Adaptive spatial partitioning and refinement for overset structured grids Original Research Article

The need for adaptive refinement for unsteady aerodynamic applications that may involve relative motion between configuration components is recognized. An efficient means of adaptive refinement within systems of overset structured grids is presented. Problem domains are segregated into near-body and off-body fields. Near-body fields are discretized via overlapping body-fitted grids that extend a short distance from body surfaces. Off-body fields are discretized via systems of overlapping uniform Cartesian (structured) grids of varying levels of refinement. A novel method of adaptive spatial partitioning and refinement that is responsive to evolving off-body flow dynamics and proximity of moving solid bodies is described. Computational advantages of structured data are reviewed. Properties of uniform Cartesian grids that lead to substantial computational advantages are identified. A grid component grouping algorithm is presented. Formal accuracy of the method is considered. The method is demonstrated for three-dimensional unsteady viscous flow applications of practical relevance.