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 2412.04239physics.flu-dyn
收藏

关于湍流代数代理模型的基于深度学习的闭合模型

On Deep-Learning-Based Closures for Algebraic Surrogate Models of Turbulent Flows

Barcelona Supercomputing Center(巴塞罗那超级计算中心);Universitat Politècnica de Catalunya(加泰罗尼亚理工大学);KTH Royal Institute of Technology(瑞典皇家理工学院)

Benet Eiximeno, Marcial Sanchís-Agudo, Arnau Miró, Ivette Rodríguez, Ricardo Vinuesa, Oriol Lehmkuhl

AI总结:

本文提出一种基于深度学习的闭合模型,利用 Transformer-encoder 和简易注意力机制预测截断 POD 模态未能捕获的波动,有效弥补了低维湍流代理模型的能量损失,显著降低了能量预测误差和速度分布的 KL 散度。

摘要:

本文提出了一种基于深度学习的闭合模型,用于解决基于本征正交分解(POD)模态的低维代理模型中的能量损失问题。该模型采用带有简易注意力机制的 Transformer-encoder 模块,预测截断 POD 模态未能捕获的波动的空间概率密度函数。该方法在偏航角为 [2.5, 5, 7.5, 10, 12.5] 的 Windsor 模型尾流上进行了演示,其中以 7.5 作为测试案例。通过使用 Hotelling T2 对时间系数的频谱特性进行聚类,基于主导频率动力学识别出关键相干模态。这些相干模态占总能量近 60%,而占所有模态的比例不到 10%。通过将训练角度的相干模态拼接并正交化该集合,创建了一个通用的 POD 基,在不丢失信息的情况下将基向量从 142 个减少到 90 个。训练了在注意力层具有不同大小(64、128 和 256)的 Transformer 来建模缺失的波动。较大的注意力大小始终能改善训练集的预测结果,但注意力层大小为 256 的 Transformer 在测试集中对波动的预测偏高,因为测试集的波动强度低于训练集。添加预测的波动弥合了重建流场与原始流场之间的能量差距,改善了对能量、均方根速度波动和瞬时流场的预测。最深的架构将平均能量误差从 37% 降低到 12%,并将速度分布的 Kullback-Leibler 散度从 KL=0.2 降低到 KL=0.026 以下。

英文摘要:

A deep-learning-based closure model to address energy loss in low-dimensional surrogate models based on proper-orthogonal-decomposition (POD) modes is introduced. Using a transformer-encoder block with easy-attention mechanism, the model predicts the spatial probability density function of fluctuations not captured by the truncated POD modes. The methodology is demonstrated on the wake of the Windsor body at yaw angles of [2.5,5,7.5,10,12.5], with 7.5 as a test case. Key coherent modes are identified by clustering them based on dominant frequency dynamics using Hotelling T2 on the spectral properties of temporal coefficients. These coherent modes account for nearly 60% of the total energy while comprising less than 10% of all modes. A common POD basis is created by concatenating coherent modes from training angles and orthonormalizing the set, reducing the basis vectors from 142 to 90 without losing information. Transformers with different size on the attention layer, (64, 128 and 256), are trained to model the missing fluctuations. Larger attention sizes always improve predictions for the training set, but the transformer with an attention layer of size 256 overshoots the fluctuations predictions in the test set because they have lower intensity than in the training cases. Adding the predicted fluctuations closes the energy gap between the reconstruction and the original flow field, improving predictions for energy, root-mean-square velocity fluctuations, and instantaneous flow fields. The deepest architecture reduces mean energy error from 37% to 12% and decreases the Kullback--Leibler divergence of velocity distributions from KL=0.2 to below KL=0.026.

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。

↑