Parameter estimation for complex thermal-fluid flows using approximate Bayesian computation

Christopher, Jason D.; Wimer, Nicholas T.; Lapointe, Caelan; Hayden, Torrey R. S.; Grooms, Ian; Rieker, Gregory B.; Hamlington, Peter E.

doi:10.1103/physrevfluids.3.104602

Cited by 16 publications

(10 citation statements)

References 58 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Buoyant jets and plumes are found in a variety of natural and engineering contexts, including industrial burners (Christopher et al 2018;Hayden et al 2019), hydrothermal vents (Gaskin & Wood 2001) and volcanic plumes (Campion et al 2018). Despite their obvious differences, each of these flows exhibit a similar 'puffing' phenomenon that results from the continuous injection of less dense fluid into a reservoir of more dense, ambient fluid.…”

Section: Introductionmentioning

confidence: 99%

Scaling of the puffing Strouhal number for buoyant jets and plumes

et al. 2020

View full text Add to dashboard Cite

Section: Introductionmentioning

confidence: 99%

Scaling of the puffing Strouhal number for buoyant jets and plumes

et al. 2020

View full text Add to dashboard Cite

“…With more pyrolysis and combustion relevant species, comparisons to the mid-infrared frequency comb measurements of Makowiecki et al (2020b) will be possible with the inclusion of the reduced biomass combustion model published by Glusman et al (2019). Moreover, coupling these computationally efficient simulations with parameter estimation methods such as approximate Bayesian computation (Christopher et al, 2018), could allow for automated estimation of simulation parameters (i.e., heat of reaction and Arrhenius reaction coefficients) as well as initial conditions (i.e., temperature and water mole fraction) and/or boundary conditions (i.e., unmeasured parameters during experimental procedure).…”

Section: Discussionmentioning

confidence: 99%

Validation of Computationally Efficient Simulations of Douglas Fir Pyrolysis and Combustion Using Time-Resolved Frequency Comb Laser Measurements

Glusman

Lapointe

Makowiecki

et al. 2022

Front. For. Glob. Change

Self Cite

View full text Add to dashboard Cite

Computational simulations have the potential to provide low-cost, low-risk insights into wildland fire structure and dynamics. Simulation accuracy is limited, however, by the difficulty of modeling physical processes that span a wide range of spatial scales. These processes include heat transfer via radiation and turbulent advection, as well as both solid- and gas-phase chemistry. In the present study, we perform large eddy simulation (LES) with adaptive mesh refinement to model the multi-phase pyrolysis and combustion of dry Douglas fir, where temperature-based lookup tables corresponding to a multi-step pyrolysis mechanism are used to represent the composition of gas-phase pyrolysis products. Gas-phase and surface temperatures, mass loss, and water vapor mole fraction from the LES are shown to compare favorably with experimental measurements of a radiatively heated Douglas fir fuel sample undergoing pyrolysis and combustion beneath a cone calorimeter. Using frequency comb laser diagnostics, optical and infrared cameras, and a load cell, the experiments provide simultaneous in situ, time-resolved measurements of chemical composition, temperature, and mass loss. The present study thus combines cutting edge computational and experimental techniques with multi-step chemical pyrolysis modeling to provide a validated computational tool for the prediction of solid fuel pyrolysis and combustion relevant to wildland fires.

show abstract

“…The ABC method was introduced, and first widely applied, in population genetics [20][21][22], and was subsequently implemented in a wide range of other scientific areas [23][24][25][26][27][28] (detailed reviews are provided by Csilléry et al [29], Marin et al [30], Lintusaari et al [31], Sisson et al [32], and Beaumont [33]). More recently, ABC has been employed in engineering contexts for the estimation of rate coefficients in chemical kinetic models [34], for the estimation of boundary conditions in complex thermal-fluid flows [35], and for determining unknown model parameters in autonomic [36] and nonlinear [37] subgrid-scale closure models for LES. Doronina et al [14] were the first to take advantage of ABC-MCMC for discovering model parameter values and uncertainties in a multi-parameter RANS closure, using relatively simple homogeneous test cases as reference data for calibration of a nonequilibrium turbulence model [38][39][40] consisting of three coupled ordinary differential equations (ODEs).…”

Section: Approximate Bayesian Computation With Markov Chain Monte Car...mentioning

confidence: 99%

Parameter Estimation for RANS Models Using Approximate Bayesian Computation

Doronina¹,

Murman²,

Hamlington³

2020

Preprint

Self Cite

View full text Add to dashboard Cite

We use approximate Bayesian computation (ABC) to estimate unknown parameter values, as well as their uncertainties, in Reynolds-averaged Navier-Stokes (RANS) simulations of turbulent flows. The ABC method approximates posterior distributions of model parameters, but does not require the direct computation, or estimation, of a likelihood function. Compared to full Bayesian analyses, ABC thus provides a faster and more flexible parameter estimation for complex models and a wide range of reference data. In this paper, we describe the ABC approach, including the use of a calibration step, adaptive proposal, and Markov chain Monte Carlo (MCMC) technique to accelerate the parameter estimation, resulting in an improved ABC approach, denoted ABC-IMCMC. As a test of the classic ABC rejection algorithm, we estimate parameters in a nonequilibrium RANS model using reference data from direct numerical simulations of periodically sheared homogeneous turbulence. We then demonstrate the use of ABC-IMCMC to estimate parameters in the Menter shear-stress-transport (SST) model using experimental reference data for an axisymmetric transonic bump. We show that the accuracy of the SST model for this test case can be improved using ABC-IMCMC, indicating that ABC-IMCMC is a promising method for the calibration of RANS models using a wide range of reference data.

show abstract

Parameter estimation for complex thermal-fluid flows using approximate Bayesian computation

Cited by 16 publications

References 58 publications

Scaling of the puffing Strouhal number for buoyant jets and plumes

Scaling of the puffing Strouhal number for buoyant jets and plumes

Validation of Computationally Efficient Simulations of Douglas Fir Pyrolysis and Combustion Using Time-Resolved Frequency Comb Laser Measurements

Parameter Estimation for RANS Models Using Approximate Bayesian Computation

Contact Info

Product

Resources

About