Deep learning-based reduced order model for three-dimensional unsteady flow using mesh transformation and stitching

Abstract

Artificial intelligence-based three-dimensional fluid modeling has gained significant attention recently. However, the accuracy of such models is often limited by the preprocessing of irregular flow data. In order to bolster the credibility of near-wall flow prediction, we present a deep learning-based reduced order model for three-dimensional unsteady flow using the transformation and stitching techniques for multi-block structured meshes. To begin with, full-order flow data is provided by numerical simulations that rely on multi-block structured meshes. The mesh transformation technique is applied to convert each structured grid with data into a corresponding uniform and orthogonal grid, which is subsequently stitched and filled. The resulting snapshots in the transformed domain contain accurate flow information for multiple meshes and can be directly fed into a structured neural network without requiring any interpolation operation. Subsequently, a network model based on a fully convolutional neural network is constructed to predict flow dynamics accurately. To validate the strategy's feasibility, the flow around a sphere with Re=300 was investigated, and the results obtained using traditional Cartesian interpolation were used as the baseline for comparison. All the results demonstrate the preservation and accurate prediction of flow details near the wall, with the pressure correlation coefficient on the wall achieving a remarkable value of 0.9997. The stitching scheme that follows the proposed standard can effectively reduce the accumulation of inferring errors. Moreover, the periodic behavior of flow fields can be faithfully predicted during long-term inference. This work demonstrates that the proposed strategy has the advantage of high efficiency while providing accuracy assurance for downstream tasks.