Computational Grid Optimization in Engineering

工程中的计算网格优化

• 批准号: 8708082
• 负责人: Graham Carey
• 金额: $ 47.81万
• 依托单位: University of Texas at Austin
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 1987
• 资助国家: 美国
• 起止时间: 1987-08-15 至 1991-07-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=8708082&HistoricalAwards=false
• 关键词: Computational; Grid; Optimization; Engineering

Grid generation, redistribution and adaptive refinement have emerged as important current aspects of computational engineering. With the rapid evolution of advanced supercomputers, large-scale 2-D and 3-D problems can now be addressed with appropriate grid resolution. To use the associated finite element or finite difference methods reliably and efficiently, we need improved grid-generation schemes and algorithms for grid redistribution and refinement. This research project is directed towards this problem of grid optimization using both redistribution and refinement strategies. The problem is approached from the viewpoint of nonlinear programming problems in optimization theory. The focus of the research will be the development of approximate methods and algorithms as well as techniques for acceleration of the grid optimization procedures including parallel processing on subdomains. The key technical issues are: grid redistribution and optimization, local adaptive grid refinement and remeshing schemes for moving grids. The institutional support is adequate and the investigation is a senior researcher with significant contributions in computational engineering.

网格的生成、重新分布和自适应细化 成为当前计算领域的重要方面 工程. 随着先进技术的快速发展， 超级计算机，大规模的二维和三维问题，现在可以 以适当的网格分辨率进行处理。 使用 相关的有限元或有限差分方法可靠 我们需要改进的网格生成方案， 网格重新分布和细化的算法。 本研究 项目是针对这个问题的网格优化 使用重新分配和细化策略。 问题 从非线性规划的观点出发， 优化理论中的问题。 研究的重点将 也是近似方法和算法的发展 作为加速电网优化的技术 包括子域上的并行处理的过程。 关键 技术问题是：网格重新分配和优化， 局部自适应网格加密和网格重划法 网格。 机构支持是足够的，调查是一个 高级研究员，在计算方面有重大贡献 工程.