Quantifying the uncertainty introduced by discretization and time‐averaging in two‐fluid model predictions

Syamlal, Madhava; Benyahia, Sofiane

doi:10.1002/aic.15868

Cited by 10 publications

(4 citation statements)

References 46 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…MFiX development has been ongoing at National Energy Technology Laboratory for over two decades, beginning with a multiparticle algorithm (Syamlal, ) based on the Eulerian‐Eulerian approach. During past years, MFiX has undergone continuous model development, verification and validation, and a wide range of application (Benyahia et al, ; Benyahia, ; Bergantz et al, ; Dartevelle, ; Dartevelle & Valentine, ; Dufek & Bergantz, ; Li et al, ; Srivastava & Sundaresan, ; M Syamlal et al, ). In this study we focus on the MFIX‐TFM (two‐fluid) model package, which describes the motion of a mixture of gas (or liquid) and solids as two or more continuum phases.…”

Section: Introductionmentioning

confidence: 99%

Continuum Modeling of Pressure‐Balanced and Fluidized Granular Flows in 2‐D: Comparison With Glass Bead Experiments and Implications for Concentrated Pyroclastic Density Currents

2019

View full text Add to dashboard Cite

Granular flows are found across multiple geophysical environments and include pyroclastic density currents, debris flows, and avalanches, among others. The key to describing transport of these hazardous flows is the rheology of these complex multiphase mixtures. Here we use the multiphase model MFIX in 2‐D for concentrated currents to examine the implications of rheological assumptions and validate this approach through comparison to experiments of both frictional and fluidized flows made of glass beads (Sauter mean grain‐size of 75 μm). Because the rheology of highly polydisperse, highly angular, polydensity granular mixtures is poorly known, we focus on simplified monodisperse or bidisperse mixtures described by the frictional flow theory of Schaeffer (1987, https://doi.org/10.1016/0022-0396(87)90038-6) and Srivastava‐Sundaresan, often referred to as the Princeton model (Srivastava & Sundaresan, 2003, https://doi.org/10.1016/S0032-5910(02)00132-8). We show that simulations including the latter model replicate well the flow shape, kinematics, and pore fluid pressure that match well‐constrained dam‐break experiments of initially fluidized or pressure‐balanced granular flows. Simulations reveal that pore fluid pressure is intrinsically modulated by dilation and compaction of the flow and hence can be generated in concentrated pyroclastic density currents. We use these simulations to interpret basal pore pressure signals from local flow properties (mixture density, solid velocity, and pore fluid pressure). Rheological changes retard these simulated flows considerably near the predicted runout, but most continuum models cannot inherently predict a zero velocity. We suggest an inertial number parameter that can be used to approximate deposition, and this approach could be a valuable tool used to validate simulations against natural pyroclastic current examples.

show abstract

Section: Introductionmentioning

confidence: 99%

Continuum Modeling of Pressure‐Balanced and Fluidized Granular Flows in 2‐D: Comparison With Glass Bead Experiments and Implications for Concentrated Pyroclastic Density Currents

2019

View full text Add to dashboard Cite

show abstract

“…Although most of the numerical scheme is only formally first-order accurate, with the application of a fine mesh, a small DEM timestep and a small sub-cycling restriction (dt CF D /dt DEM ≤ 20), we assume that the largest source of numerical error is statistical, i.e., due to finite time-averaged statis-tics. In this work, we use the method of non-overlapping bins to compute confidence intervals (CIs) on the time-averaged data [28]. Twelve temporal non-overlapping bins are used in each case.…”

Section: Modeling Strategymentioning

confidence: 99%

Benchmarking of a preliminary MFiX-Exa code

Fullmer¹,

Almgren²,

Rosso³

et al. 2019

Preprint

View full text Add to dashboard Cite

MFiX-Exa is a new code being actively developed at Lawrence Berkeley National Laboratory and the National Energy Technology Laboratory as part of the U.S. Department of Energy's Exascale Computing Project. The starting point for the MFiX-Exa code development was the extraction of basic computational fluid dynamic (CFD) and discrete element method (DEM) capabilities from the existing MFiX-DEM code which was refactored into an AMReX code architecture, herein referred to as the preliminary MFiX-Exa code. Although drastic changes to the codebase will be required to produce an exascale capable application, benchmarking of the originating code helps to establish a valid start point for future development. In this work, four benchmark cases are considered, each corresponding to experimental data sets with history of CFD-DEM validation. We find that the preliminary MFiX-Exa code compares favorably with classic MFiX-DEM simulation predictions for three slugging/bubbling fluidized beds and one spout-fluid bed. Comparison to experimental data is also acceptable (within accuracy expected from previous CFD-DEM benchmarking and validation exercises) which is comprised of several measurement techniques including particle tracking velocimetry, positron emission particle tracking and magnetic resonance imaging. The work concludes with an overview of planned developmental work and potential benchmark cases to validate new MFiX-Exa capabilities.

show abstract

“…42 Through UQ, a model is used not only to predict the best-estimate solution by using mean values of measured properties, but also to quantify the uncertainty in the model predictions, that is, model form uncertainty, by propagating measured input uncertainties through the model and into the solution itself. [46][47][48][49][50][51] In this work, we aim to provide experimental data for CFD-DEM model validation including UQ, extending the existing database 52 to a more complex system: a semicircular fluidized bed with pairs of opposing horizontal air jets. By considering UQ throughout the validation process it is also possible to extrapolate model form uncertainty from a validation phase-space (that is, the region of conditions for which experimental data exists) to a case of true model prediction in which no experimental data exists.…”

Section: Introductionmentioning

confidence: 99%

“…43 Used extensively in other fields, 44,45 a verification and validation (V&V) framework including UQ has only recently emerged within the gas-solid multiphase CFD community. [46][47][48][49][50][51] In this work, we aim to provide experimental data for CFD-DEM model validation including UQ, extending the existing database 52 to a more complex system: a semicircular fluidized bed with pairs of opposing horizontal air jets. This system is widely used in engineering processes to aid fluidization 53,54 and solid mixing, 55 to introduce gas reactants 56 or to perform particle attrition.…”

Section: Introductionmentioning

confidence: 99%

Experimental data for code validation: Horizontal air jets in a semicircular fluidized bed of Geldart Group D particles

et al. 2018

View full text Add to dashboard Cite

Experiments were conducted with 6 mm plastic beads (Geldart Group D) in a semi‐circular, gas‐fluidized bed with side jets. Attention was paid to particle characterization and bed measurements, making the resulting dataset ideal for CFD‐DEM validation and uncertainty quantification. The bed was operated slightly above and below the minimum fluidization velocity, with additional fluidization provided by one of two pairs of opposing jets located above the distributor near the flat, front face of the unit. Care is taken to report material properties and bed conditions with either measured distribution functions or uncertainty bounds. High‐speed video imaging and particle tracking velocimetry are used to extract bin‐averaged velocity profiles, which are used to extract jet penetration depths. The time‐averaged mean and standard deviation of the bed pressure drop is also reported. Finally, the lower jets are also inserted into the bed until the opposing jets merge to form a spout‐like pattern. © 2018 American Institute of Chemical Engineers AIChE J, 64: 2351–2363, 2018

show abstract

Quantifying the uncertainty introduced by discretization and time‐averaging in two‐fluid model predictions

Cited by 10 publications

References 46 publications

Continuum Modeling of Pressure‐Balanced and Fluidized Granular Flows in 2‐D: Comparison With Glass Bead Experiments and Implications for Concentrated Pyroclastic Density Currents

Continuum Modeling of Pressure‐Balanced and Fluidized Granular Flows in 2‐D: Comparison With Glass Bead Experiments and Implications for Concentrated Pyroclastic Density Currents

Benchmarking of a preliminary MFiX-Exa code

Experimental data for code validation: Horizontal air jets in a semicircular fluidized bed of Geldart Group D particles

Contact Info

Product

Resources

About