Unstructured Mesh Generation Research Issues for Real World Applications

Marcum, D. (2009). Unstructured Mesh Generation Research Issues for Real World Applications. Current & New Trends in Scientific Computing. Santiago, Chile: Universidad de Chile.

Abstract

Large-scale computational field simulation is commonly used to help solve real world problems found in industry and government. Continuous and significant advances in computational hardware and solution algorithms have allowed the scale, scope and complexity of problems that can reasonably be considered to steadily progress. This presents challenges for all aspects of the simulation process. In particular, for unstructured mesh generation there are several research and implementation issues that can impede or hinder continued progress. Many of these have persisted for some time and have simply re-emerged with applications of increasing scale and complexity. Examples will be presented and discussed that illustrate some of these issues and the relationship between research and development and application requirements. Specific unstructured mesh generation topics will include surface meshing and transformation mapping issues arising from CAD to mesh automation, discrete geometry approximations, and geometric manipulation for optimization applications, along with volume meshing issues related to robustness, metric transformations, and BL regions.