Bridging Microstructural and Continuum Theories of DispersedSystems with Parallel Computer Architectures

将分散系统的微观结构和连续体理论与并行计算机体系结构联系起来

• 批准号: 9218668
• 负责人: Sangtae Kim
• 金额: $ 25.75万
• 依托单位: University of Wisconsin-Madison
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 1993
• 资助国家: 美国
• 起止时间: 1993-04-01 至 1996-03-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=9218668&HistoricalAwards=false
• 关键词: Bridging; Microstructural; Continuum; Theories; DispersedSystems

The object of this proposal is the development of scalable parallel computing algorithms for multi-particle system flow with negligible inertial effects. The particles may have arbitrary shape and roughness. The preparation of the governing equation combined with the use of the advanced parallel computer in suspension rheology. The governing Stokes equations are transformed into a set of Fredholm integral equations of second kind with double-layer densities on the surfaces of particles and on the boundary of the system. The methodology can be applied to dilute and concentrated suspension flow, non- spherical particles and irregularly bounding walls. The emphasis of this proposal is on the computational algorithm rather than physical problems. The goal is to reduce the frequency of inter-processor communication, and to shorten the inter-processor messages, which would increase the overall computational speed. The development of an efficient computational methodology for suspension flow at low Reynolds numbers has a broad range of applications in chemical, pharmaceutical and other industrial processes that use particulate suspensions.

该提案的目标是开发可扩展的 多质点系流的并行计算算法 惯性效应可以忽略不计。 述颗粒可具有 任意形状和粗糙度。 政府的准备工作 方程结合使用先进的并行计算机 悬浮液流变学 控制斯托克斯方程是 转化为第二类Fredholm积分方程组 颗粒表面具有双层密度的一类 在系统的边界上。 方法可以是 适用于稀释和浓缩悬浮液流，非 球形颗粒和不规则边界壁。 的 该建议的重点是计算算法 而不是生理问题。 目标是减少 处理器间通信的频率，并缩短 处理器间消息，这将增加整体 计算速度 发展一种有效的计算方法， 在低雷诺数下悬浮流具有宽范围的 在化工、制药等行业的应用 使用颗粒悬浮液的工艺。