基于虚网格的格心ADT搜索法

针对重叠网格寻点问题,提出了一种直接在格心网格下操作,基于虚网格的ADT(Alternating Digital Tree)搜索方法.通过对计算网格进行扩展,建立虚边界网格,解决了格心网格因覆盖区域不完整而无法直接建立ADT数据结构的问题.搜索结果即为包含解的格心单元集合,可直接对结果列表遍历以得到合理贡献单元,故完全摒弃了可靠性差的Stencil Walk方法.由虚网格的定义,使寻点在边界附近的处理更为灵活,可以准确给出边界附近贡献单元的有效信息,同时简化了虚网格体系的构建.扩展了重叠边界类型,构建的搜索空间能完整覆盖网格范围,解决了对称面重叠问题.算例研究表明:该方法可靠性好,边界处理能力强,有效提高了重叠网格方法对大型复杂网格的解算能力.

Cell-center ADT algorithm based on ghost cell

For the point search problem in Chimera gird generation,an alternating digital tree(ADT) method based on ghost cells was presented,which operated directly on the cell-centered grid system.The problem caused by the covering area incompleteness of cell-centered grids when constituting ADT data structure can be avoided by introducing ghost cells through extension of the original grid.The result of ADT searching with this method is a list of cell-centered cells,which can be directly traversed to obtain the exact donor cell.Therefore,the Stencil Walk method can be omitted.The use of ghost cells makes point search procedures near the boundary more flexible,providing exact information of donor cell near boundaries,and simplifying the construction of ghost cell system.One major application of the proposed method is expanding the types of the overset boundaries,as the constructed searching space is able to cover the entire grid,resolving the problem for the symmetry plane overset.The test case shows that the method is very reliable,feasible for boundaries,and greatly improves the flexibility of the Chimera grid for complex configurations.