Computers & Structures最新文献_第6页

Direct two-scale finite element modeling of progressive failure in carbon fiber reinforced polymer composites with a localizing gradient damage model 基于局部梯度损伤模型的碳纤维增强聚合物复合材料渐进破坏直接双尺度有限元模拟

IF 4.8 2区工程技术 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers & Structures

Pub Date : 2025-12-26 DOI: 10.1016/j.compstruc.2025.108087

Yongyi Li , Xinyi Xu , Lianhua Ma , Biao Wang

The damage mechanisms in carbon fiber reinforced polymer composites mainly include matrix cracking, interfacial delamination, and fiber fracture. Among them, matrix failure and crack propagation constitute the principal contributors to the nonlinear response observed in the uniaxial tensile stress–strain behavior of carbon fiber reinforced polymer composites, frequently precipitating premature structural failure. Therefore, accurate simulation of matrix damage progression in carbon fiber reinforced polymer composites is crucial for predicting the failure behavior of continuous carbon fiber reinforced polymer composites. To address these issues, this study employs a direct two-scale finite element modelling framework to concurrently analyze macro-scale structural responses and meso-scale mechanical behaviors under tensile, shear, and bending loading conditions. The damage evolution in the polymer matrix is simulated through a localized gradient damage model, implemented through user-defined material and user-defined thermal analogy subroutines in commercial finite element software within a thermo-mechanical coupling framework. Furthermore, fiber–matrix interfacial debonding is characterized using a bilinear cohesive zone model. The findings demonstrate that the present direct two-scale finite element method exhibits excellent agreement with direct numerical simulation in capturing the progressive damage evolution mechanisms of carbon fiber reinforced polymer composites. Moreover, this approach offers a computationally efficient and practical tool for the design and optimization of carbon fiber reinforced polymer composites structures.

碳纤维增强聚合物复合材料的损伤机制主要包括基体开裂、界面分层和纤维断裂。其中，基体破坏和裂纹扩展是碳纤维增强聚合物复合材料单轴拉伸应力-应变行为非线性响应的主要原因，往往导致结构过早破坏。因此，准确模拟碳纤维增强聚合物复合材料的基体损伤过程对于预测连续碳纤维增强聚合物复合材料的破坏行为至关重要。为了解决这些问题，本研究采用直接的双尺度有限元建模框架，同时分析拉伸、剪切和弯曲加载条件下的宏观尺度结构响应和细观尺度力学行为。聚合物基体的损伤演化通过局部梯度损伤模型进行模拟，并通过商用有限元软件中自定义材料和自定义热模拟子程序在热-机械耦合框架内实现。此外，利用双线性黏聚区模型对纤维-基体界面脱粘进行了表征。研究结果表明，直接双尺度有限元方法与直接数值模拟方法在捕捉碳纤维增强聚合物复合材料的渐进损伤演化机制方面具有较好的一致性。此外，该方法为碳纤维增强聚合物复合材料结构的设计和优化提供了一种计算效率高、实用的工具。

{"title":"Direct two-scale finite element modeling of progressive failure in carbon fiber reinforced polymer composites with a localizing gradient damage model","authors":"Yongyi Li , Xinyi Xu , Lianhua Ma , Biao Wang","doi":"10.1016/j.compstruc.2025.108087","DOIUrl":"10.1016/j.compstruc.2025.108087","url":null,"abstract":"<div><div>The damage mechanisms in carbon fiber reinforced polymer composites mainly include matrix cracking, interfacial delamination, and fiber fracture. Among them, matrix failure and crack propagation constitute the principal contributors to the nonlinear response observed in the uniaxial tensile stress–strain behavior of carbon fiber reinforced polymer composites, frequently precipitating premature structural failure. Therefore, accurate simulation of matrix damage progression in carbon fiber reinforced polymer composites is crucial for predicting the failure behavior of continuous carbon fiber reinforced polymer composites. To address these issues, this study employs a direct two-scale finite element modelling framework to concurrently analyze macro-scale structural responses and meso-scale mechanical behaviors under tensile, shear, and bending loading conditions. The damage evolution in the polymer matrix is simulated through a localized gradient damage model, implemented through user-defined material and user-defined thermal analogy subroutines in commercial finite element software within a thermo-mechanical coupling framework. Furthermore, fiber–matrix interfacial debonding is characterized using a bilinear cohesive zone model. The findings demonstrate that the present direct two-scale finite element method exhibits excellent agreement with direct numerical simulation in capturing the progressive damage evolution mechanisms of carbon fiber reinforced polymer composites. Moreover, this approach offers a computationally efficient and practical tool for the design and optimization of carbon fiber reinforced polymer composites structures.</div></div>","PeriodicalId":50626,"journal":{"name":"Computers & Structures","volume":"321 ","pages":"Article 108087"},"PeriodicalIF":4.8,"publicationDate":"2025-12-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145840386","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

Quantile-based multiscale topology optimization of geometrically asymmetric porous sandwich structures under uncertainties 不确定条件下几何不对称多孔夹层结构的分位数多尺度拓扑优化

IF 4.8 2区工程技术 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers & Structures

Pub Date : 2025-12-26 DOI: 10.1016/j.compstruc.2025.108085

Zhimiao Zou , Zhe Ding , Kai Shan , Wei-Hsin Liao

Geometrically asymmetric porous sandwich structures (GAPSSs) often outperform conventional designs with identical face sheets under asymmetric load and boundary conditions, since more design freedoms are provided for optimization. However, most current works on the GAPSSs are based on deterministic assumptions and developed by analytical or experimental methods, while limited attention is given to the optimization for the GAPSSs under uncertainties. This study proposes a novel multiscale design approach for GAPSSs under uncertainty based on topology optimization, which can determine the optimized face-sheet thicknesses, as well as the core configuration while maintaining structural reliability. Firstly, the optimization begins at macroscale, where the optimized face-sheet thicknesses are determined through the variable thickness sheet method. Then, the configuration of periodic unit cell is optimized using the parametric level set method at microscale. At both stages, the quantile-based formulation is applied to enhance structural reliability. During the quantile-based process, Monte Carlos Simulation is used for quantile estimation with the assistance of Kriging model. To validate the proposed methodology and highlight its superior performance over conventional designs, a series of 2D and 3D numerical case studies as well as some comparative analyses are conducted.

几何不对称多孔夹层结构（gapss）在非对称载荷和边界条件下通常优于具有相同面板的传统设计，因为它为优化提供了更多的设计自由度。然而，目前关于gapss的研究大多基于确定性假设，采用分析或实验方法，而对不确定条件下gapss的优化研究较少。本研究提出了一种基于拓扑优化的不确定条件下gapss多尺度设计方法，该方法可以在保证结构可靠性的前提下确定优化的面板厚度和核心配置。首先从宏观层面入手，通过变厚法确定优化后的面板厚度；然后，在微尺度下，采用参数水平集方法对周期单元胞的结构进行优化。在这两个阶段，采用基于分位数的公式来提高结构的可靠性。在基于分位数的过程中，在Kriging模型的辅助下，使用Monte Carlos Simulation进行分位数估计。为了验证所提出的方法并突出其优于传统设计的性能，进行了一系列二维和三维数值案例研究以及一些比较分析。

{"title":"Quantile-based multiscale topology optimization of geometrically asymmetric porous sandwich structures under uncertainties","authors":"Zhimiao Zou , Zhe Ding , Kai Shan , Wei-Hsin Liao","doi":"10.1016/j.compstruc.2025.108085","DOIUrl":"10.1016/j.compstruc.2025.108085","url":null,"abstract":"<div><div>Geometrically asymmetric porous sandwich structures (GAPSSs) often outperform conventional designs with identical face sheets under asymmetric load and boundary conditions, since more design freedoms are provided for optimization. However, most current works on the GAPSSs are based on deterministic assumptions and developed by analytical or experimental methods, while limited attention is given to the optimization for the GAPSSs under uncertainties. This study proposes a novel multiscale design approach for GAPSSs under uncertainty based on topology optimization, which can determine the optimized face-sheet thicknesses, as well as the core configuration while maintaining structural reliability. Firstly, the optimization begins at macroscale, where the optimized face-sheet thicknesses are determined through the variable thickness sheet method. Then, the configuration of periodic unit cell is optimized using the parametric level set method at microscale. At both stages, the quantile-based formulation is applied to enhance structural reliability. During the quantile-based process, Monte Carlos Simulation is used for quantile estimation with the assistance of Kriging model. To validate the proposed methodology and highlight its superior performance over conventional designs, a series of 2D and 3D numerical case studies as well as some comparative analyses are conducted.</div></div>","PeriodicalId":50626,"journal":{"name":"Computers & Structures","volume":"321 ","pages":"Article 108085"},"PeriodicalIF":4.8,"publicationDate":"2025-12-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145840385","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

A finite strain elastic-corrector rate formulation for void growth models 孔洞生长模型的有限应变弹性修正率公式

IF 4.8 2区工程技术 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers & Structures

Pub Date : 2025-12-25 DOI: 10.1016/j.compstruc.2025.108072

Meijuan Zhang , Guadalupe Vadillo , Francisco J. Montáns

Recently, a novel framework for large strain elastoplasticity has been introduced. This approach preserves the multiplicative decomposition of the deformation gradient, avoiding Green-like elastoplastic decompositions of total strains, but resulting in additive predictor-corrector algorithmic implementations as in small strains, and using any form of hyperelastic stored energy. Using this approach, large strain complexities reduce to non-iterative kinematical pre- and post-processors. In this work, we extend the elastic-corrector rates formulation for dealing with non-isochoric void-growth plasticity. In contrast to other large strain formulations, in the present work the relation of the void growth with the plastic flow of the matrix is not assumed, but obtained from the multiplicative decomposition, preserving exactly the kinematics of the volumetric parts. We use the Gurson-Tvergaard-Needleman (GTN) function to demonstrate the use of the framework as well as the implementation of a fully implicit Newton algorithm. The novel GTN formulation is hyperelastic, does not need the exponential mapping, and keeps an additive algorithmic update as in the infinitesimal formulation. Other non-isochoric functions can also be employed with the same kinematics.

近年来，提出了一种新的大应变弹塑性框架。这种方法保留了变形梯度的乘法分解，避免了总应变的格林样弹塑性分解，但导致了小应变的加性预测校正算法的实现，并使用了任何形式的超弹性存储能量。使用这种方法，大的应变复杂性减少到非迭代的运动学预处理和后处理。在这项工作中，我们扩展了弹性校正率公式来处理非等时空隙生长塑性。与其他大应变公式不同的是，在本工作中，孔洞生长与基体塑性流动的关系不是假设的，而是从乘法分解中得到的，准确地保留了体积部分的运动学。我们使用Gurson-Tvergaard-Needleman （GTN）函数来演示该框架的使用以及完全隐式牛顿算法的实现。新的GTN公式是超弹性的，不需要指数映射，并且像无穷小公式一样保持加性算法更新。其他的非等时函数也可以采用相同的运动学。

{"title":"A finite strain elastic-corrector rate formulation for void growth models","authors":"Meijuan Zhang , Guadalupe Vadillo , Francisco J. Montáns","doi":"10.1016/j.compstruc.2025.108072","DOIUrl":"10.1016/j.compstruc.2025.108072","url":null,"abstract":"<div><div>Recently, a novel framework for large strain elastoplasticity has been introduced. This approach preserves the multiplicative decomposition of the deformation gradient, avoiding Green-like elastoplastic decompositions of total strains, but resulting in additive predictor-corrector algorithmic implementations as in small strains, and using any form of hyperelastic stored energy. Using this approach, large strain complexities reduce to non-iterative kinematical pre- and post-processors. In this work, we extend the elastic-corrector rates formulation for dealing with non-isochoric void-growth plasticity. In contrast to other large strain formulations, in the present work the relation of the void growth with the plastic flow of the matrix is not assumed, but obtained from the multiplicative decomposition, preserving exactly the kinematics of the volumetric parts. We use the Gurson-Tvergaard-Needleman (GTN) function to demonstrate the use of the framework as well as the implementation of a fully implicit Newton algorithm. The novel GTN formulation is hyperelastic, does not need the exponential mapping, and keeps an additive algorithmic update as in the infinitesimal formulation. Other non-isochoric functions can also be employed with the same kinematics.</div></div>","PeriodicalId":50626,"journal":{"name":"Computers & Structures","volume":"321 ","pages":"Article 108072"},"PeriodicalIF":4.8,"publicationDate":"2025-12-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145823155","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

A priori mesh error calculation method considering different constitutive models in impact and explosion simulations 一种考虑不同本构模型的冲击爆炸模拟先验网格误差计算方法

IF 4.8 2区工程技术 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers & Structures

Pub Date : 2025-12-24 DOI: 10.1016/j.compstruc.2025.108081

Boyang Zhu , Xin Bao , Gang Li , Qiushi Yan , Jingbo Liu , Shiwei Wang

The determination of optimal mesh size constitutes a critical factor in the numerical simulation of wave propagation, particularly under high-frequency excitations such as blast or impact loadings. In such contexts, mesh dimension significantly influences both the accuracy and efficiency of computations. Existing meshing strategies predominantly rely on numerical mesh independence analysis, which necessitates extensive computations to identify the requisite mesh size, resulting in diminished computational efficiency and limited generality. This study presents a novel theoretical approach for quantifying meshing-induced errors considering different constitutive models, founded upon wave field propagation theory and frequency-domain analysis. This approach enables the a priori estimation of appropriate mesh size along with its associated frequency truncation error through theoretical formulas, thereby obviating the need for exhaustive mesh sensitive studies and reducing computational expenditure. Furthermore, this study examines the impacts of input waveforms characteristics, geometric attenuation and material nonlinearity upon mesh-induced errors, providing theoretical explanations for empirical findings from previous studies. Numerical simulations of traditional engineering materials demonstrate the accuracy and engineering applicability of the proposed error estimation method. And by combining the proposed error estimation method with mesh refinement techniques, an adaptive meshing strategy that adheres to the error threshold is proposed. This research provides a more universal, efficient, and precise foundation for mesh generation in explosion simulations, potentially driving the advancement of highly refined and computationally efficient numerical modeling techniques.

确定最佳网格尺寸是波传播数值模拟中的一个关键因素，特别是在爆炸或冲击载荷等高频激励下。在这种情况下，网格尺寸对计算的精度和效率都有很大的影响。现有的网格划分策略主要依赖于数值网格无关分析，这需要大量的计算来确定所需的网格尺寸，从而降低了计算效率，限制了通用性。基于波场传播理论和频域分析，提出了一种考虑不同本构模型的网格误差量化的新理论方法。该方法能够通过理论公式先验估计合适的网格尺寸及其相关的频率截断误差，从而避免了详尽的网格敏感研究的需要，减少了计算开销。此外，本研究还考察了输入波形特性、几何衰减和材料非线性对网格误差的影响，为以往的实证研究结果提供了理论解释。对传统工程材料的数值模拟验证了误差估计方法的准确性和工程适用性。并将误差估计方法与网格细化技术相结合，提出了一种坚持误差阈值的自适应网格划分策略。该研究为爆炸模拟中的网格生成提供了一个更加通用、高效和精确的基础，有可能推动高度精细和计算效率高的数值模拟技术的发展。

{"title":"A priori mesh error calculation method considering different constitutive models in impact and explosion simulations","authors":"Boyang Zhu , Xin Bao , Gang Li , Qiushi Yan , Jingbo Liu , Shiwei Wang","doi":"10.1016/j.compstruc.2025.108081","DOIUrl":"10.1016/j.compstruc.2025.108081","url":null,"abstract":"<div><div>The determination of optimal mesh size constitutes a critical factor in the numerical simulation of wave propagation, particularly under high-frequency excitations such as blast or impact loadings. In such contexts, mesh dimension significantly influences both the accuracy and efficiency of computations. Existing meshing strategies predominantly rely on numerical mesh independence analysis, which necessitates extensive computations to identify the requisite mesh size, resulting in diminished computational efficiency and limited generality. This study presents a novel theoretical approach for quantifying meshing-induced errors considering different constitutive models, founded upon wave field propagation theory and frequency-domain analysis. This approach enables the a priori estimation of appropriate mesh size along with its associated frequency truncation error through theoretical formulas, thereby obviating the need for exhaustive mesh sensitive studies and reducing computational expenditure. Furthermore, this study examines the impacts of input waveforms characteristics, geometric attenuation and material nonlinearity upon mesh-induced errors, providing theoretical explanations for empirical findings from previous studies. Numerical simulations of traditional engineering materials demonstrate the accuracy and engineering applicability of the proposed error estimation method. And by combining the proposed error estimation method with mesh refinement techniques, an adaptive meshing strategy that adheres to the error threshold is proposed. This research provides a more universal, efficient, and precise foundation for mesh generation in explosion simulations, potentially driving the advancement of highly refined and computationally efficient numerical modeling techniques.</div></div>","PeriodicalId":50626,"journal":{"name":"Computers & Structures","volume":"321 ","pages":"Article 108081"},"PeriodicalIF":4.8,"publicationDate":"2025-12-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145823162","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

Development of isogeometric-finite element unified numerical methodology for nonlinear planar instability analysis of dual-phase SiO2-graphene nanoplatetlets reinforced shallow curved microbeams 双相二氧化硅-石墨烯纳米片增强浅弯曲微梁非线性平面失稳分析等几何-有限元统一数值方法的发展

IF 4.8 2区工程技术 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers & Structures

Pub Date : 2025-12-24 DOI: 10.1016/j.compstruc.2025.108082

Saeid Sahmani , Reza Ansari , Mohammad Kazem Hassanzadeh-Aghdam , Mona Zareichian

The main intention of the proposed multiscale framework is to employ an isogeometric analysis formulation for the size-dependent nonlinear planar instability analysis of dual-phase nanocomposite shallow curved beams at the microscale. A finite element–based micromechanics model is developed at the representative volume element level to capture the material properties. The homogenized properties obtained from the representative volume element-level finite element analysis are directly incorporated into the isogeometric model. This coupling enables accurate surveying of the small scale-dependent nonlinear in-plane stability characteristics of uniformly laterally loaded dual-phase inhomogeneous shallow curved microbeams reinforced with SiO₂ nanoparticles and graphene nanoplatelets, while embracing distinct strain gradient tensors. In this regard, cuboid-shaped representative volume elements are employed. This enables consideration of the interphase between the dual-phase nanofillers and the polymer, as well as the critical role of nanofiller agglomeration, in order to create an accurate multiscale correlation. Additionally, non-uniform rational B-splines are utilized in the relevant discretization process. This process involves distinct microstructural-dependent strain gradient tensors. The numerical results reveal that increasing the SiO₂ nanoparticle volume fraction significantly enhances both the upper and lower limit loads by nearly 69.5%. This increase does not markedly affect the axial resultant load or the lateral deflection. Conversely, increasing the SiO₂ nanoparticle diameter at a fixed volume fraction notably decreases the load-bearing capacity by about 49.5%. Similarly, a rise in graphene nanoplatelet thickness leads to an approximately 61.1% reduction in the stability limits. The inclusion of the interphase region between the nanofillers and the matrix improves the upper and lower limit loads by around 17.4%, demonstrating its reinforcing influence. Furthermore, aligning nanofillers along the beam’s longitudinal direction increases the limit loads by roughly 48.1% compared to the random dispersion case. In contrast, agglomeration has the opposite effect, reducing the load-carrying capacity by about 12.4%.

所提出的多尺度框架的主要目的是采用一种等几何分析公式来分析微尺度下双相纳米复合材料浅弯曲梁的尺寸相关非线性平面不稳定性。在具有代表性的体元水平上建立了基于有限元的细观力学模型来捕捉材料的特性。从具有代表性的体积单元级有限元分析中获得的均质性直接纳入等几何模型。这种耦合可以精确测量均匀横向加载的双相非均匀浅弯曲微梁的小尺度非线性面内稳定性特性，这些微梁由SiO2纳米颗粒和石墨烯纳米片增强，同时包含不同的应变梯度张量。在这方面，采用长方体形状的代表性体积元素。这可以考虑双相纳米填料和聚合物之间的界面，以及纳米填料团聚的关键作用，以便创建精确的多尺度关联。另外，在相关的离散化过程中采用了非均匀有理b样条。这一过程涉及不同的微观结构相关的应变梯度张量。数值结果表明，增加SiO2纳米颗粒体积分数，上、下限载荷均显著提高近69.5%。这种增加并不显著影响轴向合成载荷或横向挠度。相反，在一定体积分数下，增加SiO2纳米颗粒直径，其承载能力显著降低约49.5%。同样，石墨烯纳米板厚度的增加导致稳定性极限降低约61.1%。纳米填料与基体间相区域的加入使材料的上下限载荷提高了17.4%左右，表明了纳米填料的强化作用。此外，与随机分散的情况相比，沿梁的纵向排列纳米填料增加了大约48.1%的极限载荷。而结块则相反，其承载能力降低约12.4%。

{"title":"Development of isogeometric-finite element unified numerical methodology for nonlinear planar instability analysis of dual-phase SiO2-graphene nanoplatetlets reinforced shallow curved microbeams","authors":"Saeid Sahmani , Reza Ansari , Mohammad Kazem Hassanzadeh-Aghdam , Mona Zareichian","doi":"10.1016/j.compstruc.2025.108082","DOIUrl":"10.1016/j.compstruc.2025.108082","url":null,"abstract":"<div><div>The main intention of the proposed multiscale framework is to employ an isogeometric analysis formulation for the size-dependent nonlinear planar instability analysis of dual-phase nanocomposite shallow curved beams at the microscale. A finite element–based micromechanics model is developed at the representative volume element level to capture the material properties. The homogenized properties obtained from the representative volume element-level finite element analysis are directly incorporated into the isogeometric model. This coupling enables accurate surveying of the small scale-dependent nonlinear in-plane stability characteristics of uniformly laterally loaded dual-phase inhomogeneous shallow curved microbeams reinforced with SiO<sub>2</sub> nanoparticles and graphene nanoplatelets, while embracing distinct strain gradient tensors. In this regard, cuboid-shaped representative volume elements are employed. This enables consideration of the interphase between the dual-phase nanofillers and the polymer, as well as the critical role of nanofiller agglomeration, in order to create an accurate multiscale correlation. Additionally, non-uniform rational B-splines are utilized in the relevant discretization process. This process involves distinct microstructural-dependent strain gradient tensors. The numerical results reveal that increasing the SiO<sub>2</sub> nanoparticle volume fraction significantly enhances both the upper and lower limit loads by nearly 69.5%. This increase does not markedly affect the axial resultant load or the lateral deflection. Conversely, increasing the SiO<sub>2</sub> nanoparticle diameter at a fixed volume fraction notably decreases the load-bearing capacity by about 49.5%. Similarly, a rise in graphene nanoplatelet thickness leads to an approximately 61.1% reduction in the stability limits. The inclusion of the interphase region between the nanofillers and the matrix improves the upper and lower limit loads by around 17.4%, demonstrating its reinforcing influence. Furthermore, aligning nanofillers along the beam’s longitudinal direction increases the limit loads by roughly 48.1% compared to the random dispersion case. In contrast, agglomeration has the opposite effect, reducing the load-carrying capacity by about 12.4%.</div></div>","PeriodicalId":50626,"journal":{"name":"Computers & Structures","volume":"321 ","pages":"Article 108082"},"PeriodicalIF":4.8,"publicationDate":"2025-12-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145823163","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

Enhanced load-shape integrated measurement of plate structures based on matched section analysis 基于匹配截面分析的板式结构增强荷载-形状综合测量

IF 4.8 2区工程技术 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers & Structures

Pub Date : 2025-12-24 DOI: 10.1016/j.compstruc.2025.108083

Yanhao Guo , Yuhui Hu , Yan Zeng , Feifei Zhao , Hong Bao

Load and deformation measurement of plate structures are fundamental for structural design and health monitoring. However, existing methods are limited to independently reconstructing either deformation or load, without being capable of integrating sensing both the deformation and load in a unified sensing formulation. To address this gap, a matched-section discretization method is proposed, transforming the plate into two orthogonal sets of equivalent beam elements. The structural displacement field is formulated in terms of cross-sectional deformation parameters, internal force/moment parameters, and uniformly distributed loads. Mapping equations between discrete strain measurements and these parameters are then established. By inputting real-time discrete strain data, the least-squares inverse method is employed to simultaneously reconstruct the deformation and load distributions. The proposed method is validated through simulations and experiments on plate structures with varying thicknesses, boundary conditions, and load cases. In the simulation results, the maximum deformation sensing error is limited to 3.1%, and the load sensing error remains below 8.1% with one sensing element. These findings demonstrate the high accuracy and practical potential of the proposed approach in structural health monitoring and intelligent structural design.

板结构的荷载和变形测量是结构设计和健康监测的基础。然而，现有的方法仅限于单独重建变形或载荷，无法将变形和载荷集成到统一的传感公式中。为了解决这一缺陷，提出了一种匹配截面离散化方法，将板转化为两组正交的等效梁单元。结构位移场由截面变形参数、内力/弯矩参数和均布荷载组成。然后建立离散应变测量值与这些参数之间的映射方程。通过输入实时离散应变数据，采用最小二乘反算法同时重构变形和载荷分布。通过对不同厚度、边界条件和载荷情况下的板结构进行仿真和实验，验证了该方法的有效性。在仿真结果中，变形传感的最大误差控制在3.1%以内，负载传感的最大误差控制在8.1%以内。这些结果证明了该方法在结构健康监测和智能结构设计中的高精度和实用潜力。

{"title":"Enhanced load-shape integrated measurement of plate structures based on matched section analysis","authors":"Yanhao Guo , Yuhui Hu , Yan Zeng , Feifei Zhao , Hong Bao","doi":"10.1016/j.compstruc.2025.108083","DOIUrl":"10.1016/j.compstruc.2025.108083","url":null,"abstract":"<div><div>Load and deformation measurement of plate structures are fundamental for structural design and health monitoring. However, existing methods are limited to independently reconstructing either deformation or load, without being capable of integrating sensing both the deformation and load in a unified sensing formulation. To address this gap, a matched-section discretization method is proposed, transforming the plate into two orthogonal sets of equivalent beam elements. The structural displacement field is formulated in terms of cross-sectional deformation parameters, internal force/moment parameters, and uniformly distributed loads. Mapping equations between discrete strain measurements and these parameters are then established. By inputting real-time discrete strain data, the least-squares inverse method is employed to simultaneously reconstruct the deformation and load distributions. The proposed method is validated through simulations and experiments on plate structures with varying thicknesses, boundary conditions, and load cases. In the simulation results, the maximum deformation sensing error is limited to 3.1%, and the load sensing error remains below 8.1% with one sensing element. These findings demonstrate the high accuracy and practical potential of the proposed approach in structural health monitoring and intelligent structural design.</div></div>","PeriodicalId":50626,"journal":{"name":"Computers & Structures","volume":"321 ","pages":"Article 108083"},"PeriodicalIF":4.8,"publicationDate":"2025-12-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145823164","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

Efficient reliability-based design optimization of dynamic linear systems subjected to filtered white-noise inputs 滤波白噪声输入下动态线性系统的高效可靠性优化设计

IF 4.8 2区工程技术 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers & Structures

Pub Date : 2025-12-23 DOI: 10.1016/j.compstruc.2025.108047

Nir Itzhak Ben-Israel , Michalis Fragiadakis , Oren Lavan

This paper presents a reliability-based design optimization methodology for dynamic linear structural systems, subjected to filtered white-noise excitations. A novel double-loop approach is proposed, where both inner and outer optimization problems are efficiently solved using gradient-based algorithms, combining a specifically-tailored adjoint method for the sensitivity analyses, while the reliability analyses are carried out using the first-order-reliability-method (FORM). Key to the proposed approach is that the mean-square of the responses of interest is efficiently evaluated via the Lyapunov equations, which leads to faster and more efficient reliability-based designs. The application of the methodology is demonstrated in three examples in which control systems are optimized. First, a serviceability limit-state function is considered for a footbridge subjected to crowd load excitation. The footbridge vibrations are controlled by viscous dampers and tuned-mass dampers. Second, a limit-state function is considered for a bending moment frame, undergoing a seismic event. The interstorey drifts of the frame are controlled one time by viscous dampers and another time by tuned-mass-dampers. Third, a large benchmark frame is optimized. Verification of the probability of failure achieved via the first-order-reliability-method is performed using Monte Carlo simulations (MCS), showing good agreement between the two methods. A clear resemblance is shown between optimal design achieved via zero-order method, although with significantly reduced computational cost.

本文提出了一种基于可靠性的动态线性结构系统的优化设计方法。提出了一种新的双环方法，利用基于梯度的算法有效地解决内外优化问题，结合专门定制的伴随方法进行灵敏度分析，而可靠性分析则使用一阶可靠性方法（FORM）进行。该方法的关键是通过李雅普诺夫方程有效地评估感兴趣的响应的均方，从而实现更快，更有效的基于可靠性的设计。通过三个优化控制系统的实例说明了该方法的应用。首先，考虑了人群荷载作用下人行桥的可用性极限状态函数。人行桥的振动由粘性阻尼器和调谐质量阻尼器控制。其次，考虑了受地震作用的弯矩框架的极限状态函数。框架的层间漂移分别由粘性阻尼器和调谐质量阻尼器控制。第三，对大型基准框架进行优化。利用蒙特卡罗模拟（MCS）验证了一阶可靠度法获得的失效概率，结果表明两种方法具有较好的一致性。通过零阶方法实现的优化设计有明显的相似之处，尽管大大降低了计算成本。

{"title":"Efficient reliability-based design optimization of dynamic linear systems subjected to filtered white-noise inputs","authors":"Nir Itzhak Ben-Israel , Michalis Fragiadakis , Oren Lavan","doi":"10.1016/j.compstruc.2025.108047","DOIUrl":"10.1016/j.compstruc.2025.108047","url":null,"abstract":"<div><div>This paper presents a reliability-based design optimization methodology for dynamic linear structural systems, subjected to filtered white-noise excitations. A novel double-loop approach is proposed, where both inner and outer optimization problems are efficiently solved using gradient-based algorithms, combining a specifically-tailored adjoint method for the sensitivity analyses, while the reliability analyses are carried out using the first-order-reliability-method (FORM). Key to the proposed approach is that the mean-square of the responses of interest is efficiently evaluated via the Lyapunov equations, which leads to faster and more efficient reliability-based designs. The application of the methodology is demonstrated in three examples in which control systems are optimized. First, a serviceability limit-state function is considered for a footbridge subjected to crowd load excitation. The footbridge vibrations are controlled by viscous dampers and tuned-mass dampers. Second, a limit-state function is considered for a bending moment frame, undergoing a seismic event. The interstorey drifts of the frame are controlled one time by viscous dampers and another time by tuned-mass-dampers. Third, a large benchmark frame is optimized. Verification of the probability of failure achieved via the first-order-reliability-method is performed using Monte Carlo simulations (MCS), showing good agreement between the two methods. A clear resemblance is shown between optimal design achieved via zero-order method, although with significantly reduced computational cost.</div></div>","PeriodicalId":50626,"journal":{"name":"Computers & Structures","volume":"321 ","pages":"Article 108047"},"PeriodicalIF":4.8,"publicationDate":"2025-12-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145823167","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

Experimental and numerical analysis of angle-ply laminated plates based on a refined higher-order theory 基于改进高阶理论的角层合板试验与数值分析

IF 4.8 2区工程技术 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers & Structures

Pub Date : 2025-12-23 DOI: 10.1016/j.compstruc.2025.108078

Jie Mei , Shengbo Ling , Yunhao Huang

Accurately predicting the deformation of composite laminates is challenging due to the zig-zag effect in in-plane displacements, which arises from significant stiffness differences between layers. To address these limitations, this study proposes a novel Angle-ply Higher-order Deformation Theory coupled with a computationally efficient refined finite element, to investigate the bending response of angle-ply hybrid composite laminates. The AHDT is established via a global–local concept, utilizing Legendre polynomials to capture the zig-zag effect and interlaminar stress continuity phenomenon. The corresponding triangular element implements this model using C⁰ and C¹ continuities, ensuring high accuracy and computational efficiency. The accuracy and computational efficiency of the model are rigorously assessed through comparisons with three-dimensional elastic solutions and a three-dimensional finite element model. Furthermore, validation is provided via three-point bending tests on carbon/glass fiber hybrid composites, with full-field deformation measured using Digital Image Correlation. Both theoretical and experimental results confirm that the stacking sequence is the predominant factor affecting flexural stiffness, outweighing both material type and ply orientation. A change in the stacking sequence alone can reduce deflection by up to 70.77%. Consequently, this study conclusively shows that a rational design of the stacking sequence, material combination, and ply orientation is critical for optimizing the flexural performance of composite laminates.

由于层间显著的刚度差异导致的平面内位移呈锯齿状，因此准确预测复合材料层合板的变形具有挑战性。为了解决这些限制，本研究提出了一种新的角铺层高阶变形理论，结合计算效率高的精细有限元，来研究角铺层混杂复合材料层合板的弯曲响应。AHDT是通过全局-局部概念建立的，利用勒让德多项式来捕捉锯齿效应和层间应力连续性现象。相应的三角形单元使用C⁰和C1连续性实现该模型，确保高精度和计算效率。通过与三维弹性解和三维有限元模型的比较，严格评估了模型的精度和计算效率。此外，通过碳/玻璃纤维混合复合材料的三点弯曲测试提供了验证，并使用数字图像相关测量了全场变形。理论和实验结果均证实，叠层顺序是影响抗弯刚度的主要因素，其影响大于材料类型和铺层方向。仅改变堆叠顺序就可以减少高达70.77%的挠度。因此，本研究最终表明，合理设计叠层顺序、材料组合和层向是优化复合材料层合板抗弯性能的关键。

{"title":"Experimental and numerical analysis of angle-ply laminated plates based on a refined higher-order theory","authors":"Jie Mei , Shengbo Ling , Yunhao Huang","doi":"10.1016/j.compstruc.2025.108078","DOIUrl":"10.1016/j.compstruc.2025.108078","url":null,"abstract":"<div><div>Accurately predicting the deformation of composite laminates is challenging due to the zig-zag effect in in-plane displacements, which arises from significant stiffness differences between layers. To address these limitations, this study proposes a novel Angle-ply Higher-order Deformation Theory coupled with a computationally efficient refined finite element, to investigate the bending response of angle-ply hybrid composite laminates. The AHDT is established via a global–local concept, utilizing Legendre polynomials to capture the zig-zag effect and interlaminar stress continuity phenomenon. The corresponding triangular element implements this model using C⁰ and C<sup>1</sup> continuities, ensuring high accuracy and computational efficiency. The accuracy and computational efficiency of the model are rigorously assessed through comparisons with three-dimensional elastic solutions and a three-dimensional finite element model. Furthermore, validation is provided via three-point bending tests on carbon/glass fiber hybrid composites, with full-field deformation measured using Digital Image Correlation. Both theoretical and experimental results confirm that the stacking sequence is the predominant factor affecting flexural stiffness, outweighing both material type and ply orientation. A change in the stacking sequence alone can reduce deflection by up to 70.77%. Consequently, this study conclusively shows that a rational design of the stacking sequence, material combination, and ply orientation is critical for optimizing the flexural performance of composite laminates.</div></div>","PeriodicalId":50626,"journal":{"name":"Computers & Structures","volume":"321 ","pages":"Article 108078"},"PeriodicalIF":4.8,"publicationDate":"2025-12-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145823166","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

Structural verification of a gravity dam under seismic action using machine learning 地震作用下重力坝结构的机器学习验证

IF 4.8 2区工程技术 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers & Structures

Pub Date : 2025-12-22 DOI: 10.1016/j.compstruc.2025.108077

Neander Berto Mendes , Lineu José Pedroso

Dams play a critical role in hydroelectric power generation, a strategic and renewable energy source. However, their potential for collapse poses significant risks, including environmental damage and loss of life. Ensuring dam safety, especially under seismic loading, requires detailed analysis of the interaction between structure, fluid, and foundation soil. This study proposes an integrated approach using Artificial Intelligence and Machine Learning to assess the seismic stability of gravity dams. A finite element model was developed for a dam-reservoir-foundation system, incorporating spatial variability in the concrete’s elastic modulus through a stochastic simulation based on a continuous random field with a normalized correlation length of 0.10. The simulation results were used to train decision tree-based ML algorithms. Among them, the Extreme Gradient Boosting (XGBoost) model exhibited the highest predictive performance, achieving a correlation coefficient of 92.64 % on the test set. XGBoost was employed to estimate displacements and tensile stresses at critical locations, such as the neck and heel of the dam, where the maximum stress reached 9.36 MPa. Validation with additional simulations demonstrated prediction errors below 6.61 %, indicating strong agreement with the finite element results. This approach contributes to risk identification and supports more effective maintenance and safety management of dam infrastructure.

水力发电是一种战略性的可再生能源，水坝在水力发电中发挥着关键作用。然而，它们倒塌的可能性带来了重大风险，包括环境破坏和生命损失。确保大坝安全，特别是在地震荷载作用下，需要对结构、流体和地基土之间的相互作用进行详细的分析。本研究提出了一种利用人工智能和机器学习的综合方法来评估重力坝的地震稳定性。建立了大坝-水库-基础系统的有限元模型，通过基于归一化相关长度为0.10的连续随机场的随机模拟，考虑了混凝土弹性模量的空间变异性。仿真结果用于训练基于决策树的机器学习算法。其中，极端梯度增强（Extreme Gradient Boosting, XGBoost）模型的预测性能最高，在测试集上的相关系数达到92.64%。利用XGBoost对坝体颈部和坝后跟等关键部位的位移和拉应力进行估算，其中最大应力达到9.36 MPa。通过额外的模拟验证，预测误差低于6.61%，表明与有限元结果非常吻合。这种方法有助于风险识别，并支持更有效的大坝基础设施维护和安全管理。

{"title":"Structural verification of a gravity dam under seismic action using machine learning","authors":"Neander Berto Mendes , Lineu José Pedroso","doi":"10.1016/j.compstruc.2025.108077","DOIUrl":"10.1016/j.compstruc.2025.108077","url":null,"abstract":"<div><div>Dams play a critical role in hydroelectric power generation, a strategic and renewable energy source. However, their potential for collapse poses significant risks, including environmental damage and loss of life. Ensuring dam safety, especially under seismic loading, requires detailed analysis of the interaction between structure, fluid, and foundation soil. This study proposes an integrated approach using Artificial Intelligence and Machine Learning to assess the seismic stability of gravity dams. A finite element model was developed for a dam-reservoir-foundation system, incorporating spatial variability in the concrete’s elastic modulus through a stochastic simulation based on a continuous random field with a normalized correlation length of 0.10. The simulation results were used to train decision tree-based ML algorithms. Among them, the Extreme Gradient Boosting (XGBoost) model exhibited the highest predictive performance, achieving a correlation coefficient of 92.64 % on the test set. XGBoost was employed to estimate displacements and tensile stresses at critical locations, such as the neck and heel of the dam, where the maximum stress reached 9.36 MPa. Validation with additional simulations demonstrated prediction errors below 6.61 %, indicating strong agreement with the finite element results. This approach contributes to risk identification and supports more effective maintenance and safety management of dam infrastructure.</div></div>","PeriodicalId":50626,"journal":{"name":"Computers & Structures","volume":"321 ","pages":"Article 108077"},"PeriodicalIF":4.8,"publicationDate":"2025-12-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145813949","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

Numerical modelling of reinforced concrete using lattice particle method with extension to FSI-induced structural failure via SPH coupling 基于点阵粒子法的钢筋混凝土数值模拟，并将其推广到SPH耦合下的fsi结构破坏

IF 4.8 2区工程技术 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers & Structures

Pub Date : 2025-12-21 DOI: 10.1016/j.compstruc.2025.108075

W.C. Low , K.C. Ng , Kaiwei Tian , H.K. Ng

In this paper, a fluid structure interaction (FSI) solver based on smoothed particle hydrodynamics (SPH) and lattice particle method (LPM) is further developed to model FSI phenomena with structural failure involving concrete and reinforced concrete (RC). First, the development of the LPM structural solver to model reinforced concrete is presented. An explicit return-mapping algorithm is proposed in the framework of LPM to model the failure of reinforced concrete structures. The predictive capability and generality of the proposed LPM model is assessed against experimental results covering wide range of failure modes seen in RC beams. Key features of the flexural and shear failure in RC beams are captured by the model. With particle size refinement, the structural response and crack pattern show consistent results. Upon verification of the structural solver, the coupled SPH-LPM method is applied to simulate several FSI test cases with solid fracture. Finally, the capability of the current method in tackling complex, real-world FSI cases with structural failure is demonstrated through the test case of failure of plain and reinforced concrete wall due to tsunami-type wave.

本文基于光滑颗粒流体力学（SPH）和点阵颗粒法（LPM）进一步开发了流固耦合（FSI）求解器，用于模拟混凝土和钢筋混凝土（RC）结构破坏时的流固耦合现象。首先，介绍了用于模拟钢筋混凝土的LPM结构求解器的发展。在LPM框架下，提出了一种明确的回归映射算法来模拟钢筋混凝土结构的破坏。根据试验结果对所提出的LPM模型的预测能力和通用性进行了评估，该模型涵盖了RC梁中广泛的破坏模式。该模型捕捉了钢筋混凝土梁受弯和受剪破坏的关键特征。随着颗粒尺寸的细化，结构响应和裂纹模式呈现一致的结果。在验证结构求解器的基础上，应用耦合SPH-LPM方法模拟了多个实体断裂的FSI试验用例。最后，通过海啸型波浪造成的普通和钢筋混凝土墙体破坏的试验案例，证明了当前方法在处理复杂的、真实的结构破坏的FSI案例中的能力。

{"title":"Numerical modelling of reinforced concrete using lattice particle method with extension to FSI-induced structural failure via SPH coupling","authors":"W.C. Low , K.C. Ng , Kaiwei Tian , H.K. Ng","doi":"10.1016/j.compstruc.2025.108075","DOIUrl":"10.1016/j.compstruc.2025.108075","url":null,"abstract":"<div><div>In this paper, a fluid structure interaction (FSI) solver based on smoothed particle hydrodynamics (SPH) and lattice particle method (LPM) is further developed to model FSI phenomena with structural failure involving concrete and reinforced concrete (RC). First, the development of the LPM structural solver to model reinforced concrete is presented. An explicit return-mapping algorithm is proposed in the framework of LPM to model the failure of reinforced concrete structures. The predictive capability and generality of the proposed LPM model is assessed against experimental results covering wide range of failure modes seen in RC beams. Key features of the flexural and shear failure in RC beams are captured by the model. With particle size refinement, the structural response and crack pattern show consistent results. Upon verification of the structural solver, the coupled SPH-LPM method is applied to simulate several FSI test cases with solid fracture. Finally, the capability of the current method in tackling complex, real-world FSI cases with structural failure is demonstrated through the test case of failure of plain and reinforced concrete wall due to tsunami-type wave.</div></div>","PeriodicalId":50626,"journal":{"name":"Computers & Structures","volume":"321 ","pages":"Article 108075"},"PeriodicalIF":4.8,"publicationDate":"2025-12-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145796208","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