报告人：任杰(University of Stuttgart)

时间：2021年9月22日15：00-16:00

腾讯会议号码：376127623

报告摘要：

In this talk, we employ techniques that emerged in computer vision science to fluid mechanics. In the first part, we propose an image-based flow decomposition developed from the 2D tensor empirical wavelet transform (EWT). With the proposed method, decomposition of an instantaneous three-dimensional (3-D) flow becomes feasible without resorting to its time series. Unlike POD or DMD that extract spatial modes according to energy or frequency, EWT provides a new strategy for decomposing an instantaneous flow from its spatial scales. In the second part, we aim to provide new insights into the model reduction of travelling-wave problems by exploiting the Radon cumulative distribution transform (R-CDT). A substantial model-reduction is achieved in the R-CDT space while sustaining high accuracy in contrast to the physical space. The method is parameter-free and data-driven that lends itself to problems regardless of the dimensions or boundary conditions.

报告人简介：任杰，2007-2016清华大学，学士、博士；

2016-2017清华大学，博士后

2017-2019Delft University of Technology , Postdoc Researcher；

2019-2021 University of Nottingham, Research Fellow

2021-    University of Stuttgart, Alexander von Humboldt Research Fellow

邀请人：王义乾