Parallel Mesh Adaptation for Highly Evolving Geometries with Application to Solid Propellant Rockets

We describe our parallel 3-D surface and volume mesh modification strategy for large-scale simulation of physical systems with dynamically changing domain boundaries. Key components include an accurate, robust, and efficient surface propagation scheme, frequent mesh smoothing without topology changes, infrequent remeshing at regular intervals or when triggered by declining mesh quality, a novel hybrid geometric partitioner, accurate and conservative solution transfer to the new mesh, and a high degree of automation. We apply these techniques to simulations of internal gas flows in firing solid propellant rocket motors, as various geometrical features in the initially complex propellant configuration change dramatically due to burn-back. Smoothing and remeshing ensure that mesh quality remains high throughout these simulations without dominating the run time.

[1]  Robert D. Moser,et al.  Enabling three-dimensional unsteady SRM burn-out computations by slow-time acceleration , 2006 .

[2]  Laxmikant V. Kalé,et al.  A voxel-based parallel collision detection algorithm , 2002, ICS '02.

[3]  Patrick M. Knupp,et al.  MESQUITE design : issues in the development of a mesh quality improvement toolkit. , 2002 .

[4]  George Karypis,et al.  Multilevel k-way Partitioning Scheme for Irregular Graphs , 1998, J. Parallel Distributed Comput..

[5]  Jeffrey Grandy,et al.  Conservative Remapping and Region Overlays by Intersecting Arbitrary Polyhedra , 1999 .

[6]  Patrick M. Knupp,et al.  The Mesquite Mesh Quality Improvement Toolkit , 2003, IMR.

[7]  Houman Borouchaki,et al.  of Forming Processes in Three Dimensions A Remeshing Procedure for Numerical Simulation , 2006 .

[8]  Michael T. Heath,et al.  Building Rocstar: Simulation Science for Solid Propellant Rocket Motors , 2006 .

[9]  Laxmikant V. Kalé,et al.  ParFUM: a parallel framework for unstructured meshes for scalable dynamic physics applications , 2006, Engineering with Computers.

[10]  Michael T. Heath,et al.  A Framework for Grid Solvers , 1999 .

[11]  Patrick M. Knupp,et al.  Increasing TAU3P Abort-time via Mesh Quality Improvement , 2003, IMR.

[12]  Timothy J. Baker,et al.  Mesh Movement and Metamorphosis , 2002, Engineering with Computers.

[13]  Xiangmin Jiao,et al.  Face offsetting: A unified approach for explicit moving interfaces , 2007, J. Comput. Phys..

[14]  Vipin Kumar,et al.  A Coarse-Grain Parallel Formulation of Multilevel k-way Graph Partitioning Algorithm , 1997, PP.

[15]  Mark S. Shephard,et al.  Automated Adaptive Forming Simulations , 2003, IMR.

[16]  Houman Borouchaki,et al.  A Remeshing Procedure for Numerical Simulation of Forming Processes in Three Dimensions , 2006, IMR.

[17]  Xiangmin Jiao,et al.  Parallel Feature-Preserving Mesh Smoothing , 2005, ICCSA.

[18]  Patrick M. Knupp Mesh quality improvement for SciDAC applications , 2006 .

[19]  H. Piaggio The Mathematical Theory of Huygens' Principle , 1940, Nature.

[20]  Gary L. Miller,et al.  A Bezier-Based Moving Mesh Framework for Simulation with Elastic Membranes , 2004, IMR.

[21]  Kenji Shimada,et al.  Inverse Pre-Deformation of Finite Element Mesh for Large Deformation Analysis , 2004, IMR.

[22]  Bruce Hendrickson,et al.  Graph Partitioning and Parallel Solvers: Has the Emperor No Clother? (Extended Abstract) , 1998, IRREGULAR.

[23]  W. E. Bleick Review: B. B. Baker and E. T. Copson, The Mathematical Theory of Huygens' Principle , 1940 .

[24]  Laxmikant V. Kalé,et al.  Adaptive MPI , 2003, LCPC.

[25]  David R. White,et al.  Mesh‐based geometry , 2003 .