Computers & Structures最新文献

Energy-based methods for solving forward and inverse linear elasticity problems in 2D structures 基于能量的二维结构正、逆线性弹性问题求解方法

IF 4.4 2区工程技术

Computers & Structures Pub Date : 2025-07-25 DOI: 10.1016/j.compstruc.2025.107899

Manish Thombre , Cosmin Anitescu , BVSS Bharadwaja , Yizheng Wang , Timon Rabczuk , Alankar Alankar

{"title":"Energy-based methods for solving forward and inverse linear elasticity problems in 2D structures","authors":"Manish Thombre , Cosmin Anitescu , BVSS Bharadwaja , Yizheng Wang , Timon Rabczuk , Alankar Alankar","doi":"10.1016/j.compstruc.2025.107899","DOIUrl":"10.1016/j.compstruc.2025.107899","url":null,"abstract":"<div><div>Physics Informed Neural Networks (PINNs) and the Deep Energy Method (DEM) are two recently developed approaches for solving partial differential equations (PDEs) using deep neural networks. While PINNs aim to minimize the residual of the strong form of PDEs, DEM solvers work by minimizing the total potential energy. However, these methods have limitations in capturing the complex characteristics of displacement and stress fields. To overcome these limitations, we propose two new extensions of DEM: the deep energy method with traction-free boundary loss term (t-DEM) and the energy minimization method (EMM) with finite element method (FEM) basis. The t-DEM includes an additional loss term to enforce traction-free boundary conditions, and the EMM combines the FEM basis with the DEM to efficiently minimize the total potential energy of the system.</div><div>In our paper, we apply these techniques to solve linear elasticity forward and inverse problems and compare their performance. We demonstrate the effectiveness of these methods through various elasticity standard problems and compare the results with true solutions. In addition, we conducted a parametric study to analyze the impact of different parameter variations on the proposed frameworks. The numerical results highlight the accuracy and reliability of the proposed approaches for the forward and inverse linear elasticity problems, underlining their potential, particularly for the solution of stress concentration problems.</div></div>","PeriodicalId":50626,"journal":{"name":"Computers & Structures","volume":"316 ","pages":"Article 107899"},"PeriodicalIF":4.4,"publicationDate":"2025-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144703441","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

Form-finding design method for mesh reflector antennas considering the geometric nonlinearity of the truss 考虑桁架几何非线性的网状反射面天线寻形设计方法

IF 4.4 2区工程技术

Computers & Structures Pub Date : 2025-07-25 DOI: 10.1016/j.compstruc.2025.107903

Shunji Zhang

{"title":"Form-finding design method for mesh reflector antennas considering the geometric nonlinearity of the truss","authors":"Shunji Zhang","doi":"10.1016/j.compstruc.2025.107903","DOIUrl":"10.1016/j.compstruc.2025.107903","url":null,"abstract":"<div><div>Form-finding design serves as the crucial step towards attaining the desired electromagnetic performance of antennas, while laying the groundwork for subsequent design phases. To achieve an efficient and high-precision design of mesh antennas, a form-finding method considering the geometric nonlinearity of the truss has been proposed. In this approach, an investigation into the coupling relationship between the cable net and the truss is conducted, followed by a discussion on the deformation mechanism stemming from the geometric nonlinearity of the truss. Subsequently, a system equilibrium equation is established for the holistic design of mesh antennas, and an incremental equation, namely the tangent stiffness matrix, is further derived to enhance the iterative convergence rate and precision of the antenna’s transition from its initial state to the desired equilibrium state. On this basis, the sensitivity matrix of the antenna nodes regarding the force density vector is derived, and the functional relationship between the form-finding design objective and force density increment is established. The form-finding process is transformed into an easily solvable sequential quadratic programming function. Finally, the design accuracy and efficiency of the proposed method are verified through numerical examples.</div></div>","PeriodicalId":50626,"journal":{"name":"Computers & Structures","volume":"316 ","pages":"Article 107903"},"PeriodicalIF":4.4,"publicationDate":"2025-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144702878","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 nonlinear dynamics-informed LSTM network for response prediction of strong earthquake-excited high-rise building structures 强震作用下高层建筑结构响应预测的非线性动态LSTM网络

IF 4.4 2区工程技术

Computers & Structures Pub Date : 2025-07-24 DOI: 10.1016/j.compstruc.2025.107902

Zheng He , Jie Yang , Wenfeng Fan , Dianyou Yu

{"title":"A nonlinear dynamics-informed LSTM network for response prediction of strong earthquake-excited high-rise building structures","authors":"Zheng He , Jie Yang , Wenfeng Fan , Dianyou Yu","doi":"10.1016/j.compstruc.2025.107902","DOIUrl":"10.1016/j.compstruc.2025.107902","url":null,"abstract":"<div><div>Long short-term memory (LSTM) networks, which have attracted increasing attention, have emerged as a promising approach for predicting structural dynamic responses, particularly including earthquake-induced time–frequency characteristics. However, due to the multi-modal effects and diverse plasticity development paths in high-rise structures under strong earthquakes, existing LSTM networks fail to explicitly and simultaneously capture such complex dynamic behaviors from a structural dynamics perspective. To address this challenge, this work develops DYNLSTM-Tall, a novel nonlinear dynamics-informed LSTM network, based on two-level mapping relationships derived from the numerical substructure method and a comprehensive time–frequency domain evaluation metric. The architecture of DYNLSTM-Tall and the combined loss function are optimized through ablation experiments across five extended datasets. On this basis, the prediction accuracy and generalization ability of DYNLSTM-Tall is demonstrated by training/validation loss convergence, quantitative evaluation metrics and comparison between predicted vs. actual earthquake response histories of test samples with varying seismic damage states. DYNLSTM-Tall’s superior performance is further validated through comparison with two state-of-the-art LSTM variants across three case high-rise structures, achieving correlation coefficients of 0.95 ∼ 0.98 and accurately identifying up to six lower vibration modes. This research underscores its strong potential as an alternative for seismic risk assessment as well as structural earthquake response prediction.</div></div>","PeriodicalId":50626,"journal":{"name":"Computers & Structures","volume":"316 ","pages":"Article 107902"},"PeriodicalIF":4.4,"publicationDate":"2025-07-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144696372","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 space–time adaptive solution scheme for phase field fracture model 相场断裂模型的时空自适应求解方案

IF 4.4 2区工程技术

Computers & Structures Pub Date : 2025-07-22 DOI: 10.1016/j.compstruc.2025.107904

Yuanfeng Yu , Chi Hou

{"title":"A space–time adaptive solution scheme for phase field fracture model","authors":"Yuanfeng Yu , Chi Hou","doi":"10.1016/j.compstruc.2025.107904","DOIUrl":"10.1016/j.compstruc.2025.107904","url":null,"abstract":"<div><div>To accurately capture the crack evolution, a finer mesh is required in the region of crack propagation, resulting in a higher computational cost for the phase field model. In order to improve the computational efficiency, a space–time adaptive solution scheme is proposed in this work. An adaptive remeshing technique is used to adaptively refine the mesh in the space field, the mesh size characteristics are coupled with the phase field variable to control the time incremental changes in the time field. In this scheme, a coarse grid is first used to obtain the initial crack paths and element energy distribution, and the mesh is automatically refined based on the element energy distribution. Second, the time field alteration criterion is presented by coupling the change of the mesh size in the spatial field with the change of the phase field variable. The time increment is adaptively varied based on the changes of mesh size and phase field value in the solution region to reduce the calculated burden. The proposed scheme is validated by some examples. Whether it is a single-mode failure or a mixed-mode failure, the adaptive solution scheme can significantly reduce the number of elements and the computation time. At the same time, complex crack trajectories can be effectively simulated and accurate load–displacement curves obtained, which makes it possible to use the phase field model to the failure analysis of complex structures in practical engineering.</div></div>","PeriodicalId":50626,"journal":{"name":"Computers & Structures","volume":"316 ","pages":"Article 107904"},"PeriodicalIF":4.4,"publicationDate":"2025-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144679582","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

Combining non-linear structural modeling and machine learning for tensile behavior of multi-layered randomized architected material 结合非线性结构建模和机器学习研究多层随机结构材料的拉伸行为

IF 4.4 2区工程技术

Computers & Structures Pub Date : 2025-07-22 DOI: 10.1016/j.compstruc.2025.107896

Sagnik Paul, Ann C. Sychterz

{"title":"Combining non-linear structural modeling and machine learning for tensile behavior of multi-layered randomized architected material","authors":"Sagnik Paul, Ann C. Sychterz","doi":"10.1016/j.compstruc.2025.107896","DOIUrl":"10.1016/j.compstruc.2025.107896","url":null,"abstract":"<div><div>Architected materials allow engineers to design and develop materials with desired properties through the interplay of geometry and provide an opportunity to investigate the behavior of micro-scale structures at the macro-scale. Multi-layered Randomized Architected Material (MLRAM), inspired by polymeric structures, has the potential to act as damage detection indicators in highly redundant structures such as tensegrity structures by incorporating them into tension members of a structure. The behavior is analyzed based on the parameters, coordination number and percentage density of Long Links. This paper presents a computational model that captures the tensile behavior of this proposed architected material using a non-linear finite element method. A framework is created that captures the progressive failure of the links of MLRAM. The variation in the tensile properties is analyzed with respect to its parameters. Results show that the variation of stiffness and peak tensile capacity decreases with an increase in coordination number and percentage density of Long Links. Machine learning algorithms and artificial neural networks are evaluated to propose models that can predict the tensile properties of architected materials given their geometrical parameters.</div></div>","PeriodicalId":50626,"journal":{"name":"Computers & Structures","volume":"316 ","pages":"Article 107896"},"PeriodicalIF":4.4,"publicationDate":"2025-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144679581","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 ALE-CR formulation for nonlinear dynamics of 2D variable-length beam with unprescribed moving boundaries 非规定移动边界的二维变长梁非线性动力学的ALE-CR公式

IF 4.4 2区工程技术

Computers & Structures Pub Date : 2025-07-21 DOI: 10.1016/j.compstruc.2025.107900

Shihao Xu , Zhuyong Liu , Boyang Wang , Tingke Wu

{"title":"An ALE-CR formulation for nonlinear dynamics of 2D variable-length beam with unprescribed moving boundaries","authors":"Shihao Xu , Zhuyong Liu , Boyang Wang , Tingke Wu","doi":"10.1016/j.compstruc.2025.107900","DOIUrl":"10.1016/j.compstruc.2025.107900","url":null,"abstract":"<div><div>Conventional corotational (CR) beam formulations predominantly employ the Lagrangian description, which is inefficient in addressing moving boundaries. This study proposes a two-dimensional (2D) corotational formulation based on the arbitrary Lagrangian–Eulerian (ALE) description for the nonlinear dynamic analysis of beams with both prescribed and unprescribed moving boundaries. In the corotational framework, the length of the beam element is integrated into the local variables while the material coordinates of the nodes are incorporated into the global variables to describe the unprescribed material flow. Consequently, moving boundaries can be described accurately through moving nodes. An intermediate configuration is introduced to separate the material flow from the Lagrangian motion. The virtual work principle is used to derive the motion equations of the beam. The internal and inertial terms of the element, including those associated with material coordinates, are derived explicitly so that Gaussian quadrature is not required. The configurational force on the unprescribed moving boundary is effectively taken into account, which significantly influences the beam’s axial sliding behavior. The accuracy and efficiency are validated through examples involving moving loads, moving masses and moving supports. It shows the proposed ALE-CR beam formulation is capable of handling both prescribed and unprescribed moving boundaries.</div></div>","PeriodicalId":50626,"journal":{"name":"Computers & Structures","volume":"316 ","pages":"Article 107900"},"PeriodicalIF":4.4,"publicationDate":"2025-07-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144679580","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 method for dynamic responses, modal and buckling analysis of cylindrical periodic structures based on the group theory and Woodbury formula 基于群理论和Woodbury公式的圆柱周期结构动力响应、模态和屈曲分析的有效方法

IF 4.4 2区工程技术

Computers & Structures Pub Date : 2025-07-21 DOI: 10.1016/j.compstruc.2025.107897

Long Chen, Jiangyu Hong, Dongdong Xie, Qiang Gao

{"title":"Efficient method for dynamic responses, modal and buckling analysis of cylindrical periodic structures based on the group theory and Woodbury formula","authors":"Long Chen, Jiangyu Hong, Dongdong Xie, Qiang Gao","doi":"10.1016/j.compstruc.2025.107897","DOIUrl":"10.1016/j.compstruc.2025.107897","url":null,"abstract":"<div><div>This paper proposes an efficient method that combines group theory, the Woodbury formula, and condensation technology to solve the governing equations related to dynamic responses, modal and buckling analysis of cylindrical periodic structures. Efficient and accurate solution of linear algebraic equations is pivotal for dynamic responses, modal and buckling analysis. Based on the cyclic periodic property of the structures and group theory, the linear algebraic equations can be transformed into a series of independent subequations corresponding to one-dimensional (1D) periodic structures. The degrees of freedom (DOFs) of the unit cell corresponding to the 1D periodic structures can be divided into two boundary-surface DOFs and internal DOFs. To reduce the scale of the 1D periodic structures, we use condensation technique to condense the internal DOFs into the two boundary-surface DOFs. Thus, the scales of these subequations can be significantly reduced. Since the coefficient matrices of the small-scale subequations resemble block-circulant matrices, the Woodbury formula is used to obtain their solutions by solving some newly formed linear algebraic equations whose coefficient matrices are block-circulant. These newly formed linear algebraic equations are then efficiently solved using group theory. Numerical examples are presented to confirm the high accuracy and efficiency of the proposed method.</div></div>","PeriodicalId":50626,"journal":{"name":"Computers & Structures","volume":"316 ","pages":"Article 107897"},"PeriodicalIF":4.4,"publicationDate":"2025-07-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144670435","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

Enhancing displacement accuracy of stable node-based smoothed finite element method (SNS-FEM) with a smoothing framework based on weighted least squares (WLS) 基于加权最小二乘的光滑框架提高稳定节点光滑有限元法（SNS-FEM）位移精度

IF 4.4 2区工程技术

Computers & Structures Pub Date : 2025-07-21 DOI: 10.1016/j.compstruc.2025.107893

Samadrita Karmakar, Roland Traxl, Roman Lackner

{"title":"Enhancing displacement accuracy of stable node-based smoothed finite element method (SNS-FEM) with a smoothing framework based on weighted least squares (WLS)","authors":"Samadrita Karmakar, Roland Traxl, Roman Lackner","doi":"10.1016/j.compstruc.2025.107893","DOIUrl":"10.1016/j.compstruc.2025.107893","url":null,"abstract":"<div><div>The Node-based Smoothed Finite Element Method (NS-FEM) mitigates strain and volumetric locking in the standard Finite Element Method (FEM) by employing averaged nodal strains and nodal integration to construct the stiffness matrix. However, nodal integration may fail to constrain certain distortion modes, resulting in numerical instability. Stable NS-FEM (SNS-FEM) introduces temporary integration points that constrain these spurious modes, restoring stability. This paper proposes a new formulation that enhances the displacement accuracy of SNS-FEM, particularly on coarse meshes. Instead of relying on local strain integration and averaging, the method computes nodal strains using a Weighted Least Squares (WLS) approach. By leveraging the flexibility of WLS, it strategically selects strain sampling locations and polynomial fitting orders to modify the computed nodal strain gradients. Compared to SNS-FEM, this modification leads to a lower stiffness contribution from the stabilisation term, thereby improving displacement accuracy. The formulation is evaluated across four types of test cases, focusing primarily on displacement fields and natural frequency modes. Results demonstrate that the proposed method preserves dynamic stability while improving displacement accuracy relative to SNS-FEM—especially on coarse meshes—at comparable computational costs.</div></div>","PeriodicalId":50626,"journal":{"name":"Computers & Structures","volume":"316 ","pages":"Article 107893"},"PeriodicalIF":4.4,"publicationDate":"2025-07-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144670390","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

Influence of the manufacturing process on the stress state of curved glulam beams 制造工艺对弯曲胶合木梁应力状态的影响

IF 4.4 2区工程技术

Computers & Structures Pub Date : 2025-07-18 DOI: 10.1016/j.compstruc.2025.107890

Taoyi Yu , Ani Khaloian , Jan-Willem van de Kuilen

{"title":"Influence of the manufacturing process on the stress state of curved glulam beams","authors":"Taoyi Yu , Ani Khaloian , Jan-Willem van de Kuilen","doi":"10.1016/j.compstruc.2025.107890","DOIUrl":"10.1016/j.compstruc.2025.107890","url":null,"abstract":"<div><div>This study examines the spring-back effect and residual stresses in curved glued-laminated timber (glulam) beams during the manufacturing process. In addition to curving along the longitudinal axis (<span><math><mi>X</mi></math></span>), a cup-deformation develops in the transverse direction (<span><math><mi>Z</mi></math></span>) due to the Poisson’s effect. This deformation, combined with the glue-hardening process and the release of pressure, leads to the development of residual stresses in all three axial directions, as well as shear stresses within individual boards and at their interfaces.</div><div>Besides the well-known factors such as longitudinal elasticity, board thickness, and inner radius, the study reveals that the number of layers (<span><math><mi>n</mi></math></span>) and Poisson’s ratio (<span><math><msub><mrow><mi>ν</mi></mrow><mrow><mi>L</mi><mi>T</mi></mrow></msub></math></span>) significantly influence the magnitude of residual stresses. However, aside from longitudinal stresses, the impact of <span><math><mi>n</mi></math></span> and <span><math><msub><mrow><mi>ν</mi></mrow><mrow><mi>L</mi><mi>T</mi></mrow></msub></math></span>, as well as other residual stresses, have been scarcely studied and are not adequately addressed in current design standards.</div><div>A Monte–Carlo analysis of the growth-ring effect is conducted, taking the pith location of different board layers as the input random variable. Strong influences can be identified on the residual stresses in both radial and tangential directions, with intensified maximum values and more scattered distribution inside the cross-section. The time- and moisture-dependent relaxation analysis using the rheological wood model shows a significant influence of the temperature and relative humidity.</div></div>","PeriodicalId":50626,"journal":{"name":"Computers & Structures","volume":"316 ","pages":"Article 107890"},"PeriodicalIF":4.4,"publicationDate":"2025-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144656984","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

Prediction of physical space parameters for dynamical systems: A POD-ANN approach 动态系统物理空间参数的预测：POD-ANN方法

IF 4.4 2区工程技术

Computers & Structures Pub Date : 2025-07-18 DOI: 10.1016/j.compstruc.2025.107891

Samrul Hoda, Biswarup Bhattacharyya

{"title":"Prediction of physical space parameters for dynamical systems: A POD-ANN approach","authors":"Samrul Hoda, Biswarup Bhattacharyya","doi":"10.1016/j.compstruc.2025.107891","DOIUrl":"10.1016/j.compstruc.2025.107891","url":null,"abstract":"<div><div>Full-Order Models (FOM) parameters provide detailed representations of dynamic systems. However, they often come with high computational costs. This is especially true for large-scale or real-time applications such as structural health monitoring and digital twinning. Reduced-Order Modeling (ROM) techniques, like Proper Orthogonal Decomposition (POD), address this by reducing system dimensionality. This enables more efficient simulations while preserving essential dynamics. However, ROM approaches cannot be used directly to predict parameters (e.g., stiffness) in the physical space. These parameters are crucial for comprehensive system evaluation. This study introduces a novel framework combining POD and Artificial Neural Networks (ANN) to map reduced-order parameters to physical parameters for the case of gradual, global parameter variations. POD extracts significant modes that capture essential dynamics, projecting full-order dynamics onto the reduced space. Subsequently, ANN is trained to map reduced space to physical space parameters, addressing ROM’s limitations. The POD-ANN model is applied to linear and nonlinear dynamic problems, and <span><math><mn>50</mn><mspace></mspace><mi>%</mi></math></span> reduction in modes compared to the full-order dimension is observed in both cases. Furthermore, the achieved relative error is quite low while predicting the full-order stiffness matrix for linear and nonlinear dynamical systems. Once the model is trained, it can be used to predict parameters efficiently.</div></div>","PeriodicalId":50626,"journal":{"name":"Computers & Structures","volume":"316 ","pages":"Article 107891"},"PeriodicalIF":4.4,"publicationDate":"2025-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144656985","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