arXiv 2108.076672021-08-17physics.flu-dyncs.AI

SURFNet：使用小数据集的迁移学习的湍流超分辨率

SURFNet: Super-resolution of Turbulent Flows with Transfer Learning using Small Datasets

University of California, Irvine（加州大学尔湾分校）；Advanced Micro Devices, Inc.（超威半导体公司）

Octavi Obiols-Sales, Abhinav Vishnu, Nicholas Malaya, Aparna Chandramowlishwaran

摘要：

深度学习（DL）算法正在成为计算昂贵的CFD模拟的关键替代方案。然而，最先进的DL方法需要大量高分辨率的训练数据来学习准确的模型。这些数据集的大小和可用性是下一代数据驱动的湍流替代模型开发的主要限制。本文介绍了SURFNet，一个基于迁移学习的超分辨率流网络。SURFNet主要在低分辨率数据集上训练DL模型，并在少数高分辨率流动问题上转移学习模型-加速传统的数值求解器，而与输入大小无关。我们提出了两种用于超分辨率任务的迁移学习方法，即一次性学习和增量学习。这两种方法都只需要对一种几何形状进行迁移学习，以解决细网格流场的问题，与粗模型的微小分辨率（64 x256）相比，高分辨率输入的训练数据少15倍，大大减少了数据收集和训练的时间。我们通过求解湍流状态下的纳维-斯托克斯方程，在输入分辨率高达粗模型的256倍的情况下，对SURFNet的性能进行了经验评估。在四个测试几何形状和八个在训练过程中看不见的流动配置上，我们观察到OpenFOAM物理求解器的一致2- 2.1倍加速，与测试几何形状和分辨率大小（高达2048 x2048）无关，证明了分辨率不变性和泛化能力。我们的方法解决了从使用低分辨率输入（超分辨率）训练的粗网格模型重建高分辨率解决方案的挑战，而不会损失准确性，并且需要有限的计算资源。

英文摘要：

Deep Learning (DL) algorithms are emerging as a key alternative to computationally expensive CFD simulations. However, state-of-the-art DL approaches require large and high-resolution training data to learn accurate models. The size and availability of such datasets are a major limitation for the development of next-generation data-driven surrogate models for turbulent flows. This paper introduces SURFNet, a transfer learning-based super-resolution flow network. SURFNet primarily trains the DL model on low-resolution datasets and transfer learns the model on a handful of high-resolution flow problems - accelerating the traditional numerical solver independent of the input size. We propose two approaches to transfer learning for the task of super-resolution, namely one-shot and incremental learning. Both approaches entail transfer learning on only one geometry to account for fine-grid flow fields requiring 15x less training data on high-resolution inputs compared to the tiny resolution (64x256) of the coarse model, significantly reducing the time for both data collection and training. We empirically evaluate SURFNet's performance by solving the Navier-Stokes equations in the turbulent regime on input resolutions up to 256x larger than the coarse model. On four test geometries and eight flow configurations unseen during training, we observe a consistent 2-2.1x speedup over the OpenFOAM physics solver independent of the test geometry and the resolution size (up to 2048x2048), demonstrating both resolution-invariance and generalization capabilities. Our approach addresses the challenge of reconstructing high-resolution solutions from coarse grid models trained using low-resolution inputs (super-resolution) without loss of accuracy and requiring limited computational resources.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。

摘要：

英文摘要：

相关论文