Abstract
Large-scale computational fluid dynamics (CFD) are frequently attempted to obtain high resolution solutions or to solve large scale problems and then, bulky numerical data are generated. Data compression and reconstruction are indespensable to handle such kind of large data. The authors studied to apply flow analysis procedure to data compression and reconstruntion. In our method, the original data are compressed approximately by 1/100 by thinning out the data properly and by shortening the data bit length and reconstructed by applying an interpolation and simplefied fluid analysis technique. In this paper, we attenpt a multi-level data reconstruction to avoid the out-of-core memory problems and to reduce the processing time. By applying a local mesh refinement technique, only a part of the compressed data are reconstructed according to sub-regions to be displayed on monitor screen in this method. Furthermore, the reconstruction level is chosen appropriately depending on resolution displayed.
