New digital meshing, computational tools unveiled
By James Masters and Carolyn Woeber|September 10, 2018
The Meshing, Visualization and Computational Environments Technical Committee is concerned with interactive computer graphics systems and their application to aerospace design and computer-aided manufacturing.
Computational analysis continues to be an essential component of the aeronautics and astronautics analysis landscape. The meshing, visualization and computational environments that are necessary to create the structure within which this analysis takes place continue to advance to meet current needs as well as anticipate the needs of the future.
On the commercial meshing side, Pointwise Inc., worked with the U.S. Air Force to improve the integrated overset meshing and assembly tools in their flagship meshing software. Motivated by the need for large configuration management, Pointwise developed a system of hierarchical containers called Frameworks, which facilitate flexible model definition, overset meshing and grid assembly. Additionally, a new mesh technique, termed voxel meshing, was used to provide a unique sketch-based interface for rapid off-body meshing. These technologies, which were developed as part of an Air Force-sponsored Commercialization Readiness Program, were demonstrated to the Air Force in September and are scheduled for inclusion in an upcoming version of the commercial software. Syracuse University and the Massachusetts Institute of Technology continue to make strides in their collaborative production of Engineering Sketch Pad, or ESP, software that will quickly generate geometry for general air vehicle models. These models can be used for initial design studies and parametric analyses. ESP leverages vehicle-generation concepts, which have typically focused on manufacturing, and expands these concepts for the generation of models for aerospace design and multidisciplinary analysis and optimization. This work built on ideas presented in a paper titled “Design Sensitivity Calculations Directly on CAD-based Geometry,” which was this year’s recipient of the MVCE technical committee’s Shahyar Pirzadeh Best Paper Award. Also out of Syracuse came a novel approach for generating parametric geometry models based on existing point clouds. Because many high-fidelity legacy geometries are not readily available, this technique provides an intriguing alternative.
Work continues in the realm of in situ processing in which data is intelligently reduced, analyzed, transformed and indexed while still in memory before being written to disk or transferred over networks. A collaboration between the Defense Department and Intelligent Light brought together Kestrel, the DoD flow solver framework, and the FieldView commercial visualization tool to demonstrate a workflow using in situ data extraction that showed good scalability and acceptable overhead. This workflow produced a fast and flexible methodology that reduced costly communication while preserving data for later analysis and visualization.
Within the computational fluid dynamics community, finite element solvers able to generate high-order solutions are reaching a maturity level where they are becoming serious contenders as tools for solving production-level problems. With finite element solvers becoming more accepted, there is an impending need to generate appropriate computational meshes and adequately visualize resulting solutions. Some initial work has been done on the high-order meshing front by the University of Tennessee and Pointwise that included mesh generation by springfield and vector adding as well as weighted condition number optimization. True production level high-order mesh generation and visualization remains elusive but the need and incentive is clear and resources are being allocated to attack these important problems.
The meshing, visualization and computational environments community continues to be motivated by the finding from NASA’s “CFD Vision 2030 Study” that mesh generation is a bottleneck in the CFD workflow. To that end, a special session on Vision 2030 goals related to Computational Environments will be held at AIAA SciTech 2017 in Texas.★