SURFNet: Super-Resolution of Turbulent Flows with Transfer Learning using Small Datasets
SURFNet: Super-Resolution of Turbulent Flows with Transfer Learning using Small Datasets
复制标题
DOI：
10.1109/pact52795.2021.00031
复制
发表时间：
2021-08
期刊：
影响因子：
--
通讯作者：
Octavi Obiols-Sales;Abhinav Vishnu;Nicholas Malaya;Aparna Chandramowlishwaran
中科院分区：
文献类型：
--
作者：
Octavi Obiols-Sales;Abhinav Vishnu;Nicholas Malaya;Aparna Chandramowlishwaran
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开发下一代数据驱动的湍流代理模型的主要限制。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。这两种方法都需要仅对一种几何形状进行迁移学习，以解决细网格流场在高分辨率输入上需要的训练数据比粗模型的微小分辨率（64美元乘以256美元）少15倍，从而显着减少了时间数据收集和训练。我们通过求解湍流区域中的Navier-Stokes方程，以比粗模型大256倍的输入分辨率来经验性地评估SURFNet的性能。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，我们观察到与测试几何形状和分辨率大小无关的OpenFOAM物理求解器一致的2-2.1倍加速（高达2048 $\times 2048$），证明了分辨率不变性和泛化能力。此外，与以256 × 256$和512 × 512$网格分辨率收集大量训练数据的基线模型（aka oracle）相比，SURFNet实现了相同的性能增益，同时将合并的数据收集和训练时间分别减少了3.6倍和10.2倍。我们的方法解决了从使用低分辨率输入（即，超分辨率），而不损失准确性并且需要有限的计算资源。
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 15× less training data on high-resolution inputs compared to the tiny resolution ($64\times 256$) 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 256× larger than the coarse model. On four test geometries and eight flow configurations unseen during training, we observe a consistent 2–2.1× speedup over the OpenFOAM physics solver independent of the test geometry and the resolution size (up to $2048 \times 2048$), demonstrating both resolution-invariance and generalization capabilities. Moreover, compared to the baseline model (aka oracle) that collects large training data at $256 \times 256$ and $512 \times 512$ grid resolutions, SURFNet achieves the same performance gain while reducing the combined data collection and training time by 3.6× and 10.2×, respectively. Our approach addresses the challenge of reconstructing high-resolution solutions from coarse grid models trained using low-resolution inputs (i.e., super-resolution) without loss of accuracy and requiring limited computational resources.