1-3 层）、隐藏神经元（每层 1-50 个）和激活函数（ReLU、sigmoid 和双曲正切）开发了 1350 个网络。此外，还研究了通过遗传算法和粒子群优化技术优化的混合 FFN。少于 15 个神经元的单隐层网络提供了可接受的预测结果（G-γ 模型的 R2test 为 0.97，G/G0-γ 模型的 R2test 为 0.93，D-γ 模型的 R2test 为 0.85）。多隐藏层网络为 G 和 D 模型提供了更高的准确度，但在实际应用中更为复杂。FFN 模型的性能优于 MLR 和其他既定的经验公式，这突出表明后者的建模参数具有地点特异性。

{"title":"AI-driven predictions of the dynamic properties of fine-grained soils in South Italy based on laboratory testing","authors":"Michele Placido Antonio Gatto , Francesco Castelli , Valentina Lentini , Lorella Montrasio","doi":"10.1016/j.soildyn.2024.109009","DOIUrl":"10.1016/j.soildyn.2024.109009","url":null,"abstract":"<div><div>This study presents the use of Artificial Intelligence (AI) to predict the dynamic behaviour of fine-grained soils of South Italy based on a detailed laboratory investigation. The investigation consists of Resonant Column (RC), Cyclic Torsional Shear (CTS), and Cyclic Triaxial (CTx) tests performed on 25 specimens of fine-grained soils retrieved from 11 sites in Sicily (South Italy). To develop accurate predictive models of soil dynamic properties, essential for site response analyses and dynamic soil-structure interaction, various regression techniques were applied. These techniques range from Multiple Linear Regression (MLR) to more complex AI methods, specifically Machine Learning (ML) and Deep Learning (DL) based on FeedForward Neural networks (FFN). Three predictive models were developed to derive strain-dependent shear modulus (<em>G</em>), damping ratio (<em>D</em>), and normalized shear modulus (<em>G</em>/<em>G</em><sub><em>0</em></sub>), using four inputs: shear strain (<em>γ)</em>, plasticity index (<em>PI)</em>, confining pressure (<em>p’</em><sub><em>0</em></sub>), and the Over Consolidation Ratio (OCR). To determine the optimal FFN topology, 1350 networks were developed by varying hidden layers (1–3), hidden neurons (1–50 per layer), and activation functions (ReLU, sigmoid and hyperbolic tangent). Hybrid FFN optimised through Genetic Algorithm and Particle Swarm Optimization techniques were also investigated. Single-hidden layer networks with fewer than 15 neurons provided acceptable predictions (<em>R</em><sup><em>2</em></sup><sub><em>test</em></sub> of 0.97 for <em>G</em>-<em>γ</em>, 0.93 for <em>G</em>/<em>G</em><sub><em>0</em></sub>-<em>γ</em>, and 0.85 for <em>D</em>-<em>γ</em> models). Multiple-hidden layer networks yielded higher accuracy for <em>G</em> and <em>D</em> models but are more complex for practical use. The FFN models outperformed MLR and other established empirical formulations, highlighting the site-specificity of the modelling parameters of the latter.</div></div>","PeriodicalId":49502,"journal":{"name":"Soil Dynamics and Earthquake Engineering","volume":"187 ","pages":"Article 109009"},"PeriodicalIF":4.2,"publicationDate":"2024-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142420090","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

An optimization suggestion for site classification scheme in Chinese seismic code based on clustering analysis of site amplification 基于场地振幅聚类分析的中国地震规范场地分类方案优化建议

IF 4.2 2区工程技术 Q1 ENGINEERING, GEOLOGICAL

Soil Dynamics and Earthquake Engineering

Pub Date : 2024-10-10 DOI: 10.1016/j.soildyn.2024.109018

Ye Liu , Yefei Ren , Ruizhi Wen , Hongwei Wang , Kun Ji , Yajun Zhang , Yingxin Hui

There has been extensive discussion as to whether the scope of site classification II is too broad in current Chinese seismic code. To address this issue, this study aims to optimize the site classification scheme for Chinese seismic code using clustering analysis of site amplification. Firstly, we estimate the empirical site amplification factors of KiK-net stations by the residual analysis method, and classify them by the site classification scheme of Chinese seismic code. Next, we perform k-means clustering analysis on the stations of site class II, considering site amplification factors, equivalent shear wave velocities and thicknesses of sedimentary layers as explanatory variables, and obtain two clusters with distinct site amplification effects. Finally, we use correlation analysis and Receiver Operating Characteristic (ROC) curve to guide the optimization of site classification scheme, and suggest dividing site class II into two subclasses, II_a and II_b, by a threshold of 15m for the thickness of sedimentary layer. The proposed optimized classification scheme would be beneficial for improving the seismic design code and could be further applied to the development of ground motion models and seismic hazard analysis.

关于现行中国地震规范中场地分类 II 的范围是否过于宽泛的问题已引起广泛讨论。针对这一问题，本研究旨在通过对台站放大系数的聚类分析，优化中国地震规范的台站分类方案。首先，利用残差分析方法估算 KiK 网台站的经验台址放大系数，并按照中国地震台网规范的台址分类方案进行分类。其次，以场址放大系数、等效剪切波速和沉积层厚度为解释变量，对Ⅱ类场址台站进行 K-均值聚类分析，得到两个场址放大效应不同的聚类。最后，利用相关性分析和接收者工作特征曲线（ROC）来指导站点分类方案的优化，并建议以沉积层厚度 15 米为临界值，将站点Ⅱ类分为Ⅱa 和Ⅱb 两个子类。建议的优化分类方案将有利于改进抗震设计规范，并可进一步应用于地震动模型的开发和地震危险性分析。

{"title":"An optimization suggestion for site classification scheme in Chinese seismic code based on clustering analysis of site amplification","authors":"Ye Liu , Yefei Ren , Ruizhi Wen , Hongwei Wang , Kun Ji , Yajun Zhang , Yingxin Hui","doi":"10.1016/j.soildyn.2024.109018","DOIUrl":"10.1016/j.soildyn.2024.109018","url":null,"abstract":"<div><div>There has been extensive discussion as to whether the scope of site classification II is too broad in current Chinese seismic code. To address this issue, this study aims to optimize the site classification scheme for Chinese seismic code using clustering analysis of site amplification. Firstly, we estimate the empirical site amplification factors of KiK-net stations by the residual analysis method, and classify them by the site classification scheme of Chinese seismic code. Next, we perform <em>k</em>-means clustering analysis on the stations of site class II, considering site amplification factors, equivalent shear wave velocities and thicknesses of sedimentary layers as explanatory variables, and obtain two clusters with distinct site amplification effects. Finally, we use correlation analysis and Receiver Operating Characteristic (ROC) curve to guide the optimization of site classification scheme, and suggest dividing site class II into two subclasses, II<sub>a</sub> and II<sub>b</sub>, by a threshold of 15m for the thickness of sedimentary layer. The proposed optimized classification scheme would be beneficial for improving the seismic design code and could be further applied to the development of ground motion models and seismic hazard analysis.</div></div>","PeriodicalId":49502,"journal":{"name":"Soil Dynamics and Earthquake Engineering","volume":"187 ","pages":"Article 109018"},"PeriodicalIF":4.2,"publicationDate":"2024-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142420081","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Parameter-normalized probabilistic seismic demand model considering the structural design strength for structural response assessment 考虑结构设计强度的参数归一化概率地震需求模型，用于结构响应评估

IF 4.2 2区工程技术 Q1 ENGINEERING, GEOLOGICAL

Soil Dynamics and Earthquake Engineering

Pub Date : 2024-10-10 DOI: 10.1016/j.soildyn.2024.109023

Jian-Cheng Dai , Dong-Sheng Wang , Wei-Jian Tang , Yu-He Zou , Ying-Xin Hui , Ya-Jun Zhang

The Probabilistic Seismic Demand Model (PSDM) is a crucial component of the performance-based seismic design framework when establishing the relationship between the ground motion intensity measure (IM) and the engineering demand parameter (EDP). The definitions of IMs and EDPs introduce varying degrees of uncertainty into the PSDM and notes different fragility or hazard analysis results. In accordance with the elastic limit state of the structural seismic response, this study normalizes two key parameters, the IM and EDP, within the PSDM. Normalized EDP (EDP^N) is the ratio of the structural response to the elastic limit state of the structure, as defined by the onset of the strength yielding of the main structural element. Similarly, the IM (IM^N) is normalized based on corresponding ground motions (scaled) that cause the structure to offer an elastic limit state response. This means that structural design strength is considered in IM^N following the construction of a parameter-normalized PSDM. The study examined two typical isolated bridges presented their hazard curves with IM^N. The results show that IM^N can unify the efficiency and sufficiency of different IMs and reduce uncertainty in the PSDM. The assessment error of the structural elastic limit state for its design strength had little effect on the parameter-normalized PSDM, so the model is robust. Additionally, the IM^N outperformed traditional IMs for efficiency and sufficiency in most instances.

概率地震需求模型（PSDM）是基于性能的抗震设计框架的重要组成部分，用于建立地震动烈度（IM）和工程需求参数（EDP）之间的关系。IM 和 EDP 的定义为 PSDM 带来了不同程度的不确定性，并指出了不同的脆性或危险性分析结果。根据结构地震反应的弹性极限状态，本研究将 PSDM 中的两个关键参数 IM 和 EDP 归一化。归一化 EDP（EDPN）是结构响应与结构弹性极限状态的比值，由主要结构元件强度屈服开始时定义。同样，IM（IMN）也是根据导致结构产生弹性极限状态响应的相应地面运动（按比例）进行归一化的。这意味着在构建参数归一化 PSDM 后，IMN 将考虑结构设计强度。研究考察了两座典型的孤立桥梁，并用 IMN 展示了它们的危险曲线。结果表明，IMN 可以统一不同 IM 的效率和充分性，减少 PSDM 的不确定性。结构弹性极限状态设计强度的评估误差对参数归一化 PSDM 的影响很小，因此该模型是稳健的。此外，在大多数情况下，IMN 在效率和充分性方面都优于传统的 IM。

{"title":"Parameter-normalized probabilistic seismic demand model considering the structural design strength for structural response assessment","authors":"Jian-Cheng Dai , Dong-Sheng Wang , Wei-Jian Tang , Yu-He Zou , Ying-Xin Hui , Ya-Jun Zhang","doi":"10.1016/j.soildyn.2024.109023","DOIUrl":"10.1016/j.soildyn.2024.109023","url":null,"abstract":"<div><div>The Probabilistic Seismic Demand Model (PSDM) is a crucial component of the performance-based seismic design framework when establishing the relationship between the ground motion intensity measure (IM) and the engineering demand parameter (EDP). The definitions of IMs and EDPs introduce varying degrees of uncertainty into the PSDM and notes different fragility or hazard analysis results. In accordance with the elastic limit state of the structural seismic response, this study normalizes two key parameters, the IM and EDP, within the PSDM. Normalized EDP (<em>EDP</em><sup>N</sup>) is the ratio of the structural response to the elastic limit state of the structure, as defined by the onset of the strength yielding of the main structural element. Similarly, the IM (<em>IM</em><sup>N</sup>) is normalized based on corresponding ground motions (scaled) that cause the structure to offer an elastic limit state response. This means that structural design strength is considered in <em>IM</em><sup>N</sup> following the construction of a parameter-normalized PSDM. The study examined two typical isolated bridges presented their hazard curves with <em>IM</em><sup>N</sup>. The results show that <em>IM</em><sup>N</sup> can unify the efficiency and sufficiency of different IMs and reduce uncertainty in the PSDM. The assessment error of the structural elastic limit state for its design strength had little effect on the parameter-normalized PSDM, so the model is robust. Additionally, the <em>IM</em><sup>N</sup> outperformed traditional IMs for efficiency and sufficiency in most instances.</div></div>","PeriodicalId":49502,"journal":{"name":"Soil Dynamics and Earthquake Engineering","volume":"187 ","pages":"Article 109023"},"PeriodicalIF":4.2,"publicationDate":"2024-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142420089","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Seismic behavior of shallow buried water reservoirs via large scale three-dimensional numerical models 通过大尺度三维数值模型研究浅埋水层的地震行为

IF 4.2 2区工程技术 Q1 ENGINEERING, GEOLOGICAL

Soil Dynamics and Earthquake Engineering

Pub Date : 2024-10-10 DOI: 10.1016/j.soildyn.2024.109005

Karim AlKhatib , Youssef MA. Hashash , Katerina Ziotopoulou

As buried water reservoirs are increasingly being utilized to store and deliver water, they are now regarded as critical infrastructures that must continue to operate in the event of an earthquake. This paper presents the results of a large-scale numerical parametric study that was carried out to advance our understanding of the seismic fluid-structure-soil interaction (FSSI) response of buried water reservoirs. Advanced nonlinear three-dimensional (3D) FSSI numerical models of reservoirs were employed while considering reservoir size, embedment depth, soil profile, and ground motion variability. The study showed that, unlike other conventional underground structures, the peak ground acceleration (PGA) has the strongest correlation to the reservoir seismic response. Increasing the embedment depth or reservoir size was found to generally increase the demands on the structural elements while reducing the base and backfill slippage. Softer sites were found to cause an increase in the roof racking and including the vertical component of the motion increased the water dynamic pressures. Among the columns, the ones closest to the center were found to experience the highest demands and the ones at the corner the lowest. In fact, in some extreme cases, a total collapse of the reservoir was initiated by column failure due to the lack of structural redundancy. The roof in-plane shear stresses were observed to accumulate near the walls, indicating a diaphragm behavior. The reservoir's unique seismic response compared to other underground structures makes generalizing the commonly used simplified design procedures inapplicable. Instead, 3D FSSI numerical models were demonstrated to be a reliable tool for the seismic design of buried reservoirs.

随着地下水库越来越多地被用于储水和输水，它们现在被视为地震时必须继续运行的关键基础设施。本文介绍了一项大规模数值参数研究的结果，该研究旨在加深我们对地下水库地震流体-结构-土壤相互作用（FSSI）响应的理解。研究采用了先进的非线性三维 FSSI 数值模型，同时考虑了水库规模、埋深、土壤剖面和地动变化。研究表明，与其他常规地下结构不同，地面峰值加速度（PGA）与水库地震响应的相关性最强。研究发现，增加埋设深度或水库规模通常会增加对结构元件的要求，同时减少基底和回填土的滑动。研究发现，较松软的场地会导致顶架增加，包括运动的垂直分量也会增加水动压力。在柱子中，最靠近中心的柱子承受的压力最大，而位于角落的柱子承受的压力最小。事实上，在一些极端情况下，由于缺乏结构冗余，蓄水池的完全坍塌是由支柱失效引起的。观察到屋顶平面剪应力在墙壁附近累积，表明存在隔墙行为。与其他地下结构相比，该水库的地震响应非常独特，因此无法采用常用的简化设计程序。相反，三维 FSSI 数值模型被证明是地下水库抗震设计的可靠工具。

{"title":"Seismic behavior of shallow buried water reservoirs via large scale three-dimensional numerical models","authors":"Karim AlKhatib , Youssef MA. Hashash , Katerina Ziotopoulou","doi":"10.1016/j.soildyn.2024.109005","DOIUrl":"10.1016/j.soildyn.2024.109005","url":null,"abstract":"<div><div>As buried water reservoirs are increasingly being utilized to store and deliver water, they are now regarded as critical infrastructures that must continue to operate in the event of an earthquake. This paper presents the results of a large-scale numerical parametric study that was carried out to advance our understanding of the seismic fluid-structure-soil interaction (FSSI) response of buried water reservoirs. Advanced nonlinear three-dimensional (3D) FSSI numerical models of reservoirs were employed while considering reservoir size, embedment depth, soil profile, and ground motion variability. The study showed that, unlike other conventional underground structures, the peak ground acceleration (PGA) has the strongest correlation to the reservoir seismic response. Increasing the embedment depth or reservoir size was found to generally increase the demands on the structural elements while reducing the base and backfill slippage. Softer sites were found to cause an increase in the roof racking and including the vertical component of the motion increased the water dynamic pressures. Among the columns, the ones closest to the center were found to experience the highest demands and the ones at the corner the lowest. In fact, in some extreme cases, a total collapse of the reservoir was initiated by column failure due to the lack of structural redundancy. The roof in-plane shear stresses were observed to accumulate near the walls, indicating a diaphragm behavior. The reservoir's unique seismic response compared to other underground structures makes generalizing the commonly used simplified design procedures inapplicable. Instead, 3D FSSI numerical models were demonstrated to be a reliable tool for the seismic design of buried reservoirs.</div></div>","PeriodicalId":49502,"journal":{"name":"Soil Dynamics and Earthquake Engineering","volume":"187 ","pages":"Article 109005"},"PeriodicalIF":4.2,"publicationDate":"2024-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142420088","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Axial kinematic response of an end-bearing pile subjected to seismic P-wave excitation in a double-layered soil 双层土中受地震 P 波激励的端承桩的轴向运动学响应

IF 4.2 2区工程技术 Q1 ENGINEERING, GEOLOGICAL

Soil Dynamics and Earthquake Engineering

Pub Date : 2024-10-10 DOI: 10.1016/j.soildyn.2024.109012

Denghui Dai , Kai Xue , Yufeng Gao , M. Hesham El Naggar , Xinyu Du , Ning Zhang

This paper presents an analytical solution for the axial kinematic response of an end-bearing pile under seismic P-wave excitation in a double-layered soil. The motions of the pile and adjacent soil layers are determined using the derived series solution. The seismic wave scattering effect caused by the combined action of the soil layers and pile foundations is accounted for. To validate the accuracy of the derived solution, the kinematic responses obtained from the proposed analytical solution are compared with the results from an existing solution for a pile in a single soil layer. The validated solution is then utilized to conduct a parametric study to investigate the impacts of the modulus and thickness ratios between the two soil layers on response characteristics. These characteristics include the kinematic response factor, amplification factor, frictional force exerted on the pile body, pile displacement with depth, and motion of soil around the pile.

本文给出了双层土中端承桩在地震 P 波激励下的轴向运动响应的分析解。利用推导出的序列解确定了桩和相邻土层的运动。其中考虑了土层和桩基的共同作用引起的地震波散射效应。为了验证推导解决方案的准确性，我们将根据建议的分析解决方案得到的运动响应与现有的单土层桩基解决方案的结果进行了比较。然后，利用验证后的解决方案进行参数研究，探讨两个土层之间的模量和厚度比对响应特性的影响。这些特征包括运动响应系数、放大系数、施加在桩身上的摩擦力、桩随深度的位移以及桩周围土壤的运动。

引用次数: 0

Seismic resilience evaluation of confined masonry school buildings retrofitted by shotcrete method 采用喷射混凝土法改造的受限砌体学校建筑的抗震性评估

IF 4.2 2区工程技术 Q1 ENGINEERING, GEOLOGICAL

Soil Dynamics and Earthquake Engineering

Pub Date : 2024-10-09 DOI: 10.1016/j.soildyn.2024.108980

Ali Sadeghi Raveshti , Morteza Raissi Dehkordi , Mahdi Eghbali , Delbaz Samadian

This study addresses the urgent need for retrofitting masonry schools in Iran, where over 89 % of schools are constructed using masonry (107000 Schools consisted of 563000 classrooms). Focusing on seismic performance evaluation, this research is of paramount importance as the current state of these schools poses a significant risk to the safety of millions of students. Without proper evaluation, the consequences could be catastrophic. To illustrate the significance of this study, a confined masonry school in Tehran was selected as a representative case study. The seismic resilience index was calculated, considering two hazard levels (2 % and 10 %) over a 50-year period, with and without near-fault pulse-type ground motions. Furthermore, a comparative analysis was conducted by modeling the retrofitted structure, which involved the application of shotcrete on walls, to assess the benefits of retrofitting. The seismic resilience index, derived from analytical functions that account for probable hazards and the recovery process, serves as a comprehensive measure of the structure's vulnerability. Through this evaluation, valuable insights into the retrofitting and rehabilitation of confined masonry schools can be gained. The results obtained from this research will aid in informed decision-making regarding the mitigation of seismic risks in similar structures. In conclusion, this paper indicates applying shotcrete to masonry walls can be substantially efficient as the resilience index and functionality of the building increases by a significant margin. Furthermore, the building damage was halved after the retrofitting operation and fragility curves admit it by showing better performance in higher drift ratios.

伊朗 89% 以上的学校（10.7 万所学校，56.3 万间教室）都是用砌体建造的，本研究解决了伊朗砌体学校改造的迫切需求。这项研究以抗震性能评估为重点，具有极其重要的意义，因为这些学校的现状对数百万学生的安全构成了重大威胁。如果不进行适当的评估，后果可能是灾难性的。为了说明这项研究的重要性，我们选择了德黑兰的一所封闭式砌体学校作为代表性案例。计算了抗震指数，考虑了 50 年期间的两个危险等级（2 % 和 10 %），以及近断层脉冲型地动和非近断层脉冲型地动。此外，还对改造后的结构进行了建模比较分析，包括在墙体上喷射混凝土，以评估改造的益处。抗震指数由考虑到可能发生的灾害和恢复过程的分析函数得出，可作为结构脆弱性的综合衡量标准。通过这一评估，可以对封闭式砌体学校的改造和修复获得有价值的见解。这项研究的结果将有助于在减轻类似结构的地震风险方面做出明智的决策。总之，本文表明，在砌体墙体上喷射混凝土可以大大提高效率，因为建筑物的回弹指数和功能性都有显著提高。此外，改造后建筑物的损坏程度减半，脆性曲线也证明了这一点，在较高的漂移率下表现更佳。

{"title":"Seismic resilience evaluation of confined masonry school buildings retrofitted by shotcrete method","authors":"Ali Sadeghi Raveshti , Morteza Raissi Dehkordi , Mahdi Eghbali , Delbaz Samadian","doi":"10.1016/j.soildyn.2024.108980","DOIUrl":"10.1016/j.soildyn.2024.108980","url":null,"abstract":"<div><div>This study addresses the urgent need for retrofitting masonry schools in Iran, where over 89 % of schools are constructed using masonry (107000 Schools consisted of 563000 classrooms). Focusing on seismic performance evaluation, this research is of paramount importance as the current state of these schools poses a significant risk to the safety of millions of students. Without proper evaluation, the consequences could be catastrophic. To illustrate the significance of this study, a confined masonry school in Tehran was selected as a representative case study. The seismic resilience index was calculated, considering two hazard levels (2 % and 10 %) over a 50-year period, with and without near-fault pulse-type ground motions. Furthermore, a comparative analysis was conducted by modeling the retrofitted structure, which involved the application of shotcrete on walls, to assess the benefits of retrofitting. The seismic resilience index, derived from analytical functions that account for probable hazards and the recovery process, serves as a comprehensive measure of the structure's vulnerability. Through this evaluation, valuable insights into the retrofitting and rehabilitation of confined masonry schools can be gained. The results obtained from this research will aid in informed decision-making regarding the mitigation of seismic risks in similar structures. In conclusion, this paper indicates applying shotcrete to masonry walls can be substantially efficient as the resilience index and functionality of the building increases by a significant margin. Furthermore, the building damage was halved after the retrofitting operation and fragility curves admit it by showing better performance in higher drift ratios.</div></div>","PeriodicalId":49502,"journal":{"name":"Soil Dynamics and Earthquake Engineering","volume":"187 ","pages":"Article 108980"},"PeriodicalIF":4.2,"publicationDate":"2024-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142420079","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Rapid and automated seismic design of cable restrainer for simply supported bridges crossing fault rupture zones using explainable machine learning 利用可解释的机器学习，为跨越断层破裂带的简支桥梁快速自动设计缆索约束装置

IF 4.2 2区工程技术 Q1 ENGINEERING, GEOLOGICAL

Soil Dynamics and Earthquake Engineering

Pub Date : 2024-10-09 DOI: 10.1016/j.soildyn.2024.109011

Fan Zhang , Yuguang Fu , Jingquan Wang

Earthquakes in recent decades have demonstrated that fault-crossing simply supported bridges were susceptible to damage caused by the fault-induced permanent ground dislocation. Cable restrainer can potentially reduce the relative displacement of bridge spans, but the current seismic design method for restrainer is time-consuming and labor-intensive. This study aims to develop a rapid and automated seismic design method for cable restrainer using explainable machine learning (ML) models. To do this, a large database was first generated based on the current design approach. ML algorithms were utilized to develop classification models to determine the design classes and then regression models to estimate the restrainer stiffness for the fault-crossing bridges. Furthermore, SHapley Additive exPlanations (SHAP) analysis was utilized to provide interpretations for the best regression model. In particular, an empirical formula and two explainable prediction models by combining the empirical formula with simplified ML models were finally proposed to facilitate the design for engineers. Results show that the proposed design method can provide accurate and robust results of bridge restrainers. Within the method, artificial neural network was selected among nine ML models, because of its highest accuracy for both classification and regression. The SHAP analysis reveals that, the allowable displacement has a negative nonlinear effect, while permanent ground dislocation and initial relative displacement present positive nonlinear effects. The proposed empirical formula for restrainer design can provide conservative estimations with an accuracy of 79 %, whereas the proposed explainable prediction models have a high accuracy of 94 % and are significantly efficient and user-friendly.

近几十年来发生的地震表明，跨越断层的简支桥梁很容易受到断层引起的永久地表位移的破坏。拉索约束装置有可能减少桥跨的相对位移，但目前的约束装置抗震设计方法耗时耗力。本研究旨在利用可解释的机器学习（ML）模型，开发一种快速、自动化的缆索约束抗震设计方法。为此，首先根据当前的设计方法生成了一个大型数据库。利用 ML 算法开发分类模型来确定设计类别，然后利用回归模型来估算跨断层桥梁的约束刚度。此外，还利用 SHapley Additive exPlanations（SHAP）分析法为最佳回归模型提供解释。特别是，通过将经验公式与简化的 ML 模型相结合，最终提出了一个经验公式和两个可解释的预测模型，以方便工程师进行设计。结果表明，所提出的设计方法可以为桥梁约束装置提供准确、稳健的结果。在该方法中，人工神经网络在分类和回归方面都具有最高的准确性，因此在九个 ML 模型中被选中。SHAP 分析表明，容许位移具有负非线性效应，而永久地面错位和初始相对位移则具有正非线性效应。所提出的约束装置设计经验公式可以提供保守的估计，准确率为 79%，而所提出的可解释预测模型的准确率高达 94%，并且非常高效和易于使用。

{"title":"Rapid and automated seismic design of cable restrainer for simply supported bridges crossing fault rupture zones using explainable machine learning","authors":"Fan Zhang , Yuguang Fu , Jingquan Wang","doi":"10.1016/j.soildyn.2024.109011","DOIUrl":"10.1016/j.soildyn.2024.109011","url":null,"abstract":"<div><div>Earthquakes in recent decades have demonstrated that fault-crossing simply supported bridges were susceptible to damage caused by the fault-induced permanent ground dislocation. Cable restrainer can potentially reduce the relative displacement of bridge spans, but the current seismic design method for restrainer is time-consuming and labor-intensive. This study aims to develop a rapid and automated seismic design method for cable restrainer using explainable machine learning (ML) models. To do this, a large database was first generated based on the current design approach. ML algorithms were utilized to develop classification models to determine the design classes and then regression models to estimate the restrainer stiffness for the fault-crossing bridges. Furthermore, SHapley Additive exPlanations (SHAP) analysis was utilized to provide interpretations for the best regression model. In particular, an empirical formula and two explainable prediction models by combining the empirical formula with simplified ML models were finally proposed to facilitate the design for engineers. Results show that the proposed design method can provide accurate and robust results of bridge restrainers. Within the method, artificial neural network was selected among nine ML models, because of its highest accuracy for both classification and regression. The SHAP analysis reveals that, the allowable displacement has a negative nonlinear effect, while permanent ground dislocation and initial relative displacement present positive nonlinear effects. The proposed empirical formula for restrainer design can provide conservative estimations with an accuracy of 79 %, whereas the proposed explainable prediction models have a high accuracy of 94 % and are significantly efficient and user-friendly.</div></div>","PeriodicalId":49502,"journal":{"name":"Soil Dynamics and Earthquake Engineering","volume":"187 ","pages":"Article 109011"},"PeriodicalIF":4.2,"publicationDate":"2024-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142420082","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Seismic failure analysis of a high arch dam-foundation multiple nonlinear coupling system 高拱坝-地基多重非线性耦合系统的地震破坏分析

IF 4.2 2区工程技术 Q1 ENGINEERING, GEOLOGICAL

Soil Dynamics and Earthquake Engineering

Pub Date : 2024-10-07 DOI: 10.1016/j.soildyn.2024.109001

Chunli Yan , Jin Tu , Hui Liang , Shengshan Guo , Deyu Li

In this study, a high arch dam-foundation system model with more than ten million degrees of freedom was constructed. The model innovatively incorporates multiple nonlinear couplings of the strength failure of the dam body and stability failure of dam abutment blocks for the first time. A nonlinear dynamic response analysis of the coupling system was performed at different overload coefficients. The maximum damage depth-thickness ratio and sliding area ratio are proposed as performance evaluation indices. The failure mechanism of the model under strong earthquakes was elucidated. The residual displacement of the dam crest relative to the dam bottom in the stream direction is proposed as another performance evaluation index. Sudden changes and rapid growth are suggested as evaluation criteria to assess the ultimate seismic capacity of arch dams based on proposed multi-nonlinear coupled model. The results show that the strength failure of the dam body and stability failure of the dam abutments vary dynamically with the duration and intensity of the earthquake. Earthquake energy can be fully released by only one failure mode at low seismic intensity, whereas it is gradually released by both failure modes as the seismic intensity increases. The overload coefficient corresponding to the ultimate seismic capacity of the dam is concluded to be 2.0.

本研究构建了一个自由度超过 1000 万的高拱坝-地基系统模型。该模型首次创新性地加入了坝体强度破坏和坝基稳定性破坏的多重非线性耦合。在不同超载系数下，对耦合系统进行了非线性动态响应分析。提出了最大破坏深度厚度比和滑动面积比作为性能评价指标。阐明了模型在强震下的破坏机理。提出了坝顶相对于坝底在流向上的残余位移作为另一个性能评价指标。根据所提出的多非线性耦合模型，提出了突变和快速增长作为评估拱坝极限抗震能力的评价标准。结果表明，坝体的强度破坏和坝基的稳定性破坏随地震持续时间和烈度的变化而动态变化。在地震烈度较低时，只有一种破坏模式可以完全释放地震能量，而随着地震烈度的增加，两种破坏模式都会逐渐释放地震能量。与大坝极限抗震能力相对应的超载系数为 2.0。

{"title":"Seismic failure analysis of a high arch dam-foundation multiple nonlinear coupling system","authors":"Chunli Yan , Jin Tu , Hui Liang , Shengshan Guo , Deyu Li","doi":"10.1016/j.soildyn.2024.109001","DOIUrl":"10.1016/j.soildyn.2024.109001","url":null,"abstract":"<div><div>In this study, a high arch dam-foundation system model with more than ten million degrees of freedom was constructed. The model innovatively incorporates multiple nonlinear couplings of the strength failure of the dam body and stability failure of dam abutment blocks for the first time. A nonlinear dynamic response analysis of the coupling system was performed at different overload coefficients. The maximum damage depth-thickness ratio and sliding area ratio are proposed as performance evaluation indices. The failure mechanism of the model under strong earthquakes was elucidated. The residual displacement of the dam crest relative to the dam bottom in the stream direction is proposed as another performance evaluation index. Sudden changes and rapid growth are suggested as evaluation criteria to assess the ultimate seismic capacity of arch dams based on proposed multi-nonlinear coupled model. The results show that the strength failure of the dam body and stability failure of the dam abutments vary dynamically with the duration and intensity of the earthquake. Earthquake energy can be fully released by only one failure mode at low seismic intensity, whereas it is gradually released by both failure modes as the seismic intensity increases. The overload coefficient corresponding to the ultimate seismic capacity of the dam is concluded to be 2.0.</div></div>","PeriodicalId":49502,"journal":{"name":"Soil Dynamics and Earthquake Engineering","volume":"187 ","pages":"Article 109001"},"PeriodicalIF":4.2,"publicationDate":"2024-10-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142420078","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

An enhanced hybrid approach for spatial distribution of seismic liquefaction characteristics by integrating physics-based simulation and machine learning 基于物理的模拟与机器学习相结合的地震液化特征空间分布增强型混合方法

IF 4.2 2区工程技术 Q1 ENGINEERING, GEOLOGICAL

Soil Dynamics and Earthquake Engineering

Pub Date : 2024-10-07 DOI: 10.1016/j.soildyn.2024.109007

Zhenning Ba , Shujuan Han , Mengtao Wu , Yan Lu , Jianwen Liang

This study aims to propose an enhanced hybrid approach that combines physics-based simulation and machine learning to investigate the spatial distribution of seismic liquefaction characteristics. This innovative approach comprises two main components: Firstly, the physics-based frequency-wavenumber method is employed to construct the spatial-temporal field of ground motion in the study area, which provides ground motion quantities for assessing the liquefaction characteristic (e.g., liquefaction potential index) of the site. Subsequently, the seismic liquefaction parameters of the region are predicted using a machine learning (ML)-based SSA-XGBoost model. Due to the integration of physics-based simulation and machine learning techniques, which consider the effects of near-fault ground motion characteristics on seismic liquefaction, the proposed solution enables the evaluation of the spatial distribution of seismic liquefaction parameters under scenario earthquakes. In this study, the SSA-XGBoost model, constructed using the sparrow search algorithm (SSA) to automate and optimize the hyper-parameter tuning of the eXtreme gradient boosting (XGBoost), incorporates factors such as peak ground acceleration, magnitude scaling factor, ground water level, soil depth, vertical total overburden stress, vertical effective overburden stress, and fine content to evaluate their influence on liquefaction potential index. To demonstrate the effectiveness of the enhanced hybrid approach, the Jinnan district of Tianjin is taken as an example to evaluate liquefaction potential under various scenario earthquakes (M_w = 5.0, 5.5 and 6.0). The results show that the constructed SSA-XGBoost model has excellent predictive ability and is suitable for evaluating the liquefaction potential index of large-scale site soils. In the case of M_w 6.0 earthquake, most of the northern region of Jinnan district has the possibility of liquefaction, and some areas are seriously liquefied, and the liquefaction grade gradually decreases from the north to the south. These findings distinctly illustrate the spatial distribution of liquefaction characteristic parameters across the entire region, providing new insights and methods for similar studies and serving as a decision-making basis for the prevention and control of seismic liquefaction hazards.

本研究旨在提出一种增强型混合方法，结合基于物理的模拟和机器学习来研究地震液化特征的空间分布。这种创新方法包括两个主要部分：首先，采用基于物理的频率-波数方法构建研究区域的地动时空场，为评估场地的液化特征（如液化潜力指数）提供地动量。随后，使用基于机器学习（ML）的 SSA-XGBoost 模型对该地区的地震液化参数进行预测。由于基于物理的模拟与机器学习技术相结合，考虑了近断层地动特征对地震液化的影响，所提出的解决方案能够评估情景地震下地震液化参数的空间分布。在本研究中，利用麻雀搜索算法（SSA）构建了 SSA-XGBoost 模型，用于自动优化极限梯度提升（XGBoost）的超参数调整，将地表加速度峰值、震级缩放因子、地下水位、土层深度、垂直总覆土应力、垂直有效覆土应力和细粒含量等因素纳入模型，以评估它们对液化潜在指数的影响。为证明增强型混合方法的有效性，以天津市津南区为例，评估了各种情景地震（Mw = 5.0、5.5 和 6.0）下的液化潜力。结果表明，所构建的 SSA-XGBoost 模型具有出色的预测能力，适用于评估大型场地土壤的液化潜力指数。在 Mw 6.0 地震中，津南区北部大部分地区有液化的可能，部分地区液化严重，且液化等级由北向南逐渐降低。这些研究结果清楚地说明了全区液化特征参数的空间分布情况，为类似研究提供了新的认识和方法，也为防治地震液化灾害提供了决策依据。

{"title":"An enhanced hybrid approach for spatial distribution of seismic liquefaction characteristics by integrating physics-based simulation and machine learning","authors":"Zhenning Ba , Shujuan Han , Mengtao Wu , Yan Lu , Jianwen Liang","doi":"10.1016/j.soildyn.2024.109007","DOIUrl":"10.1016/j.soildyn.2024.109007","url":null,"abstract":"<div><div>This study aims to propose an enhanced hybrid approach that combines physics-based simulation and machine learning to investigate the spatial distribution of seismic liquefaction characteristics. This innovative approach comprises two main components: Firstly, the physics-based frequency-wavenumber method is employed to construct the spatial-temporal field of ground motion in the study area, which provides ground motion quantities for assessing the liquefaction characteristic (e.g., liquefaction potential index) of the site. Subsequently, the seismic liquefaction parameters of the region are predicted using a machine learning (ML)-based SSA-XGBoost model. Due to the integration of physics-based simulation and machine learning techniques, which consider the effects of near-fault ground motion characteristics on seismic liquefaction, the proposed solution enables the evaluation of the spatial distribution of seismic liquefaction parameters under scenario earthquakes. In this study, the SSA-XGBoost model, constructed using the sparrow search algorithm (SSA) to automate and optimize the hyper-parameter tuning of the eXtreme gradient boosting (XGBoost), incorporates factors such as peak ground acceleration, magnitude scaling factor, ground water level, soil depth, vertical total overburden stress, vertical effective overburden stress, and fine content to evaluate their influence on liquefaction potential index. To demonstrate the effectiveness of the enhanced hybrid approach, the Jinnan district of Tianjin is taken as an example to evaluate liquefaction potential under various scenario earthquakes (<em>M</em><sub>w</sub> = 5.0, 5.5 and 6.0). The results show that the constructed SSA-XGBoost model has excellent predictive ability and is suitable for evaluating the liquefaction potential index of large-scale site soils. In the case of <em>M</em><sub>w</sub> 6.0 earthquake, most of the northern region of Jinnan district has the possibility of liquefaction, and some areas are seriously liquefied, and the liquefaction grade gradually decreases from the north to the south. These findings distinctly illustrate the spatial distribution of liquefaction characteristic parameters across the entire region, providing new insights and methods for similar studies and serving as a decision-making basis for the prevention and control of seismic liquefaction hazards.</div></div>","PeriodicalId":49502,"journal":{"name":"Soil Dynamics and Earthquake Engineering","volume":"187 ","pages":"Article 109007"},"PeriodicalIF":4.2,"publicationDate":"2024-10-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142420080","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0