上海交通大学 a. 船舶海洋与建筑工程学院； b. 海洋工程国家重点实验室， 上海 200240
出版日期:2020-06-28发布日期:2020-07-03通讯作者:邹早建，男，教授，博士生导师，E-mail: zjzou@sjtu.edu.cn.作者简介:夏立 (1990-)，男，浙江省舟山市人，博士生，研究方向为不确定度量化.Uncertainty Quantification for CFD Simulation of Stochastic Drag Flow Based on Non-Intrusive Polynomial Chaos Method
XIA Li a,ZOU Zaojian a,b,YUAN Shuai a,ZENG Zhihua aa. School of Naval Architecture, Ocean and Civil Engineering; b. State Key Laboratory of Ocean Engineering, Shanghai Jiao Tong University, Shanghai 200240, China
Online:2020-06-28Published:2020-07-03摘要/Abstract
摘要： 比较了计算流体动力学不确定度分析中的验证与确认方法和不确定度量化方法之间的区别.介绍了一种最新的不确定度量化方法——非侵入式混沌多项式法，并应用该方法和传统的蒙特卡洛法分别对二维随机阻曳流进行不确定度量化，其中蒙特卡洛法选用了随机采样法和拉丁超立方采样法两种方法.研究了当进出口压力为服从均匀分布的随机变量时所引起的流动的不确定性.研究结果表明，应用蒙特卡洛法进行不确定度量化时，随机采样法和拉丁超立方采样法效果差别不大；混沌多项式方法相较于蒙特卡洛法可以更高效地模拟不确定性在流场中的传播.
关键词: 不确定度量化, 计算流体动力学, 混沌多项式方法, 拉丁超立方采样
Abstract: In this paper, verification & validation and uncertainty quantification in uncertainty analysis for computational fluid dynamics (CFD) simulation are compared. A state-of-the-art method for uncertainty quantification problems, i.e., the non-intrusive polynomial chaos (NIPC) method, is introduced and applied to quantifying the uncertainty of two-dimensional stochastic drag flow, together with the Monte-Carlo (MC) method. For the MC method, the random sampling (RS) method and the Latin hypercube sampling (LHS) method are adopted. The uncertainty of the stochastic drag flow induced by the inlet and outlet pressure boundaries is studied, with the boundaries treated as stochastic variables with uniform distribution. It is shown that there is no big difference between LHS and RS, and the NIPC method can simulate the uncertainty propagation better.
Key words: uncertainty quantification, computational fluid dynamics (CFD), polynomial chaos method, Latin hypercube sampling
PDF全文下载地址:
点我下载PDF
