首页 > 最新文献

Engineering with Computers最新文献

英文 中文
A multiscale stabilized physics informed neural networks with weakly imposed boundary conditions transfer learning method for modeling advection dominated flow 用于平流主导流建模的多尺度稳定物理信息神经网络与弱边界条件迁移学习法
IF 8.7 2区 工程技术 Q1 Mathematics Pub Date : 2024-05-06 DOI: 10.1007/s00366-024-01981-5
Tsung-Yeh Hsieh, Tsung-Hui Huang

Physics informed neural network (PINN) frameworks have been developed as a powerful technique for solving partial differential equations (PDEs) with potential data integration. However, when applied to advection based PDEs, PINNs confront challenges such as parameter sensitivity in boundary condition enforcement and diminished learning capability due to an ill-conditioned system resulting from the strong advection. In this study, we present a multiscale stabilized PINN formulation with a weakly imposed boundary condition (WBC) method coupled with transfer learning that can robustly model the advection diffusion equation. To address key challenges, we use an advection-flux-decoupling technique to prescribe the Dirichlet boundary conditions, which rectifies the imbalanced training observed in PINN with conventional penalty and strong enforcement methods. A multiscale approach under the least squares functional form of PINN is developed that introduces a controllable stabilization term, which can be regarded as a special form of Sobolev training that augments the learning capacity. The efficacy of the proposed method is demonstrated through the resolution of a series of benchmark problems of forward modeling, and the outcomes affirm the potency of the methodology proposed.

物理信息神经网络(PINN)框架已被开发为一种强大的技术,用于解决具有潜在数据整合能力的偏微分方程(PDE)。然而,当应用于基于平流的偏微分方程时,PINNs 面临着各种挑战,如边界条件执行中的参数敏感性,以及强平流导致的条件不良系统造成的学习能力减弱。在本研究中,我们提出了一种多尺度稳定 PINN 方案,该方案采用弱施加边界条件(WBC)方法,并结合迁移学习,可以对平流扩散方程进行稳健建模。为了应对关键挑战,我们使用了一种平流-通量-解耦技术来规定 Dirichlet 边界条件,从而纠正了在 PINN 中使用传统惩罚和强执行方法所观察到的不平衡训练。我们在 PINN 的最小二乘函数形式下开发了一种多尺度方法,引入了一个可控的稳定项,它可以被视为一种特殊形式的 Sobolev 训练,可以增强学习能力。通过解决一系列前向建模的基准问题,展示了所提方法的功效,结果肯定了所提方法的有效性。
{"title":"A multiscale stabilized physics informed neural networks with weakly imposed boundary conditions transfer learning method for modeling advection dominated flow","authors":"Tsung-Yeh Hsieh, Tsung-Hui Huang","doi":"10.1007/s00366-024-01981-5","DOIUrl":"https://doi.org/10.1007/s00366-024-01981-5","url":null,"abstract":"<p>Physics informed neural network (PINN) frameworks have been developed as a powerful technique for solving partial differential equations (PDEs) with potential data integration. However, when applied to advection based PDEs, PINNs confront challenges such as parameter sensitivity in boundary condition enforcement and diminished learning capability due to an ill-conditioned system resulting from the strong advection. In this study, we present a multiscale stabilized PINN formulation with a weakly imposed boundary condition (WBC) method coupled with transfer learning that can robustly model the advection diffusion equation. To address key challenges, we use an advection-flux-decoupling technique to prescribe the Dirichlet boundary conditions, which rectifies the imbalanced training observed in PINN with conventional penalty and strong enforcement methods. A multiscale approach under the least squares functional form of PINN is developed that introduces a controllable stabilization term, which can be regarded as a special form of Sobolev training that augments the learning capacity. The efficacy of the proposed method is demonstrated through the resolution of a series of benchmark problems of forward modeling, and the outcomes affirm the potency of the methodology proposed.</p>","PeriodicalId":11696,"journal":{"name":"Engineering with Computers","volume":"17 1","pages":""},"PeriodicalIF":8.7,"publicationDate":"2024-05-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140887984","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
Fast parametric analysis of trimmed multi-patch isogeometric Kirchhoff-Love shells using a local reduced basis method 使用局部还原基方法对修剪多补丁等几何基尔霍夫-洛夫壳进行快速参数分析
IF 8.7 2区 工程技术 Q1 Mathematics Pub Date : 2024-04-29 DOI: 10.1007/s00366-024-01980-6
Margarita Chasapi, Pablo Antolin, Annalisa Buffa

This contribution presents a model order reduction framework for real-time efficient solution of trimmed, multi-patch isogeometric Kirchhoff-Love shells. In several scenarios, such as design and shape optimization, multiple simulations need to be performed for a given set of physical or geometrical parameters. This step can be computationally expensive in particular for real world, practical applications. We are interested in geometrical parameters and take advantage of the flexibility of splines in representing complex geometries. In this case, the operators are geometry-dependent and generally depend on the parameters in a non-affine way. Moreover, the solutions obtained from trimmed domains may vary highly with respect to different values of the parameters. Therefore, we employ a local reduced basis method based on clustering techniques and the Discrete Empirical Interpolation Method to construct affine approximations and efficient reduced order models. In addition, we discuss the application of the reduction strategy to parametric shape optimization. Finally, we demonstrate the performance of the proposed framework to parameterized Kirchhoff-Love shells through benchmark tests on trimmed, multi-patch meshes including a complex geometry. The proposed approach is accurate and achieves a significant reduction of the online computational cost in comparison to the standard reduced basis method.

本文提出了一个模型阶次缩减框架,用于实时高效地解决修剪、多补丁等几何基尔霍夫-洛夫壳。在设计和形状优化等多种情况下,需要对给定的物理或几何参数集进行多次模拟。这一步骤的计算成本很高,尤其是在现实世界的实际应用中。我们对几何参数感兴趣,并利用花键的灵活性来表示复杂的几何形状。在这种情况下,算子与几何相关,通常以非曲线的方式依赖于参数。此外,从修剪域得到的解可能会因参数值的不同而有很大差异。因此,我们采用基于聚类技术和离散经验插值法的局部还原基础方法来构建仿射近似和高效的还原阶模型。此外,我们还讨论了还原策略在参数形状优化中的应用。最后,我们通过对包括复杂几何体在内的修剪过的多补丁网格进行基准测试,证明了所提出的框架在参数化 Kirchhoff-Love 壳体方面的性能。与标准的还原基方法相比,所提出的方法非常精确,并显著降低了在线计算成本。
{"title":"Fast parametric analysis of trimmed multi-patch isogeometric Kirchhoff-Love shells using a local reduced basis method","authors":"Margarita Chasapi, Pablo Antolin, Annalisa Buffa","doi":"10.1007/s00366-024-01980-6","DOIUrl":"https://doi.org/10.1007/s00366-024-01980-6","url":null,"abstract":"<p>This contribution presents a model order reduction framework for real-time efficient solution of trimmed, multi-patch isogeometric Kirchhoff-Love shells. In several scenarios, such as design and shape optimization, multiple simulations need to be performed for a given set of physical or geometrical parameters. This step can be computationally expensive in particular for real world, practical applications. We are interested in geometrical parameters and take advantage of the flexibility of splines in representing complex geometries. In this case, the operators are geometry-dependent and generally depend on the parameters in a non-affine way. Moreover, the solutions obtained from trimmed domains may vary highly with respect to different values of the parameters. Therefore, we employ a local reduced basis method based on clustering techniques and the Discrete Empirical Interpolation Method to construct affine approximations and efficient reduced order models. In addition, we discuss the application of the reduction strategy to parametric shape optimization. Finally, we demonstrate the performance of the proposed framework to parameterized Kirchhoff-Love shells through benchmark tests on trimmed, multi-patch meshes including a complex geometry. The proposed approach is accurate and achieves a significant reduction of the online computational cost in comparison to the standard reduced basis method.</p>","PeriodicalId":11696,"journal":{"name":"Engineering with Computers","volume":"75 1","pages":""},"PeriodicalIF":8.7,"publicationDate":"2024-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140811184","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
The Trefftz methods for 3D biharmonic equation using directors and in-plane biharmonic functions 使用导演和平面内双谐函数的三维双谐方程的特雷弗茨方法
IF 8.7 2区 工程技术 Q1 Mathematics Pub Date : 2024-04-29 DOI: 10.1007/s00366-024-01977-1
Chein-Shan Liu, Chung-Lun Kuo

Because the complete set of Trefftz functions for the 3D biharmonic equation is not yet well established, a multiple-direction Trefftz method (MDTM) and an in-plane biharmonic functions method (IPBFM) are deduced in the paper. Inspired by the Trefftz method for the 2D biharmonic equation, a novel MDTM incorporates planar directors into the 2D like Trefftz functions to solve the 3D biharmonic equation. These functions being a series of biharmonic polynomials of different degree, automatically satisfying the 3D biharmonic equation, are taken as the bases to expand the solution. Then, we derive a quite large class solution of the 3D biharmonic equation in terms of 3D harmonic functions, and 2D biharmonic functions in three sub-planes. The 2D biharmonic functions are formulated as the Trefftz functions in terms of the polar coordinates for each sub-plane. Introducing a projective variable, we can obtain the projective type general solution for the 3D Laplace equation, which is used to generate the 3D Trefftz type harmonic functions. Several numerical examples confirm the efficiency and accuracy of the proposed MDTM and IPBFM.

由于三维双谐波方程的全套特雷弗兹函数尚未完全建立,本文推导出一种多方向特雷弗兹方法(MDTM)和一种平面内双谐波函数方法(IPBFM)。受用于二维双谐波方程的 Trefftz 方法的启发,一种新型 MDTM 将平面导向融入类似 Trefftz 函数的二维双谐波方程中,以求解三维双谐波方程。这些函数是一系列不同度数的双谐波多项式,自动满足三维双谐波方程,并以此为基础展开求解。然后,我们用三维谐函数和三个子平面上的二维谐函数推导出了三维双谐方程的大类解。二维双谐函数是以每个子平面的极坐标表示的 Trefftz 函数。通过引入一个投影变量,我们可以得到三维拉普拉斯方程的投影型通解,并用它来生成三维特雷弗兹型谐函数。几个数值示例证实了所提出的 MDTM 和 IPBFM 的效率和准确性。
{"title":"The Trefftz methods for 3D biharmonic equation using directors and in-plane biharmonic functions","authors":"Chein-Shan Liu, Chung-Lun Kuo","doi":"10.1007/s00366-024-01977-1","DOIUrl":"https://doi.org/10.1007/s00366-024-01977-1","url":null,"abstract":"<p>Because the complete set of Trefftz functions for the 3D biharmonic equation is not yet well established, a multiple-direction Trefftz method (MDTM) and an in-plane biharmonic functions method (IPBFM) are deduced in the paper. Inspired by the Trefftz method for the 2D biharmonic equation, a novel MDTM incorporates planar directors into the 2D like Trefftz functions to solve the 3D biharmonic equation. These functions being a series of biharmonic polynomials of different degree, automatically satisfying the 3D biharmonic equation, are taken as the bases to expand the solution. Then, we derive a quite large class solution of the 3D biharmonic equation in terms of 3D harmonic functions, and 2D biharmonic functions in three sub-planes. The 2D biharmonic functions are formulated as the Trefftz functions in terms of the polar coordinates for each sub-plane. Introducing a projective variable, we can obtain the projective type general solution for the 3D Laplace equation, which is used to generate the 3D Trefftz type harmonic functions. Several numerical examples confirm the efficiency and accuracy of the proposed MDTM and IPBFM.</p>","PeriodicalId":11696,"journal":{"name":"Engineering with Computers","volume":"1 1","pages":""},"PeriodicalIF":8.7,"publicationDate":"2024-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140811303","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
Unstructured mesh tools for magnetically confined fusion system simulations 用于磁约束聚变系统模拟的非结构网格工具
IF 8.7 2区 工程技术 Q1 Mathematics Pub Date : 2024-04-28 DOI: 10.1007/s00366-024-01976-2
Mark S. Shephard, Jacob Merson, Onkar Sahni, Angel E. Castillo, Aditya Y. Joshi, Dhyanjyoti D. Nath, Usman Riaz, E. Seegyoung Seol, Cameron W. Smith, Chonglin Zhang, Mark W. Beall, Ottmar Klaas, Rocco Nastasia, Saurabh Tendulkar

As fusion simulation codes increasingly account for the full geometric complexity of magnetically confined fusion systems, a need arose to provide tailored unstructured mesh technologies to address the specific needs of fusion plasma simulation codes and their coupling to other physics simulation codes. This paper presents a high level overview of a set of unstructured mesh developments that include; methods to effectively employ manufacturing CAD models in the construction of quality analysis model geometries, specialized mesh generation and adaptation tools; an infrastructure to support parallel particle-in-cell calculation on unstructured meshes; and an infrastructure for coupling of massively parallel mesh-based fusion codes.

随着核聚变模拟代码越来越多地考虑到磁约束核聚变系统的全部几何复杂性,需要提供量身定制的非结构化网格技术,以满足核聚变等离子体模拟代码及其与其他物理模拟代码耦合的特定需求。本文对一系列非结构化网格开发进行了高度概述,其中包括:在构建质量分析模型几何图形时有效使用制造 CAD 模型的方法、专业网格生成和适配工具、支持非结构化网格上并行粒子入胞计算的基础架构,以及用于耦合基于网格的大规模并行聚变代码的基础架构。
{"title":"Unstructured mesh tools for magnetically confined fusion system simulations","authors":"Mark S. Shephard, Jacob Merson, Onkar Sahni, Angel E. Castillo, Aditya Y. Joshi, Dhyanjyoti D. Nath, Usman Riaz, E. Seegyoung Seol, Cameron W. Smith, Chonglin Zhang, Mark W. Beall, Ottmar Klaas, Rocco Nastasia, Saurabh Tendulkar","doi":"10.1007/s00366-024-01976-2","DOIUrl":"https://doi.org/10.1007/s00366-024-01976-2","url":null,"abstract":"<p>As fusion simulation codes increasingly account for the full geometric complexity of magnetically confined fusion systems, a need arose to provide tailored unstructured mesh technologies to address the specific needs of fusion plasma simulation codes and their coupling to other physics simulation codes. This paper presents a high level overview of a set of unstructured mesh developments that include; methods to effectively employ manufacturing CAD models in the construction of quality analysis model geometries, specialized mesh generation and adaptation tools; an infrastructure to support parallel particle-in-cell calculation on unstructured meshes; and an infrastructure for coupling of massively parallel mesh-based fusion codes.</p>","PeriodicalId":11696,"journal":{"name":"Engineering with Computers","volume":"6 1","pages":""},"PeriodicalIF":8.7,"publicationDate":"2024-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140811005","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 novel peridynamics refinement method with dual-horizon peridynamics 采用双水平周流体力学的新型周流体力学细化方法
IF 8.7 2区 工程技术 Q1 Mathematics Pub Date : 2024-04-28 DOI: 10.1007/s00366-024-01983-3
Zhixin Zeng, Xiong Zhang

A novel adaptive refinement method is proposed for ordinary state-based PD in both 2D and 3D simulation. We perform an efficient adaptive refinement by splitting a parent particle into several child particles directly which does not require any information from its adjacent particles. The state variables and neighbor list of the child particles can be obtained directly from their parent particles, so that the proposed adaptive refinement method is very efficient and can be easily applied in both 2D and 3D cases without complex adjacent particle list building. To maintain simulation accuracy, the deformation of the parent particles need to be taken into consideration in the refinement process. Therefore, the accumulated strain of a PD particle is defined to calculate the position vector of its child particles in current configuration. With the consideration of the deformation of the parent particle, the fake damage is avoid. And a new criterion is established based on the particle accumulated strain to efficiently determine the particles need to be refined. The numerical examples studied show that the proposed adaptive refinement correctly captures the complicated crack propagation process and eliminates the fake crack growth with slight extra simulation cost, compared to the coarse discretization.

针对二维和三维模拟中基于状态的普通 PD,我们提出了一种新颖的自适应细化方法。我们通过将一个父粒子直接拆分成多个子粒子来执行高效的自适应细化,而无需从其相邻粒子中获取任何信息。子粒子的状态变量和相邻粒子列表可直接从父粒子中获取,因此所提出的自适应细化方法非常高效,可轻松应用于二维和三维情况,而无需建立复杂的相邻粒子列表。为了保持仿真精度,在细化过程中需要考虑父粒子的变形。因此,需要定义一个 PD 粒子的累积应变,以计算其子粒子在当前配置中的位置矢量。考虑到父粒子的变形,可以避免假损伤。并根据粒子累积应变建立了一个新的标准,以有效地确定需要细化的粒子。研究的数值示例表明,与粗离散化相比,所提出的自适应细化方法能正确捕捉复杂的裂纹扩展过程,并消除假裂纹的增长,但仿真成本略有增加。
{"title":"A novel peridynamics refinement method with dual-horizon peridynamics","authors":"Zhixin Zeng, Xiong Zhang","doi":"10.1007/s00366-024-01983-3","DOIUrl":"https://doi.org/10.1007/s00366-024-01983-3","url":null,"abstract":"<p>A novel adaptive refinement method is proposed for ordinary state-based PD in both 2D and 3D simulation. We perform an efficient adaptive refinement by splitting a parent particle into several child particles directly which does not require any information from its adjacent particles. The state variables and neighbor list of the child particles can be obtained directly from their parent particles, so that the proposed adaptive refinement method is very efficient and can be easily applied in both 2D and 3D cases without complex adjacent particle list building. To maintain simulation accuracy, the deformation of the parent particles need to be taken into consideration in the refinement process. Therefore, the accumulated strain of a PD particle is defined to calculate the position vector of its child particles in current configuration. With the consideration of the deformation of the parent particle, the fake damage is avoid. And a new criterion is established based on the particle accumulated strain to efficiently determine the particles need to be refined. The numerical examples studied show that the proposed adaptive refinement correctly captures the complicated crack propagation process and eliminates the fake crack growth with slight extra simulation cost, compared to the coarse discretization.</p>","PeriodicalId":11696,"journal":{"name":"Engineering with Computers","volume":"16 1","pages":""},"PeriodicalIF":8.7,"publicationDate":"2024-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140810872","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
Comprehensive polygonal topology optimization for triplet thermo-mechanical-pressure multi-material systems 三重热-机-压多材料系统的综合多边形拓扑优化
IF 8.7 2区 工程技术 Q1 Mathematics Pub Date : 2024-04-27 DOI: 10.1007/s00366-024-01982-4
Thanh T. Banh, Dongkyu Lee

Current advancements in topology optimization research have extensively explored thermoelastic problems. Yet, notable limitations persist in effectively handling design-dependent fluidic pressure loads, particularly in the realm of coupled thermo-mechanical systems. To bridge this gap, this study proposes a novel and consistent methodology that comprehensively accommodates these challenges. The principal contributions of this research are threefold: (1) presenting an innovative and comprehensive solution for triplet thermo-mechanical-pressure problems, achieved through the establishment of a specific pressure field using Darcy’s law and a drainage term, (2) broadening the scope to incorporate flexible polygonal meshes within generalized Solid Isotropic Material with Penalization (SIMP)-based multi-material systems, and (3) introducing an alternative interpolated model related to independent material properties, specifically the general thermal stress coefficient, to simplify the complexity during sensitivity calculations of thermal-strain load in generalized SIMP-based multi-material problems. Additionally, within the scope of this study, several investigations into penalty parameters in solids, not unified in previous coupled thermo-mechanical multi-material problems, are also conducted. Three additional adjoint vectors are introduced using the adjoint variable technique for sensitivity analysis to enhance computational efficiency in the gradient-based mathematical programming algorithm. The effectiveness and reliability of this method are validated through numerical examples, demonstrating its efficiency, robustness, and practicality.

目前,拓扑优化研究的进展已经广泛地探索了热弹性问题。然而,在有效处理与设计相关的流体压力负荷方面,尤其是在热机械耦合系统领域,仍然存在明显的局限性。为了弥补这一不足,本研究提出了一种新颖、一致的方法,以全面应对这些挑战。本研究的主要贡献有三方面:(1) 通过使用达西定律和排水项建立特定压力场,为三重热力-机械-压力问题提出了创新而全面的解决方案;(2) 拓宽了范围,将灵活的多边形网格纳入基于泛化固体各向同性材料与惩罚(SIMP)的多材料系统、(3) 引入与独立材料属性(特别是一般热应力系数)相关的替代插值模型,以简化基于广义 SIMP 的多材料问题中热应变载荷灵敏度计算的复杂性。此外,在本研究范围内,还对固体中的一些惩罚参数进行了研究,这些参数在以往的多材料热机械耦合问题中并不统一。在基于梯度的数学编程算法中,利用用于灵敏度分析的辅助变量技术引入了三个额外的辅助向量,以提高计算效率。通过数值实例验证了该方法的有效性和可靠性,证明了其高效性、鲁棒性和实用性。
{"title":"Comprehensive polygonal topology optimization for triplet thermo-mechanical-pressure multi-material systems","authors":"Thanh T. Banh, Dongkyu Lee","doi":"10.1007/s00366-024-01982-4","DOIUrl":"https://doi.org/10.1007/s00366-024-01982-4","url":null,"abstract":"<p>Current advancements in topology optimization research have extensively explored thermoelastic problems. Yet, notable limitations persist in effectively handling design-dependent fluidic pressure loads, particularly in the realm of coupled thermo-mechanical systems. To bridge this gap, this study proposes a novel and consistent methodology that comprehensively accommodates these challenges. The principal contributions of this research are threefold: (1) presenting an innovative and comprehensive solution for triplet thermo-mechanical-pressure problems, achieved through the establishment of a specific pressure field using Darcy’s law and a drainage term, (2) broadening the scope to incorporate flexible polygonal meshes within generalized Solid Isotropic Material with Penalization (SIMP)-based multi-material systems, and (3) introducing an alternative interpolated model related to independent material properties, specifically the general thermal stress coefficient, to simplify the complexity during sensitivity calculations of thermal-strain load in generalized SIMP-based multi-material problems. Additionally, within the scope of this study, several investigations into penalty parameters in solids, not unified in previous coupled thermo-mechanical multi-material problems, are also conducted. Three additional adjoint vectors are introduced using the adjoint variable technique for sensitivity analysis to enhance computational efficiency in the gradient-based mathematical programming algorithm. The effectiveness and reliability of this method are validated through numerical examples, demonstrating its efficiency, robustness, and practicality.</p>","PeriodicalId":11696,"journal":{"name":"Engineering with Computers","volume":"1 1","pages":""},"PeriodicalIF":8.7,"publicationDate":"2024-04-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140811004","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 novel normalized reduced-order physics-informed neural network for solving inverse problems 用于解决逆问题的新型归一化降阶物理信息神经网络
IF 8.7 2区 工程技术 Q1 Mathematics Pub Date : 2024-04-20 DOI: 10.1007/s00366-024-01971-7
Khang A. Luong, Thang Le-Duc, Seunghye Lee, Jaehong Lee

The utilization of Physics-informed Neural Networks (PINNs) in deciphering inverse problems has gained significant attention in recent years. However, the PINN training process for inverse problems is notably restricted due to gradient failures provoked by magnitudes of partial differential equations (PDEs) parameters or source functions. To address these matters, normalized reduced-order physics-informed neural network (nr-PINN) is developed in this study. The goal of the nr-PINN is to reconfigure the original PDE into a system of normalized lower-order PDEs through two sequential steps. To start with, self-homeomorphisms of the PDEs are implemented via scaling factors determined based on measured data. Afterward, each normalized PDE is transformed into a system of lower-order PDEs by primary and secondary variables. Besides, a technique to exactly impose many types of boundary conditions (BCs) by redefining NNs outputs is developed in the context of reduced-order method. The advantages of the nr-PINN model over the original one regarding solution accuracy and training cost are demonstrated through several inverse problems in solid mechanics with different types of PDEs and BCs.

近年来,利用物理信息神经网络(PINNs)破解逆问题的研究受到了广泛关注。然而,由于偏微分方程(PDE)参数或源函数的大小导致梯度失效,逆问题的 PINN 训练过程明显受到限制。为了解决这些问题,本研究开发了归一化降阶物理信息神经网络(nr-PINN)。nr-PINN 的目标是通过两个连续步骤将原始 PDE 重构为归一化低阶 PDE 系统。首先,根据测量数据确定的缩放因子实现 PDE 的自同构。然后,通过主变量和次变量将每个归一化 PDE 转化为低阶 PDE 系统。此外,在降阶方法的背景下,还开发了一种通过重新定义 NNs 输出来精确施加多种类型边界条件 (BC) 的技术。nr-PINN 模型与原始模型相比,在求解精度和训练成本方面的优势通过几个具有不同类型 PDE 和 BC 的固体力学逆问题得到了证明。
{"title":"A novel normalized reduced-order physics-informed neural network for solving inverse problems","authors":"Khang A. Luong, Thang Le-Duc, Seunghye Lee, Jaehong Lee","doi":"10.1007/s00366-024-01971-7","DOIUrl":"https://doi.org/10.1007/s00366-024-01971-7","url":null,"abstract":"<p>The utilization of Physics-informed Neural Networks (PINNs) in deciphering inverse problems has gained significant attention in recent years. However, the PINN training process for inverse problems is notably restricted due to gradient failures provoked by magnitudes of partial differential equations (PDEs) parameters or source functions. To address these matters, normalized reduced-order physics-informed neural network (nr-PINN) is developed in this study. The goal of the nr-PINN is to reconfigure the original PDE into a system of normalized lower-order PDEs through two sequential steps. To start with, self-homeomorphisms of the PDEs are implemented via scaling factors determined based on measured data. Afterward, each normalized PDE is transformed into a system of lower-order PDEs by primary and secondary variables. Besides, a technique to exactly impose many types of boundary conditions (BCs) by redefining NNs outputs is developed in the context of reduced-order method. The advantages of the nr-PINN model over the original one regarding solution accuracy and training cost are demonstrated through several inverse problems in solid mechanics with different types of PDEs and BCs.</p>","PeriodicalId":11696,"journal":{"name":"Engineering with Computers","volume":"25 1","pages":""},"PeriodicalIF":8.7,"publicationDate":"2024-04-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140625917","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
Element-free Galerkin analysis of MHD duct flow problems at arbitrary and high Hartmann numbers 对任意和高哈特曼数下的 MHD 管道流动问题进行无元素 Galerkin 分析
IF 8.7 2区 工程技术 Q1 Mathematics Pub Date : 2024-04-15 DOI: 10.1007/s00366-024-01969-1
Xiaolin Li, Shuling Li

A stabilized element-free Galerkin (EFG) method is proposed in this paper for numerical analysis of the generalized steady MHD duct flow problems at arbitrary and high Hartmann numbers up to (10^{16}). Computational formulas of the EFG method for MHD duct flows are derived by using Nitsche’s technique to facilitate the implementation of Dirichlet boundary conditions. The reproducing kernel gradient smoothing integration technique is incorporated into the EFG method to accelerate the solution procedure impaired by Gauss quadrature rules. A stabilized Nitsche-type EFG weak formulation of MHD duct flows is devised to enhance the performance damaged by high Hartmann numbers. Several benchmark MHD duct flow problems are solved to testify the stability and the accuracy of the present EFG method. Numerical results show that the range of the Hartmann number Ha in the present EFG method is (1le Hale 10^{16}), which is much larger than that in existing numerical methods.

本文提出了一种稳定的无元素伽勒金(EFG)方法,用于数值分析任意高哈特曼数(10^{16})下的广义稳定 MHD 管道流问题。通过使用 Nitsche 技术推导出了 MHD 管道流 EFG 方法的计算公式,从而方便了 Dirichlet 边界条件的实施。在 EFG 方法中加入了再现核梯度平滑积分技术,以加速受高斯正交规则影响的求解过程。设计了 MHD 管道流的稳定 Nitsche 型 EFG 弱公式,以提高受高哈特曼数影响的性能。解决了几个基准 MHD 管道流问题,以证明本 EFG 方法的稳定性和准确性。数值结果表明,本EFG方法的哈特曼数Ha范围为(1le Hale 10^{16}),远大于现有数值方法。
{"title":"Element-free Galerkin analysis of MHD duct flow problems at arbitrary and high Hartmann numbers","authors":"Xiaolin Li, Shuling Li","doi":"10.1007/s00366-024-01969-1","DOIUrl":"https://doi.org/10.1007/s00366-024-01969-1","url":null,"abstract":"<p>A stabilized element-free Galerkin (EFG) method is proposed in this paper for numerical analysis of the generalized steady MHD duct flow problems at arbitrary and high Hartmann numbers up to <span>(10^{16})</span>. Computational formulas of the EFG method for MHD duct flows are derived by using Nitsche’s technique to facilitate the implementation of Dirichlet boundary conditions. The reproducing kernel gradient smoothing integration technique is incorporated into the EFG method to accelerate the solution procedure impaired by Gauss quadrature rules. A stabilized Nitsche-type EFG weak formulation of MHD duct flows is devised to enhance the performance damaged by high Hartmann numbers. Several benchmark MHD duct flow problems are solved to testify the stability and the accuracy of the present EFG method. Numerical results show that the range of the Hartmann number <i>Ha</i> in the present EFG method is <span>(1le Hale 10^{16})</span>, which is much larger than that in existing numerical methods.</p>","PeriodicalId":11696,"journal":{"name":"Engineering with Computers","volume":"52 1","pages":""},"PeriodicalIF":8.7,"publicationDate":"2024-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140582417","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
Cut-PFEM: a Particle Finite Element Method using unfitted boundary meshes Cut-PFEM:使用非拟合边界网格的粒子有限元方法
IF 8.7 2区 工程技术 Q1 Mathematics Pub Date : 2024-04-12 DOI: 10.1007/s00366-024-01956-6
Rubén Zorrilla, Alessandro Franci

In this work, we present a novel unfitted mesh boundary strategy in the context of the Particle Finite Flement Method (PFEM) aiming to improve endemic limitations of the PFEM relative to boundary conditions treatment and mass conservation. In this new methodology, which we called Cut-PFEM, the fluid–wall interaction is not performed by adding interface elements, as is done in the standard PFEM boundaries. Instead, we use an implicit representation of (all or some of) the boundaries by introducing the use of a level set function. Such distance function detects the elements trespassing the (virtual) contours of the domain to equip them with opportunely boundary conditions, which are variationally enforced using Nitsche’s method. The proposed Cut-PFEM circumvents important issues associated with the standard PFEM contact detection algorithm, such as the artificial addition of mass to the computational domain and the anticipation of contact time. Furthermore, the Cut-PFEM represents a natural ground for the imposition of alternative wall boundary conditions (e.g., pure slip) which pose significant difficulties in a standard PFEM framework. Several numerical examples, featuring both no-slip and slip boundary conditions, are presented to prove the accuracy and robustness of the method in two-dimensional and three-dimensional scenarios.

在这项工作中,我们在粒子有限元法(PFEM)的背景下提出了一种新的非拟合网格边界策略,旨在改善粒子有限元法在边界条件处理和质量守恒方面的局限性。我们将这种新方法称为 "切割-PFEM",它不像标准 PFEM 边界那样通过添加界面元素来实现流体与壁面的相互作用。相反,我们通过引入使用水平集函数来隐式表示(全部或部分)边界。这种距离函数可以检测到侵入域(虚拟)轮廓的元素,从而为它们配备合适的边界条件,这些边界条件通过尼采方法可变地执行。所提出的剪切-PFEM 避开了与标准 PFEM 接触检测算法相关的重要问题,如人为增加计算域质量和预计接触时间。此外,Cut-PFEM 是施加替代壁边界条件(如纯滑移)的自然基础,而这些条件在标准 PFEM 框架中会造成很大困难。本文介绍了几个以无滑移和滑移边界条件为特征的数值示例,以证明该方法在二维和三维场景中的准确性和稳健性。
{"title":"Cut-PFEM: a Particle Finite Element Method using unfitted boundary meshes","authors":"Rubén Zorrilla, Alessandro Franci","doi":"10.1007/s00366-024-01956-6","DOIUrl":"https://doi.org/10.1007/s00366-024-01956-6","url":null,"abstract":"<p>In this work, we present a novel unfitted mesh boundary strategy in the context of the Particle Finite Flement Method (PFEM) aiming to improve endemic limitations of the PFEM relative to boundary conditions treatment and mass conservation. In this new methodology, which we called Cut-PFEM, the fluid–wall interaction is not performed by adding interface elements, as is done in the standard PFEM boundaries. Instead, we use an implicit representation of (all or some of) the boundaries by introducing the use of a level set function. Such distance function detects the elements trespassing the (virtual) contours of the domain to equip them with opportunely boundary conditions, which are variationally enforced using Nitsche’s method. The proposed Cut-PFEM circumvents important issues associated with the standard PFEM contact detection algorithm, such as the artificial addition of mass to the computational domain and the anticipation of contact time. Furthermore, the Cut-PFEM represents a natural ground for the imposition of alternative wall boundary conditions (<i>e.g.</i>, pure slip) which pose significant difficulties in a standard PFEM framework. Several numerical examples, featuring both no-slip and slip boundary conditions, are presented to prove the accuracy and robustness of the method in two-dimensional and three-dimensional scenarios.</p>","PeriodicalId":11696,"journal":{"name":"Engineering with Computers","volume":"51 1","pages":""},"PeriodicalIF":8.7,"publicationDate":"2024-04-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140582470","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
Brain-inspired spiking neural networks in Engineering Mechanics: a new physics-based self-learning framework for sustainable Finite Element analysis 工程力学中的脑启发尖峰神经网络:基于物理学的可持续有限元分析自学新框架
IF 8.7 2区 工程技术 Q1 Mathematics Pub Date : 2024-04-12 DOI: 10.1007/s00366-024-01967-3
Saurabh Balkrishna Tandale, Marcus Stoffel

The present study aims to develop a sustainable framework employing brain-inspired neural networks for solving boundary value problems in Engineering Mechanics. Spiking neural networks, known as the third generation of artificial neural networks, are proposed for physics-based artificial intelligence. Accompanied by a new pseudo-explicit integration scheme based on spiking recurrent neural networks leading to a spike-based pseudo explicit integration scheme, the underlying differential equations are solved with a physics-informed strategy. We propose additionally a third-generation spike-based Legendre Memory Unit that handles large sequences. These third-generation networks can be implemented on the coming-of-age neuromorphic hardware resulting in less energy and memory consumption. The proposed framework, although implicit, is viewed as a pseudo-explicit scheme since it requires almost no or fewer online training steps to achieve a converged solution even for unseen loading sequences. The proposed framework is deployed in a Finite Element solver for plate structures undergoing cyclic loading and a Xylo-Av2 SynSense neuromorphic chip is used to assess its energy performance. An acceleration of more than 40% when compared to classical Finite Element Method simulations and the capability of online training is observed. We also see a reduction in energy consumption down to the thousandth order.

本研究旨在开发一个可持续的框架,利用脑启发神经网络解决工程力学中的边界值问题。尖峰神经网络被称为第三代人工神经网络,是为基于物理的人工智能而提出的。在尖峰递归神经网络的基础上,我们提出了一种新的基于尖峰的伪显式积分方案,以物理学为基础的策略求解底层微分方程。此外,我们还提出了可处理大型序列的第三代基于尖峰的 Legendre 存储单元。这些第三代网络可以在即将问世的神经形态硬件上实现,从而降低能耗和内存消耗。所提出的框架虽然是隐式的,但被视为一种伪显式方案,因为它几乎不需要或只需要较少的在线训练步骤,就能获得收敛的解决方案,即使对于未见过的加载序列也是如此。该框架被部署在一个有限元求解器中,用于对承受循环加载的板结构进行求解,并使用 Xylo-Av2 SynSense 神经形态芯片来评估其能量性能。与传统的有限元法模拟相比,该方法的速度提高了 40% 以上,并具备了在线训练的能力。我们还发现能耗降低到了千分之一。
{"title":"Brain-inspired spiking neural networks in Engineering Mechanics: a new physics-based self-learning framework for sustainable Finite Element analysis","authors":"Saurabh Balkrishna Tandale, Marcus Stoffel","doi":"10.1007/s00366-024-01967-3","DOIUrl":"https://doi.org/10.1007/s00366-024-01967-3","url":null,"abstract":"<p>The present study aims to develop a sustainable framework employing brain-inspired neural networks for solving boundary value problems in Engineering Mechanics. Spiking neural networks, known as the third generation of artificial neural networks, are proposed for physics-based artificial intelligence. Accompanied by a new pseudo-explicit integration scheme based on spiking recurrent neural networks leading to a spike-based pseudo explicit integration scheme, the underlying differential equations are solved with a physics-informed strategy. We propose additionally a third-generation spike-based Legendre Memory Unit that handles large sequences. These third-generation networks can be implemented on the coming-of-age neuromorphic hardware resulting in less energy and memory consumption. The proposed framework, although implicit, is viewed as a pseudo-explicit scheme since it requires almost no or fewer online training steps to achieve a converged solution even for unseen loading sequences. The proposed framework is deployed in a Finite Element solver for plate structures undergoing cyclic loading and a Xylo-Av2 SynSense neuromorphic chip is used to assess its energy performance. An acceleration of more than 40% when compared to classical Finite Element Method simulations and the capability of online training is observed. We also see a reduction in energy consumption down to the thousandth order.</p>","PeriodicalId":11696,"journal":{"name":"Engineering with Computers","volume":"11 1","pages":""},"PeriodicalIF":8.7,"publicationDate":"2024-04-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140582648","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
期刊
Engineering with Computers
全部 Acc. Chem. Res. ACS Applied Bio Materials ACS Appl. Electron. Mater. ACS Appl. Energy Mater. ACS Appl. Mater. Interfaces ACS Appl. Nano Mater. ACS Appl. Polym. Mater. ACS BIOMATER-SCI ENG ACS Catal. ACS Cent. Sci. ACS Chem. Biol. ACS Chemical Health & Safety ACS Chem. Neurosci. ACS Comb. Sci. ACS Earth Space Chem. ACS Energy Lett. ACS Infect. Dis. ACS Macro Lett. ACS Mater. Lett. ACS Med. Chem. Lett. ACS Nano ACS Omega ACS Photonics ACS Sens. ACS Sustainable Chem. Eng. ACS Synth. Biol. Anal. Chem. BIOCHEMISTRY-US Bioconjugate Chem. BIOMACROMOLECULES Chem. Res. Toxicol. Chem. Rev. Chem. Mater. CRYST GROWTH DES ENERG FUEL Environ. Sci. Technol. Environ. Sci. Technol. Lett. Eur. J. Inorg. Chem. IND ENG CHEM RES Inorg. Chem. J. Agric. Food. Chem. J. Chem. Eng. Data J. Chem. Educ. J. Chem. Inf. Model. J. Chem. Theory Comput. J. Med. Chem. J. Nat. Prod. J PROTEOME RES J. Am. Chem. Soc. LANGMUIR MACROMOLECULES Mol. Pharmaceutics Nano Lett. Org. Lett. ORG PROCESS RES DEV ORGANOMETALLICS J. Org. Chem. J. Phys. Chem. J. Phys. Chem. A J. Phys. Chem. B J. Phys. Chem. C J. Phys. Chem. Lett. Analyst Anal. Methods Biomater. Sci. Catal. Sci. Technol. Chem. Commun. Chem. Soc. Rev. CHEM EDUC RES PRACT CRYSTENGCOMM Dalton Trans. Energy Environ. Sci. ENVIRON SCI-NANO ENVIRON SCI-PROC IMP ENVIRON SCI-WAT RES Faraday Discuss. Food Funct. Green Chem. Inorg. Chem. Front. Integr. Biol. J. Anal. At. Spectrom. J. Mater. Chem. A J. Mater. Chem. B J. Mater. Chem. C Lab Chip Mater. Chem. Front. Mater. Horiz. MEDCHEMCOMM Metallomics Mol. Biosyst. Mol. Syst. Des. Eng. Nanoscale Nanoscale Horiz. Nat. Prod. Rep. New J. Chem. Org. Biomol. Chem. Org. Chem. Front. PHOTOCH PHOTOBIO SCI PCCP Polym. Chem.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1