An adaptive angular quadrature for the discrete transfer method based on error estimation

The ray effect is a serious problem in radiative heat transfer computations. Continuously varying radiation fields are approximated numerically by sampling a limited number of angular directions. The discrete transfer method (DTM) is a conceptually simple technique suitable for general-purpose calcu...

Full description

Saved in:
Bibliographic Details
Main Authors: Hendrik Versteeg, Jonathan C. Henson, Weeratunge Malalasekera
Format: Default Conference proceeding
Published: 2001
Subjects:
Online Access:https://hdl.handle.net/2134/5513
Tags: Add Tag
No Tags, Be the first to tag this record!
id rr-article-9547412
record_format Figshare
spelling rr-article-95474122001-01-01T00:00:00Z An adaptive angular quadrature for the discrete transfer method based on error estimation Hendrik Versteeg (1248945) Jonathan C. Henson (7201703) Weeratunge Malalasekera (1258755) Mechanical engineering not elsewhere classified untagged Mechanical Engineering not elsewhere classified The ray effect is a serious problem in radiative heat transfer computations. Continuously varying radiation fields are approximated numerically by sampling a limited number of angular directions. The discrete transfer method (DTM) is a conceptually simple technique suitable for general-purpose calculations of thermal radiation in complex geometries. Over the years a large variety of quadratures based on fixed ray firing patterns has been suggested for use in conjunction with the DTM and recently an adaptive quadrature has been proposed by Cumber (2000). Arguably, in absence of a comprehensive error analysis, the efficacy of all these quadratures has only been proved for limited collections of radiation problems. In recent work we have established sharp error bounds for the heat flux integral in the DTM for irradiation distributions of three different continuity classes: smooth fields, fields with discontinuous angular derivatives and piecewise constant fields (Versteeg et al, 1999a,b). The resulting error formulae have paved the way for a new adaptive quadrature strategy. We show results of its application to an idealised jet flame and to radiative exchanges inside a cube-shaped enclosure. We also briefly comment on the viability of this approach in general-purpose CFD/radiation computations. Our work demonstrates that the new adaptive angular quadrature has the following capabilities: • Evaluation of DTM heat flux integrals to a pre-specified tolerance for sufficiently smooth intensity distributions. • Excellent accuracy with very low ray numbers for irradiation due small view factor sources. • Good heat flux estimates for piecewise constant sources, provided that the truncation criterion is slightly adjusted and care is taken in specifying the starting mesh. 2001-01-01T00:00:00Z Text Conference contribution 2134/5513 https://figshare.com/articles/conference_contribution/An_adaptive_angular_quadrature_for_the_discrete_transfer_method_based_on_error_estimation/9547412 CC BY-NC-ND 4.0
institution Loughborough University
collection Figshare
topic Mechanical engineering not elsewhere classified
untagged
Mechanical Engineering not elsewhere classified
spellingShingle Mechanical engineering not elsewhere classified
untagged
Mechanical Engineering not elsewhere classified
Hendrik Versteeg
Jonathan C. Henson
Weeratunge Malalasekera
An adaptive angular quadrature for the discrete transfer method based on error estimation
description The ray effect is a serious problem in radiative heat transfer computations. Continuously varying radiation fields are approximated numerically by sampling a limited number of angular directions. The discrete transfer method (DTM) is a conceptually simple technique suitable for general-purpose calculations of thermal radiation in complex geometries. Over the years a large variety of quadratures based on fixed ray firing patterns has been suggested for use in conjunction with the DTM and recently an adaptive quadrature has been proposed by Cumber (2000). Arguably, in absence of a comprehensive error analysis, the efficacy of all these quadratures has only been proved for limited collections of radiation problems. In recent work we have established sharp error bounds for the heat flux integral in the DTM for irradiation distributions of three different continuity classes: smooth fields, fields with discontinuous angular derivatives and piecewise constant fields (Versteeg et al, 1999a,b). The resulting error formulae have paved the way for a new adaptive quadrature strategy. We show results of its application to an idealised jet flame and to radiative exchanges inside a cube-shaped enclosure. We also briefly comment on the viability of this approach in general-purpose CFD/radiation computations. Our work demonstrates that the new adaptive angular quadrature has the following capabilities: • Evaluation of DTM heat flux integrals to a pre-specified tolerance for sufficiently smooth intensity distributions. • Excellent accuracy with very low ray numbers for irradiation due small view factor sources. • Good heat flux estimates for piecewise constant sources, provided that the truncation criterion is slightly adjusted and care is taken in specifying the starting mesh.
format Default
Conference proceeding
author Hendrik Versteeg
Jonathan C. Henson
Weeratunge Malalasekera
author_facet Hendrik Versteeg
Jonathan C. Henson
Weeratunge Malalasekera
author_sort Hendrik Versteeg (1248945)
title An adaptive angular quadrature for the discrete transfer method based on error estimation
title_short An adaptive angular quadrature for the discrete transfer method based on error estimation
title_full An adaptive angular quadrature for the discrete transfer method based on error estimation
title_fullStr An adaptive angular quadrature for the discrete transfer method based on error estimation
title_full_unstemmed An adaptive angular quadrature for the discrete transfer method based on error estimation
title_sort adaptive angular quadrature for the discrete transfer method based on error estimation
publishDate 2001
url https://hdl.handle.net/2134/5513
_version_ 1794843704142331904