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

TL;DR

This paper introduces a deep learning reduced order model for 3D unsteady flow that uses mesh transformation and stitching to improve near-wall flow prediction accuracy without interpolation.

Contribution

It proposes a novel mesh transformation and stitching approach combined with a convolutional neural network for accurate 3D unsteady flow prediction.

Findings

High accuracy in near-wall flow prediction with a pressure correlation coefficient of 0.9985

Effective preservation of flow details during long-term flow field prediction

Validation on flow around a sphere at Re=300 demonstrates method's robustness

Abstract

Artificial intelligence-based three-dimensional(3D) fluid modeling has gained significant attention in recent years. However, the accuracy of such models is often limited by the processing of irregular flow data. In order to bolster the credibility of near-wall flow prediction, this paper presents a deep learning-based reduced order model for three-dimensional unsteady flow using the transformation and stitching of multi-block structured meshes. To begin with, full-order flow data is provided by numerical simulations that rely on multi-block structured meshes. A 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 an impressive value of 0.9985. Moreover, the periodic behavior of flow fields can be faithfully predicted during long-term inference.