Convergence of POD Modes for Cylinder Wake Flow

Author:

Hlavatý T.,Belda M.,Isoz M.

Abstract

Proper orthogonal decomposition (POD) is a popular technique for both flow analysis and model order reduction. However, its applicability is strongly dependent on the amount of input data. In the present contribution, we investigate the convergence of numerical POD modes for the canonical case of turbulent flow in the wake behind a circular cylinder. In particular, the cylinder placed at a crossflow at Reynolds number of 4815 was simulated using detached eddy simulation (DES) with sufficient mesh resolution to achieve a large eddy simulation (LES) and direct numerical simulation (DNS) blend in the region of interest located directly behind the examined cylinder. The convergence of POD modes with respect to (i) the number of temporal snapshots used, (ii) the enforcement of the assumed symmetries, and (iii) spatial resolution of the CFD results was examined. Only the first eleven modes can be assumed converged when using standard POD and 12000 snapshots sampled at 2 kHz. However, the results may be significantly improved by enforcing symmetries in the data, which enables to converge even the fifteenth mode with a significantly lower number of snapshots. Finally, simple norm-based techniques did not proved to be a sufficient tool for POD toposes comparison. Therefore the quantitative evaluation of the modes convergence remains an open question.

Publisher

Institute of Thermomechanics of the Czech Academy of Sciences; CTU in Prague Faculty of Mech. Engineering Dept. Tech. Mathematics

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3