A compressing and saving method for structured grid data based on block construction

Abstract

The increasing grid data in CFD simulation has brought some new difficulties and challenges, such as high storage cost, low transmission efficiency. In order to overcome these problems, a novel method for compressing and saving the structured grid are proposed. In the present method, the geometric coordinates of the six logical domains of one grid block is saved instead of all grid vertex coordinates to reduce the size of the structured grid file when the grid is compressed. And all grid vertex coordinates are recovered from the compressed data with the use of the transfinite interpolation algorithm when the grid is decompressed. Firstly, single-block grid cases with different edge vertexes are tested to investigate the compression effect. The test results show that a higher compression ratio will be obtained on a larger grid. Secondly, further theoretical analysis is carried out to investigate the effects of parameters on grid compression. The analysis on single-block grid compression shows that the compression ratio is proportionate to the cubic root of the number of total vertexes. The highest compression ratio of single-block grid is obtained when the numbers of vertexes in three logical directions are equal. The analysis on multi-block grid compression shows that a higher compression ratio will be obtained when a larger difference of total vertexes number exists among the grid blocks. Finally, multi-blockgrids of two industrial aircraft configurations are compressed to validate the method. The compression results demonstrate that the present method has an excellent ability on structured grid compression. For a million-vertex structured grid, more than 80 percent disk space can be saved after compression.