CRII: CHS: Robust Algorithms Modeling Frictional Contact with Industrial, Medical and Computer Graphics Applications

CRII：CHS：工业、医疗和计算机图形应用中摩擦接触建模的鲁棒算法

• 批准号: 1755544
• 负责人: Chenfanfu Jiang
• 金额: $ 17.5万
• 依托单位: University of Pennsylvania
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2018
• 资助国家: 美国
• 起止时间: 2018-04-01 至 2020-03-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1755544&HistoricalAwards=false
• 关键词: CRII; CHS; Robust; Algorithms; Modeling

Visual simulation methods for natural phenomena involving complex solid and fluid dynamics have been widely applied in digital animation and effects. Computer graphics and computational mechanics have also proved to be a powerful combination for pushing the boundaries of science and engineering in diverse fields such as soil mechanics, geophysics, biomechanics and engineering design. The current research aims to both alleviate existing computational bottlenecks and improve simulation resolution, with an emphasis on interacting coupled virtual materials with elaborate geometry and intricate frictional contact. Project outcomes will include algorithms with broad impact in various applications such as the analysis of geo-mechanical phenomena (landslides, debris flows, avalanches), the simulation of fully coupled human body parts for medical training (skin, muscle, organ, bone), and the design of granular material processing (in food, agriculture, mining, and the pharmaceuticals industry). The project will collaborate with researchers from these related science and engineering fields, and will advocate for interdisciplinary collaborations among undergraduate and graduate students and researchers. It will take steps to engage and attract both design and medical science students and will encourage the research careers of women and minority students through educational events, exchange programs, and other social activities. The project will focus on the development of highly robust and efficient new algorithms for modeling frictional contact under complex settings through innovative numerical discretization schemes and continuum elastoplastic models. The work will utilize the proven hybrid Lagrangian / Eulerian Material Point Method (MPM). A primary focus will be on the simulation of granular materials with nontrivial geometries in multi-physics settings. The methods will leverage spatial / temporal adaptivity, multiscale modeling, novel discretization schemes, efficient data structures, and low-level optimization exploiting hardware architecture. The techniques developed in this work will offer seamless coupling between rigid bodies, granular materials, deformable objects, and fluids; they will establish a solid foundation for a unified, versatile multi-material physics solver to handle complex phenomena in a way that requires minimal user interference.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

涉及复杂固流动力学的自然现象的可视化仿真方法已广泛应用于数字动画和特效中。计算机图形学和计算力学也被证明是推动不同领域的科学和工程边界的强大组合，如土壤力学、地球物理、生物力学和工程设计。目前的研究旨在缓解现有的计算瓶颈和提高模拟分辨率，重点是与复杂几何和复杂摩擦接触的耦合虚拟材料的相互作用。项目成果将包括在各种应用中具有广泛影响的算法，例如对地球力学现象（滑坡、泥石流、雪崩）的分析，用于医学训练的完全耦合人体部位（皮肤、肌肉、器官、骨骼）的模拟，以及颗粒材料加工（食品、农业、采矿和制药工业）的设计。该项目将与这些相关科学和工程领域的研究人员合作，并将提倡本科生和研究生与研究人员之间的跨学科合作。它将采取措施吸引和吸引设计和医学专业的学生，并将通过教育活动、交流项目和其他社会活动鼓励妇女和少数民族学生从事研究事业。该项目将重点开发高度稳健和高效的新算法，通过创新的数值离散化方案和连续弹塑性模型来模拟复杂设置下的摩擦接触。这项工作将利用经过验证的拉格朗日/欧拉混合材料点法（MPM）。一个主要的焦点将是在多物理场设置中具有非平凡几何形状的颗粒材料的模拟。这些方法将利用空间/时间适应性、多尺度建模、新颖的离散化方案、高效的数据结构和利用硬件架构的低级优化。在这项工作中开发的技术将提供刚体，颗粒材料，可变形物体和流体之间的无缝耦合；它们将为一个统一的、通用的多材料物理解算器奠定坚实的基础，以一种需要最小用户干扰的方式处理复杂现象。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

项目成果

A Temporally Adaptive Material Point Method with Regional Time Stepping

• DOI: 10.1111/cgf.13524
• 发表时间: 2018-09
• 期刊: Computer Graphics Forum
• 影响因子: 2.5
• 作者: Yu Fang;Yuanming Hu;Shimin Hu;Chenfanfu Jiang

Animating fluid sediment mixture in particle-laden flows

• DOI: 10.1145/3197517.3201309
• 发表时间: 2018-07
• 期刊: ACM Transactions on Graphics (TOG)
• 作者: Ming Gao;Andre Pradhana;Xuchen Han;Q. Guo;G. Kot;Eftychios Sifakis;Chenfanfu Jiang

Efficient and Conservative Fluids Using Bidirectional Mapping

• DOI: 10.1145/3306346.3322945
• 发表时间: 2019-07-01
• 期刊: ACM TRANSACTIONS ON GRAPHICS
• 影响因子: 6.2
• 作者: Qu, Ziyin;Zhang, Xinxin;Chen, Baoquan
• 通讯作者: Chen, Baoquan

Decomposed optimization time integrator for large-step elastodynamics

• DOI: 10.1145/3306346.3322951
• 发表时间: 2019-07
• 期刊: ACM Transactions on Graphics (TOG)
• 作者: Minchen Li;Ming Gao;Timothy R. Langlois;Chenfanfu Jiang;D. Kaufman

GPU optimization of material point methods

• DOI: 10.1145/3272127.3275044
• 发表时间: 2018-12
• 期刊: ACM Transactions on Graphics (TOG)
• 作者: Ming Gao;Xinlei Wang;Kui Wu;Andre Pradhana;Eftychios Sifakis;Cem Yuksel;Chenfanfu Jiang