Advances in computational science and engineering最新文献

{"title":"LOCAL DIVERGENCE-FREE IMMERSED FINITE ELEMENT-DIFFERENCE METHOD USING COMPOSITE B-SPLINES.","authors":"Lianxia Li, Cole Gruninger, Jae H Lee, Boyce E Griffith","doi":"10.3934/acse.2025011","DOIUrl":"https://doi.org/10.3934/acse.2025011","url":null,"abstract":"<p><p>In the class of immersed boundary (IB) methods, the choice of the regularized delta function plays a crucial role in transferring information between fluid and solid domains through interpolation and spreading operators. Most prior work using the IB method has used isotropic kernels that do not preserve the divergence-free condition of the velocity field, leading to loss of incompressibility of the solid when interpolating the Eulerian velocity to Lagrangian markers. One approach to addressing this issue in IB simulations involving large deformations of immersed incompressible elastic structures is to use a volumetric stabilization approach, such as adding a volumetric energy term and using modified invariants in the structure's constitutive model. Composite B-spline (CBS) kernels offer an alternative approach by inherently maintaining the discrete divergence-free property. This work evaluates the performance of CBS kernels in terms of their volume conservation and accuracy, comparing them with several traditional isotropic kernel functions using a construction introduced by Peskin (referred to as IB kernels) and B-spline (BS) kernels. Benchmark tests include pressure-loaded and shear-dominated flows, such as an elastic band under differential pressure loads, a pressurized membrane, a compressed block, Cook's membrane, a slanted channel flow, and a modified Turek-Hron problem. Additionally, we validate our methodology using a complex fluid-structure interaction model of bioprosthetic heart valve dynamics in a pulse duplicator. Results demonstrate that CBS kernels achieve superior volume conservation compared to conventional isotropic kernels, eliminating the need for additional volumetric stabilization techniques typically required to address instabilities arising from volume conservation errors. Further, it is common that the accuracy provided by CBS kernels on coarser grids is comparable to that provided by IB and BS kernels on finer grids. Unlike IB and BS kernels, which perform better with larger mesh ratio factors between solid and fluid grids, CBS kernels show improved results with smaller mesh ratio factors. Additionally, the study reveals that although wider kernels provide more accurate results across all methods, CBS kernels are less sensitive to variations in relative grid spacings than isotropic kernels. This study highlights the advantages of CBS kernels in achieving stable, accurate, and efficient FSI simulations without requiring specialized volumetric stabilization treatments when simulating large deformations of elastic solids immersed in fluid.</p>","PeriodicalId":520995,"journal":{"name":"Advances in computational science and engineering","volume":"4 ","pages":"16-56"},"PeriodicalIF":0.0,"publicationDate":"2025-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12365689/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144987199","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"DIRECT MEDICAL IMAGE TO SIMULATION USING AUTO-SEGMENTATION AND POINT CLOUD-BASED CFD.","authors":"Ashton M Corpuz, Monu Jaiswal, Pan Du, Abhay B Ramachandra, Jian-Xun Wang, Ming-Chen Hsu","doi":"10.3934/acse.2025006","DOIUrl":"10.3934/acse.2025006","url":null,"abstract":"<p><p>Cardiovascular disease (CVD) remains one of the leading causes of mortality worldwide. Computational medicine and digital twins hold promise in mitigating the impact and prevalence of CVD. Recent advances in image-based computational methods have enabled the quantification of functional and biologically important metrics that would otherwise be difficult to obtain from the standard of care. However, significant challenges remain due to the manual/semi-automated nature of the processes and the domain expertise required to perform them. This paper addresses these challenges by proposing a novel framework that builds on our recently developed direct point cloud-to-CFD approach using immersogeometric analysis. The proposed method leverages advanced auto-segmentation techniques to extract medically relevant geometries as point clouds, which are then directly used for CFD simulations. The framework is validated using benchmark flow problems with analytical and computational solutions and is subsequently applied to patient-specific images to demonstrate its capabilities. The results highlight the method's ability to facilitate rapid CFD simulations directly on point clouds derived from patient scans, underscoring its potential to accelerate the image-to-simulation pipeline and enable the tractability of cardiovascular digital twins.</p>","PeriodicalId":520995,"journal":{"name":"Advances in computational science and engineering","volume":"3 ","pages":"95-124"},"PeriodicalIF":0.0,"publicationDate":"2025-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12927703/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147287049","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}