Leveraging Bayesian analysis to improve accuracy of approximate models

DOI10.1016/J.JCP.2019.05.015MaRDI QIDQ2222288zbMATH OpenOpenAlexWikidataFDO

Authors Chiyu Jiang, D. Livescu, Balasubramanya T. Nadiga

Publication date 26 January 2021

Published in Journal of Computational Physics (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1905.08227

zbMATH Keywords

Bayesian analysisneural networkreduced order modelingReynolds-averaged Navier Stokessurrogate modelingturbulence modeling

Mathematics Subject Classification ID

Bayesian inference (62F15) Artificial neural networks and deep learning (68T07) Statistical solutions of Navier-Stokes and related equations (76D06) Statistical turbulence modeling (76F55)

Abstract: We focus on improving the accuracy of an approximate model of a multiscale dynamical system that uses a set of parameter-dependent terms to account for the effects of unresolved or neglected dynamics on resolved scales. We start by considering various methods of calibrating and analyzing such a model given a few well-resolved simulations. After presenting results for various point estimates and discussing some of their shortcomings, we demonstrate (a) the potential of hierarchical Bayesian analysis to uncover previously unanticipated physical dependencies in the approximate model, and (b) how such insights can then be used to improve the model. In effect parametric dependencies found from the Bayesian analysis are used to improve structural aspects of the model. While we choose to illustrate the procedure in the context of a closure model for buoyancy-driven, variable-density turbulence, the statistical nature of the approach makes it more generally applicable. Towards addressing issues of increased computational cost associated with the procedure, we demonstrate the use of a neural network based surrogate in accelerating the posterior sampling process and point to recent developments in variational inference as an alternative methodology for greatly mitigating such costs. We conclude by suggesting that modern validation and uncertainty quantification techniques such as the ones we consider have a valuable role to play in the development and improvement of approximate models.

Recommendations

• New insights into approximate Bayesian computation
• scientific article; zbMATH DE number 270205
• Applied Bayesian Modelling
• Applied Bayesian Modelling
• Approximation enhancement for stochastic Bayesian inference
• Approximate Bayesian computational methods
• Improved approximate Bayesian computation methods via empirical likelihood

Cites work

• A Bayesian framework for adaptive selection, calibration, and validation of coarse-grained models of atomistic systems
• An adaptive Metropolis algorithm
• Application of a second-moment closure model to mixing processes involving multicomponent miscible fluids
• Approximation by superpositions of a sigmoidal function
• Bayesian estimates of parameter variability in the \(k-\varepsilon\) turbulence model
• Buoyancy-driven variable-density turbulence
• Dynamic mode decomposition of numerical and experimental data
• High-Reynolds number Rayleigh–Taylor turbulence
• Identification of almost invariant aggregates in reversible nearly uncoupled Markov chains
• MODEL STRUCTURAL INFERENCE USING LOCAL DYNAMIC OPERATORS
• Near-wall turbulence closure modeling without ``damping functions
• Optimal tuning of the hybrid Monte Carlo algorithm
• Orientation of eddy fluxes in geostrophic turbulence
• Predictions of Channel and Boundary-Layer Flows with a Low-Reynolds-Number Turbulence Model
• Predictive RANS simulations via Bayesian model-scenario averaging
• Proper orthogonal decomposition closure models for turbulent flows: a numerical comparison
• Response surface methodology: A neural network approach
• Spalart-Allmaras
• Stochastic parameterization for large eddy simulation of geophysical flows
• Toward a direct and scalable identification of reduced models for categorical processes
• Turbulent Flows
• Verification and validation in computational engineering and science: basic concepts

Cited in

(3)

• Improved modeling by coupling imperfect models
• Combining machine learning and data assimilation to forecast dynamical systems from noisy partial observations
• Leveraging initial conditions memory for modelling Rayleigh-Taylor turbulence

Describes a project that uses

Uses Software

• Spalart-Allmaras

This page was built for publication: Leveraging Bayesian analysis to improve accuracy of approximate models

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2222288)