Computer Physics Communications最新文献

{"title":"Deep operator networks for Bayesian parameter estimation in PDEs","authors":"Amogh Raj,&nbsp;Sakol Bun,&nbsp;Keerthana Srinivasa,&nbsp;Carol Eunice Gudumotou,&nbsp;Arash Sarshar","doi":"10.1016/j.cpc.2025.109853","DOIUrl":"10.1016/j.cpc.2025.109853","url":null,"abstract":"<div><div>We present a novel framework combining Deep Operator Networks (DeepONets) with Physics-Informed Neural Networks (PINNs) to solve partial differential equations (PDEs) while estimating their unknown parameters. By integrating data-driven learning with physical constraints, our method achieves robust and accurate solutions across diverse scenarios. Bayesian training is implemented through variational inference, allowing for comprehensive uncertainty quantification for both data and model uncertainties. This ensures reliable prediction and parameter estimates even in noisy conditions or when some of the physical equations governing the problem are missing. The framework demonstrates its efficacy in solving forward and inverse problems, including the 1D unsteady heat equation, 2D reaction-diffusion equations, 3D eigenvalue problem, and various regression tasks with sparse, noisy observations. This approach provides a computationally efficient and generalizable method for addressing uncertainty quantification in PDE surrogate modeling.</div></div>","PeriodicalId":285,"journal":{"name":"Computer Physics Communications","volume":"317 ","pages":"Article 109853"},"PeriodicalIF":3.4,"publicationDate":"2025-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145109517","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A compact gas-kinetic scheme with scalable hp multigrid acceleration for steady-state computation on 3D unstructured meshes","authors":"Hongyu Liu ,&nbsp;Xing Ji ,&nbsp;Yunpeng Mao ,&nbsp;Yuan Ding ,&nbsp;Kun Xu","doi":"10.1016/j.cpc.2025.109820","DOIUrl":"10.1016/j.cpc.2025.109820","url":null,"abstract":"<div><div>In this paper, we present an advanced high-order compact gas-kinetic scheme (CGKS) for 3D unstructured mixed-element meshes, augmented with a <em>hp</em> multigrid technique to accelerate steady-state convergence. The scheme evolves cell-averaged flow variables and their gradients on the original mesh. Mesh coarsening employs a two-step parallel agglomeration algorithm, utilizing a random hash for cell interface selection and a geometric skewness metric for deletion confirmation, thereby ensuring both efficiency and robustness. For the coarser meshes, first-order kinetic flux vector splitting (KFVS) schemes with explicit or implicit time-stepping are used. The proposed multigrid CGKS is tested across various flow regimes on hybrid unstructured meshes, demonstrating significant improvements. A three-level V-cycle multigrid strategy, coupled with an explicit forward Euler method on coarser levels, results in a convergence rate up to ten times faster than standard CGKS. In contrast, the implicit lower-upper symmetric Gauss-Seidel (LU-SGS) method offers limited convergence acceleration. Scalability tests have demonstrated that GMG-CGKS exhibits consistent performance across varying numbers of CPU cores, highlighting its outstanding scalability. Our findings indicate that the explicit multigrid CGKS is highly scalable and effective for large-scale computations, marking a substantial step forward in computational fluid dynamics.</div></div><div><h3>Program summary</h3><div><em>Program title:</em> GMG-CGKS-v0.1</div><div><em>CPC Library link to program files:</em> <span><span>https://doi.org/10.17632/t97mh5c78g.1</span><svg><path></path></svg></span></div><div><em>Developer's repository link:</em> <span><span>https://github.com/kevinhongyu/GMG-CGKS-v0.1.git</span><svg><path></path></svg></span></div><div><em>Programming language:</em> C++</div><div><em>Licensing provisions:</em> GPLv2</div><div><em>External Libraries:</em> METIS, MPI, HDF5</div><div><em>Nature of problem:</em> The program is designed to solve the compressible Euler and Navier-Stokes equations, which are widely used in aerodynamics. The program provides steady-state acceleration techniques for third-order CGKS.</div><div><em>Solution method:</em> A three-level geometric multigrid scheme is adopted to improve the convergence rate of CGKS.</div></div>","PeriodicalId":285,"journal":{"name":"Computer Physics Communications","volume":"317 ","pages":"Article 109820"},"PeriodicalIF":3.4,"publicationDate":"2025-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145095832","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Nyström type exponential integrators for strongly magnetized charged particle dynamics","authors":"Tri P. Nguyen ,&nbsp;Ilon Joseph ,&nbsp;Mayya Tokman","doi":"10.1016/j.cpc.2025.109848","DOIUrl":"10.1016/j.cpc.2025.109848","url":null,"abstract":"<div><div>Solving for charged particle motion in electromagnetic fields (i.e. the particle pushing problem) is a computationally intensive component of particle-in-cell (PIC) methods for plasma physics simulations. This task is especially challenging when the plasma is strongly magnetized due numerical stiffness arising from the wide range of time scales between highly oscillatory gyromotion and long term macroscopic behavior. A promising approach to solve these problems is by a class of methods known as exponential integrators that can solve linear problems exactly and are A-stable. This work extends the standard exponential integration framework to derive Nyström-type exponential integrators that integrates the Newtonian equations of motion as a second-order differential equation directly. In particular, we derive second-order and third-order Nyström-type exponential integrators for strongly magnetized particle pushing problems. Numerical experiments show that the Nyström-type exponential integrators exhibit significant improvement in computation speed over the standard exponential integrators.</div></div>","PeriodicalId":285,"journal":{"name":"Computer Physics Communications","volume":"317 ","pages":"Article 109848"},"PeriodicalIF":3.4,"publicationDate":"2025-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145095935","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Systematic development of an equivalent particle method for efficient simulation of dense granular flows","authors":"Yilong Liu ,&nbsp;Xiping Yu","doi":"10.1016/j.cpc.2025.109862","DOIUrl":"10.1016/j.cpc.2025.109862","url":null,"abstract":"<div><div>Development of a highly efficient model is very important to expand the applicability of discrete element method (DEM) to large-scale granular flows that often include a tremendous number of granular particles. An equivalent particle method is rigorously developed for such a purpose in this study. The kinetic theory for granular flows is taken advantage to understand the relationship between the original particle system and the equivalent particle system, with a focus on conservation of mass and momentum. With the newly established equivalent particle method, the averaged particle velocity, density and volume concentration remain the same as in the original system. Scaling factors for other physical quantities, particularly those describing particle contact processes, are introduced to satisfy the geometric, kinematic and dynamic similarities. Verification of the equivalent particle model are performed by applying it to the computation of granular collapses on both horizontal and inclined bottoms. The computational results on deformation of granular profiles show that existing coarse grain or representative particle models, which were developed for the similar purpose as the present equivalent particle model, underestimate the granular material’s mobility. The numerical results from the present model agree much better with experimental data, indicating a major advancement in this kind of model development. The efficiency is drastically improved by tremendously reducing the number of computed particles, as compared to the standard DEM model.</div></div>","PeriodicalId":285,"journal":{"name":"Computer Physics Communications","volume":"317 ","pages":"Article 109862"},"PeriodicalIF":3.4,"publicationDate":"2025-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145095937","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Ray-tracing laser-deposition model for plasma particle-in-cell simulation","authors":"A. Hyder ,&nbsp;W. Fox ,&nbsp;K.V. Lezhnin ,&nbsp;S.R. Totorica","doi":"10.1016/j.cpc.2025.109847","DOIUrl":"10.1016/j.cpc.2025.109847","url":null,"abstract":"<div><div>We develop a ray-tracing model for laser-plasma interaction suitable for coupling in-line into kinetic particle-in-cell plasma simulations. The model is based on inverse Bremsstrahlung absorption and includes oblique incidence effects and reflection at the critical surface. The energy deposition is given to electrons by randomized kicks to momentum. The model is verified against analytic solutions and a 2-D laser ray-tracing code.</div></div>","PeriodicalId":285,"journal":{"name":"Computer Physics Communications","volume":"318 ","pages":"Article 109847"},"PeriodicalIF":3.4,"publicationDate":"2025-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145156858","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Speeding up fermionic lattice calculations with photonic accelerated inverters","authors":"Felipe Attanasio ,&nbsp;Marc Bauer ,&nbsp;Jelle Dijkstra ,&nbsp;Timoteo Lee ,&nbsp;Jan M. Pawlowski ,&nbsp;Wolfram Pernice ,&nbsp;Frank Brückerhoff-Plückelmann","doi":"10.1016/j.cpc.2025.109825","DOIUrl":"10.1016/j.cpc.2025.109825","url":null,"abstract":"<div><div>Lattice field theory (LFT) is the standard non-perturbative method to perform numerical calculations of quantum field theory. However, the typical bottleneck of fermionic lattice calculations is the inversion of the Dirac matrix. This inversion is solved by iterative methods, like the conjugate gradient algorithm, where matrix-vector multiplications (MVMs) are the main operation. Photonic integrated circuits excel in performing quick and energy-efficient MVMs, but at the same time, they are known to have low accuracy. This can be overcome by using mixed precision methods. In this paper, we explore the idea of using photonic technology to fulfil the demand for computational power of fermionic lattice calculations. These methods have the potential to reduce computation costs by one order of magnitude. Because of the hybrid nature of these methods, we call these ‘photonic accelerated inverters (PAIs)’.</div></div>","PeriodicalId":285,"journal":{"name":"Computer Physics Communications","volume":"317 ","pages":"Article 109825"},"PeriodicalIF":3.4,"publicationDate":"2025-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145095933","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An advanced fully-implicit solver for heterogeneous porous media based on foam-extend","authors":"Roberto Lange,&nbsp;Gabriel M. Magalhães,&nbsp;Franciane F. Rocha,&nbsp;Hélio Ribeiro Neto","doi":"10.1016/j.cpc.2025.109842","DOIUrl":"10.1016/j.cpc.2025.109842","url":null,"abstract":"&lt;div&gt;&lt;div&gt;Multiphase flow in porous media is present in many engineering applications, including hydrogeology, oil recovery, and CO&lt;sub&gt;2&lt;/sub&gt; sequestration. Accurate predictions of fluid behavior in these systems can improve process efficiency while mitigating environmental and health risks. Commercial simulators and open source software, such as the &lt;span&gt;porousMultiphaseFoam&lt;/span&gt; repository based on the OpenFOAM framework, have been developed to model this type of problem. However, simulating heterogeneous porous media with heterogeneous porosity and permeability distributions poses significant numerical challenges. We introduce &lt;span&gt;coupledMatrixFoam&lt;/span&gt;, an OpenFOAM-based solver designed for enhanced numerical stability and robustness. &lt;span&gt;coupledMatrixFoam&lt;/span&gt; integrates the Eulerian multi-fluid formulation for phase fractions with Darcy's law for porous media flow, applying a fully implicit, block-coupled solution for pressure and phase fractions. The solver is based on foam-extend 5.0, leveraging the latest &lt;span&gt;fvBlockMatrix&lt;/span&gt; developments to improve computational efficiency. This approach enables a significant increase in time step sizes, particularly in cases involving capillary pressure effects and other complex physical interactions. This work details the formulation, implementation and validation of &lt;span&gt;coupledMatrixFoam&lt;/span&gt;, including comparisons with &lt;span&gt;porousMultiphaseFoam&lt;/span&gt; that uses a segregated approach, to assess performance improvements. Additionally, a scalability analysis is conducted, demonstrating the solver's ability for high-performance computing (HPC) applications, which are essential for large-scale, real-world simulations.&lt;/div&gt;&lt;/div&gt;&lt;div&gt;&lt;h3&gt;Program summary&lt;/h3&gt;&lt;div&gt;&lt;em&gt;Program Title:&lt;/em&gt; coupledMatrixFoam&lt;/div&gt;&lt;div&gt;&lt;em&gt;CPC Library link to program files:&lt;/em&gt; &lt;span&gt;&lt;span&gt;https://doi.org/10.17632/3d3xdh4x89.1&lt;/span&gt;&lt;svg&gt;&lt;path&gt;&lt;/path&gt;&lt;/svg&gt;&lt;/span&gt;&lt;/div&gt;&lt;div&gt;&lt;em&gt;Developer's repository link:&lt;/em&gt; &lt;span&gt;&lt;span&gt;https://gitlab.com/wikki.brasil/porousmedia&lt;/span&gt;&lt;svg&gt;&lt;path&gt;&lt;/path&gt;&lt;/svg&gt;&lt;/span&gt;&lt;/div&gt;&lt;div&gt;&lt;em&gt;Licensing provisions:&lt;/em&gt; GPLv3&lt;/div&gt;&lt;div&gt;&lt;em&gt;Programming language:&lt;/em&gt; C++&lt;/div&gt;&lt;div&gt;&lt;em&gt;Supplementary material:&lt;/em&gt; Available in the repository &lt;span&gt;&lt;span&gt;porousMedia&lt;/span&gt;&lt;svg&gt;&lt;path&gt;&lt;/path&gt;&lt;/svg&gt;&lt;/span&gt;.&lt;/div&gt;&lt;div&gt;&lt;em&gt;Nature of problem:&lt;/em&gt; This software solves multiphase flow in porous media.&lt;/div&gt;&lt;div&gt;&lt;em&gt;Solution method:&lt;/em&gt; Fully implicit solver based on the Euler-Euler multifluid formulation combined with Darcy's law developed in the OpenFOAM framework, that is based on the Finite Volume Method (FVM). Implicit coupling of phase fraction and pressure equations allowing significantly larger time steps. Complex physical phenomena are accounted for, including capillary pressure effects, gravitational forces, compressibility, and heterogeneous media properties. The porous medium is modeled as a stationary phase, and nonlinear terms are linearized using Taylor series expansions. P","PeriodicalId":285,"journal":{"name":"Computer Physics Communications","volume":"317 ","pages":"Article 109842"},"PeriodicalIF":3.4,"publicationDate":"2025-09-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145095830","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Fast automatically differentiable matrix functions and applications in molecular simulations","authors":"Tina Torabi ,&nbsp;Timon S. Gutleb ,&nbsp;Christoph Ortner","doi":"10.1016/j.cpc.2025.109832","DOIUrl":"10.1016/j.cpc.2025.109832","url":null,"abstract":"<div><div>We describe efficient differentiation methods for computing Jacobians and gradients of a large class of matrix functions including the matrix logarithm <span><math><mi>log</mi><mo>⁡</mo><mo>(</mo><mi>A</mi><mo>)</mo></math></span> and <em>p</em>-th roots <span><math><msup><mrow><mi>A</mi></mrow><mrow><mfrac><mrow><mn>1</mn></mrow><mrow><mi>p</mi></mrow></mfrac></mrow></msup></math></span>. We exploit contour integrals and conformal maps as described by Hale et al. (2008) <span><span>[34]</span></span> for evaluation and differentiation and analyze the computational complexity as well as numerical accuracy compared to high accuracy finite difference methods. As a demonstrator application we compute properties of structural defects in silicon crystals at positive temperatures, requiring efficient and accurate gradients of matrix trace-logarithms.</div></div>","PeriodicalId":285,"journal":{"name":"Computer Physics Communications","volume":"317 ","pages":"Article 109832"},"PeriodicalIF":3.4,"publicationDate":"2025-09-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145026728","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Scaling up to multivariate rational function reconstruction","authors":"Andreas Maier","doi":"10.1016/j.cpc.2025.109827","DOIUrl":"10.1016/j.cpc.2025.109827","url":null,"abstract":"<div><div>I present an algorithm for the reconstruction of multivariate rational functions from black-box probes. The arguably most important application in high-energy physics is the calculation of multi-loop and multi-leg amplitudes, where rational functions appear as coefficients in the integration-by-parts reduction to basis integrals. I show that for a dense coefficient the algorithm is nearly optimal, in the sense that the number of required probes is close to the number of unknowns.</div></div>","PeriodicalId":285,"journal":{"name":"Computer Physics Communications","volume":"317 ","pages":"Article 109827"},"PeriodicalIF":3.4,"publicationDate":"2025-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145019629","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"GPU-based compressible lattice Boltzmann simulations on non-uniform grids using standard C++ parallelism: From best practices to aerodynamics, aeroacoustics and supersonic flow simulations","authors":"Christophe Coreixas ,&nbsp;Jonas Latt","doi":"10.1016/j.cpc.2025.109833","DOIUrl":"10.1016/j.cpc.2025.109833","url":null,"abstract":"<div><div>Despite decades of research, creating accurate, robust, and efficient lattice Boltzmann methods (LBM) on non-uniform grids with seamless GPU acceleration remains challenging. This work introduces a novel strategy to address this challenge by integrating simple yet effective components: (1) parallel algorithms in modern C++, (2) conservative cell-centered grid refinement, (3) local boundary conditions, and (4) robust collision models. Our framework supports multiple lattices (D2Q9, D2Q13, D2Q21, D2Q37, D3Q27, etc) tailored to various flow conditions. It includes collision models with polynomial and numerical equilibria, a second distribution for polyatomic behavior, a Jameson-like shock sensor, and generalizes Rohde's refinement strategy.</div><div>The framework's accuracy and robustness is validated across diverse benchmarks, including lid-driven cavity flows, Aeolian noise, 30P30N airfoil aerodynamics, inviscid Riemann problems, and viscous flows past a NACA airfoil in transonic and supersonic regimes. Modern C++ further enables our framework to reach GPU-native performance, while ensuring high portability, modularity, and ease of implementation. Notably, weakly compressible LBMs achieve state-of-the-art GPU efficiency on non-uniform grids, while fully compressible LBMs benefit from acceleration equivalent to thousands of CPU cores in the most compute-intensive cases. Our advanced performance models incorporate neighbor-list and asynchronous time-stepping effects, providing new insights into the performance decomposition of LB simulations on non-uniform grids.</div><div>Overall, this study sets a new standard for portable, tree-based LBMs, demonstrating that a combination of well-chosen components can achieve high performance, accuracy, and robustness across various flow conditions. As a final proof-of-concept, adaptive mesh refinement is proposed for subsonic and supersonic applications.</div></div>","PeriodicalId":285,"journal":{"name":"Computer Physics Communications","volume":"317 ","pages":"Article 109833"},"PeriodicalIF":3.4,"publicationDate":"2025-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145026729","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}