arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2108.13220physics.comp-ph
收藏

基于机器学习的湍流动力系统统计闭包模型

Machine Learning-Based Statistical Closure Models for Turbulent Dynamical Systems

Di Qi, John Harlim

摘要:

我们提出了一个机器学习(ML)非马尔可夫封闭建模框架,用于精确预测湍流动力系统受外力作用的统计响应。这个统计闭包问题的困难之一是缺乏训练数据,这是一种在神经网络模型的监督学习中不可取的配置。在40维Lorenz-96模型的这项研究中,数据的短缺(在时间上)是由于去相关时间之外的统计数据的平稳性,因此,训练数据中唯一的信息内容是关于短时瞬态统计数据。我们采用了一个统一的封闭框架的各种截断制度,包括和排除详细的动力学方程的方差。闭包框架采用长短期记忆体系结构来表示高阶未解决的统计反馈,并仔细考虑了内在的不稳定性,同时产生稳定的长期预测。我们发现,这种统一的不可知ML方法在各种截断场景下表现良好。在数值上,ML闭包模型可以准确地预测受到各种时间依赖性外力的长期统计响应,这些外力不在训练数据集中(并且最大强迫幅度相对大于这些外力)。

英文摘要:

We propose a Machine Learning (ML) non-Markovian closure modeling framework for accurate predictions of statistical responses of turbulent dynamical systems subjected to external forcings. One of the difficulties in this statistical closure problem is the lack of training data, which is a configuration that is not desirable in supervised learning with neural network models. In this study with the 40-dimensional Lorenz-96 model, the shortage of data (in temporal) is due to the stationarity of the statistics beyond the decorrelation time, thus, the only informative content in the training data is on short-time transient statistics. We adopted a unified closure framework on various truncation regimes, including and excluding the detailed dynamical equations for the variances. The closure frameworks employ a Long-Short-Term-Memory architecture to represent the higher-order unresolved statistical feedbacks with careful consideration to account for the intrinsic instability yet producing stable long-time predictions. We found that this unified agnostic ML approach performs well under various truncation scenarios. Numerically, the ML closure model can accurately predict the long-time statistical responses subjected to various time-dependent external forces that are not (and maximum forcing amplitudes that are relatively larger than those) in the training dataset.

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。

↑