A Surrogate Bayesian Inference Approach with Application to Large Eddy Simulation Turbulence Models

来源 :第八届工业与应用数学国际大会 | 被引量 : 0次 | 上传用户：magicarpet

【摘要】

：

　　In this effort,we developed an adaptive hierarchical sparse-grid(AHSG)surrogate modeling approach to Bayesian inference of large eddy simulation(LES)models,

【作者】

：

WebsterClayton;TranHoang;ZhangGuannan;

【机构】

：

Oak Ridge National Laboratory

【出处】

：

第八届工业与应用数学国际大会

【发表日期】

：

2015年期

下载到本地 , 更方便阅读

下载此文赞助VIP

声明 : 本文档内容版权归属内容提供方 , 如果您对本文有版权争议 , 可与客服联系进行内容授权或下架

论文部分内容阅读

　　In this effort,we developed an adaptive hierarchical sparse-grid(AHSG)surrogate modeling approach to Bayesian inference of large eddy simulation(LES)models,which,through a numerical demonstration of the Smagorinsky turbulence model of two-dimensional flow around a cylinder at sub-critical Reynolds number,is proven to significantly reduce the number of costly LES executions without losing much accuracy in the posterior probability estimation.

其他文献

Multi-scale Methods for Elliptic Equations with Highly Oscillatory Periodic Diffusion Coefficients a

　　We study the limiting distribution of the random part of the homogenization error of elliptic equations with highly oscillatory periodic diffusion coefficie

会议

Multiscale Approximations for Mixed Finite Element Methods

　　We propose and analyze a mixed finite element method for solving elliptic multiscale problems.The method is based on a localized orthogonal decomposition(LO

会议

Fast Multiscale Gaussian Wavepacket Transforms and Multiscale Gaussian Beams for Wave Equations

　　We develop new multiscale Gaussian beam methods for initial boundary value problems of wave equations in the high frequency regime.The starting tools are fa

会议

Eliminating the Pollution Effect in Helmholtz Problems by Local Subscale Correction

　　We introduce a new Petrov-Galerkin multiscale method for the numerical approximation of the Helmholtz equation with large wave number k in bounded domains.

会议

Fourier Cosine Expansions for Backward Stochastic Differential Equations in Finance

　　In this presentation we will explain how Fourier cosine expansions can be used to solve BSDEs originating from finance.By solving BSDEs,we can find the solu

会议

Numerical Approximation of Switching Problems

　　We use the representation of Switching Problems as obliquely reflected BSDEs to obtain a numerical approximation of the solution.We thus focus on the discre

会议

Error Estimate of the Crank-Nicolson Scheme for Solving the Decoupled FBSDEs

　　In this talk,by the theory of the multiple Malliavin calculus,we will introduce Crank-Nicolson scheme for solving decoupled forward-backward stochastic diff

会议

Efficient Numerical Methods for Backward Forward Doubly Stochastic Differential Equations and Their

　　Nonlinear filtering problems arise in many application fields such as financial mathematics,signal processing,target tracking,and biological sciences.

会议

Numerical Analysis for Forward-Backward Stochastic Differential Equations and Related Problems

　　Backward stochastic differential equations(BSDE's)were first introduced by J.M.Bismut in 1973 and generalized to the nonlinear form by Pardoux and Peng in 1

会议

A Dual-mixed Finite Element Method for the Brinkman Problem

　　In this talk,we develop a dual mixed finite element method for the Brinkman problem.In this approach,the velocity and the deviatoric part of the velocity gr

会议

A Surrogate Bayesian Inference Approach with Application to Large Eddy Simulation Turbulence Models

其他学术论文