BriefGPT.xyz
Jun, 2022
大涡模拟中的深度强化学习用于湍流建模
Deep Reinforcement Learning for Turbulence Modeling in Large Eddy Simulations
HTML
PDF
Marius Kurz, Philipp Offenhäuser, Andrea Beck
TL;DR
本论文基于强化学习理论,通过卷积神经网络建立了大涡模拟中湍流参数、长时间稳定的模拟、精度等方面都优于传统分析模型的涡黏度动态自适应模型,并将其推广到不同分辨率和离散化的场景,为完成大涡模拟提供一种持续、准确和稳定的参数化框架。
Abstract
Over the last years,
supervised learning
(SL) has established itself as the state-of-the-art for data-driven
turbulence modeling
. In the SL paradigm, models are trained based on a dataset, which is typically comp
→