Quantifying uncertainty of a reacting multiphase flow in a bench-scale fluidized bed gasifier: A Bayesian approach

Aytekin Gel, Mehrdad Shahnam, Arun K. Subramaniyan

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

Adequate assessment of the uncertainties is becoming an integral part of the engineering design. This study is based on several efforts that are underway to investigate the uncertainty quantification (UQ) methods that are applicable to multiphase flows. Although particular emphasis has been given on identifying, characterizing and managing uncertainties in computational fluid dynamics (CFD) simulations, experimental data used in the validation of the CFD model is as critical. Hence, the goal of this paper is to demonstrate application of non-intrusive Bayesian uncertainty quantification methodology in multiphase (gas-solid) flows with experimental data as part of our research efforts to determine the most suited approach for UQ. For this purpose, a bench scale fluidized bed gasifier experiment from a prior study is used to demonstrate the applicability of Bayesian techniques on sparse data. Global sensitivity analysis performed as part of the UQ study shows that among the three operating factors, steam to oxygen ratio has the most influence on syngas composition in the gasifier. An analysis for forward propagation of uncertainties shows that an increase in steam to oxygen ratio leads to an increase in H2 mole fraction and a decrease in CO mole fraction. These findings are in agreement with the reference experimental study. Another contribution in addition to the UQ analysis is the optimization-based approach employed to identify next best set of additional experimental samples. Hence, the surrogate models constructed as part of the UQ analysis is employed to improve the information gain and make incremental recommendation, should the possibility to add more experiments arise. The insight gained from this study has been extensively used in the follow-up UQ study with the CFD modeling of the same gasifier configuration.

Original languageEnglish (US)
Pages (from-to)484-495
Number of pages12
JournalPowder Technology
Volume311
DOIs
StatePublished - Apr 15 2017
Externally publishedYes

Fingerprint

Multiphase flow
Fluidized beds
Computational fluid dynamics
Uncertainty analysis
Steam
Oxygen
Flow of solids
Uncertainty
Carbon Monoxide
Sensitivity analysis
Dynamic models
Gases
Experiments
Computer simulation
Chemical analysis

Keywords

  • Bench-scale experiments
  • Data-fitted surrogate models
  • Fluidized bed gasifier
  • Non-intrusive Bayesian uncertainty quantification
  • Reacting multiphase flows

ASJC Scopus subject areas

  • Chemical Engineering(all)

Cite this

Quantifying uncertainty of a reacting multiphase flow in a bench-scale fluidized bed gasifier : A Bayesian approach. / Gel, Aytekin; Shahnam, Mehrdad; Subramaniyan, Arun K.

In: Powder Technology, Vol. 311, 15.04.2017, p. 484-495.

Research output: Contribution to journalArticle

Gel, Aytekin ; Shahnam, Mehrdad ; Subramaniyan, Arun K. / Quantifying uncertainty of a reacting multiphase flow in a bench-scale fluidized bed gasifier : A Bayesian approach. In: Powder Technology. 2017 ; Vol. 311. pp. 484-495.
@article{3620c86b4c9348fdb52a6909112abcc3,
title = "Quantifying uncertainty of a reacting multiphase flow in a bench-scale fluidized bed gasifier: A Bayesian approach",
abstract = "Adequate assessment of the uncertainties is becoming an integral part of the engineering design. This study is based on several efforts that are underway to investigate the uncertainty quantification (UQ) methods that are applicable to multiphase flows. Although particular emphasis has been given on identifying, characterizing and managing uncertainties in computational fluid dynamics (CFD) simulations, experimental data used in the validation of the CFD model is as critical. Hence, the goal of this paper is to demonstrate application of non-intrusive Bayesian uncertainty quantification methodology in multiphase (gas-solid) flows with experimental data as part of our research efforts to determine the most suited approach for UQ. For this purpose, a bench scale fluidized bed gasifier experiment from a prior study is used to demonstrate the applicability of Bayesian techniques on sparse data. Global sensitivity analysis performed as part of the UQ study shows that among the three operating factors, steam to oxygen ratio has the most influence on syngas composition in the gasifier. An analysis for forward propagation of uncertainties shows that an increase in steam to oxygen ratio leads to an increase in H2 mole fraction and a decrease in CO mole fraction. These findings are in agreement with the reference experimental study. Another contribution in addition to the UQ analysis is the optimization-based approach employed to identify next best set of additional experimental samples. Hence, the surrogate models constructed as part of the UQ analysis is employed to improve the information gain and make incremental recommendation, should the possibility to add more experiments arise. The insight gained from this study has been extensively used in the follow-up UQ study with the CFD modeling of the same gasifier configuration.",
keywords = "Bench-scale experiments, Data-fitted surrogate models, Fluidized bed gasifier, Non-intrusive Bayesian uncertainty quantification, Reacting multiphase flows",
author = "Aytekin Gel and Mehrdad Shahnam and Subramaniyan, {Arun K.}",
year = "2017",
month = "4",
day = "15",
doi = "10.1016/j.powtec.2017.01.034",
language = "English (US)",
volume = "311",
pages = "484--495",
journal = "Powder Technology",
issn = "0032-5910",
publisher = "Elsevier",

}

TY - JOUR

T1 - Quantifying uncertainty of a reacting multiphase flow in a bench-scale fluidized bed gasifier

T2 - A Bayesian approach

AU - Gel, Aytekin

AU - Shahnam, Mehrdad

AU - Subramaniyan, Arun K.

PY - 2017/4/15

Y1 - 2017/4/15

N2 - Adequate assessment of the uncertainties is becoming an integral part of the engineering design. This study is based on several efforts that are underway to investigate the uncertainty quantification (UQ) methods that are applicable to multiphase flows. Although particular emphasis has been given on identifying, characterizing and managing uncertainties in computational fluid dynamics (CFD) simulations, experimental data used in the validation of the CFD model is as critical. Hence, the goal of this paper is to demonstrate application of non-intrusive Bayesian uncertainty quantification methodology in multiphase (gas-solid) flows with experimental data as part of our research efforts to determine the most suited approach for UQ. For this purpose, a bench scale fluidized bed gasifier experiment from a prior study is used to demonstrate the applicability of Bayesian techniques on sparse data. Global sensitivity analysis performed as part of the UQ study shows that among the three operating factors, steam to oxygen ratio has the most influence on syngas composition in the gasifier. An analysis for forward propagation of uncertainties shows that an increase in steam to oxygen ratio leads to an increase in H2 mole fraction and a decrease in CO mole fraction. These findings are in agreement with the reference experimental study. Another contribution in addition to the UQ analysis is the optimization-based approach employed to identify next best set of additional experimental samples. Hence, the surrogate models constructed as part of the UQ analysis is employed to improve the information gain and make incremental recommendation, should the possibility to add more experiments arise. The insight gained from this study has been extensively used in the follow-up UQ study with the CFD modeling of the same gasifier configuration.

AB - Adequate assessment of the uncertainties is becoming an integral part of the engineering design. This study is based on several efforts that are underway to investigate the uncertainty quantification (UQ) methods that are applicable to multiphase flows. Although particular emphasis has been given on identifying, characterizing and managing uncertainties in computational fluid dynamics (CFD) simulations, experimental data used in the validation of the CFD model is as critical. Hence, the goal of this paper is to demonstrate application of non-intrusive Bayesian uncertainty quantification methodology in multiphase (gas-solid) flows with experimental data as part of our research efforts to determine the most suited approach for UQ. For this purpose, a bench scale fluidized bed gasifier experiment from a prior study is used to demonstrate the applicability of Bayesian techniques on sparse data. Global sensitivity analysis performed as part of the UQ study shows that among the three operating factors, steam to oxygen ratio has the most influence on syngas composition in the gasifier. An analysis for forward propagation of uncertainties shows that an increase in steam to oxygen ratio leads to an increase in H2 mole fraction and a decrease in CO mole fraction. These findings are in agreement with the reference experimental study. Another contribution in addition to the UQ analysis is the optimization-based approach employed to identify next best set of additional experimental samples. Hence, the surrogate models constructed as part of the UQ analysis is employed to improve the information gain and make incremental recommendation, should the possibility to add more experiments arise. The insight gained from this study has been extensively used in the follow-up UQ study with the CFD modeling of the same gasifier configuration.

KW - Bench-scale experiments

KW - Data-fitted surrogate models

KW - Fluidized bed gasifier

KW - Non-intrusive Bayesian uncertainty quantification

KW - Reacting multiphase flows

UR - http://www.scopus.com/inward/record.url?scp=85013223505&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85013223505&partnerID=8YFLogxK

U2 - 10.1016/j.powtec.2017.01.034

DO - 10.1016/j.powtec.2017.01.034

M3 - Article

AN - SCOPUS:85013223505

VL - 311

SP - 484

EP - 495

JO - Powder Technology

JF - Powder Technology

SN - 0032-5910

ER -