Point Mean Beam Length - a New Concept to Enhance the Computational Efficiency of Multi- Dimensional Non-Gray Radiative Heat Transfer

Published

December 3, 2020

Author(s)

Walter W. Yuen, Andy Tam

Abstract

A new concept of point mean beam length (PMBL) is introduced. For enclosures with simple geometry, this concept provides a fundamental self-consistent interpretation on the various different definition of the conventional mean beam length. The concept is further demonstrated to be effective in enhancing the computational efficiency for multi-dimensional radiative heat transfer in non-gray media. In the evaluation of radiative exchange between two perpendicular areas with a common edge, the use of PMBL leads to a factor of 100 to 400 reduction in computational effort compared to the direct integration approach. For practical applications, PMBL is combined with RADNNET (a neural network correlation for a one-dimensional CO2/H2O/soot combustion mixture) to generate two highly efficient and accurate solvers for the evaluation of exchange factors between two parallel or perpendicular rectangular areas of arbitrary dimensions with an intervening combustion mixture.

Citation

Numerical Heat Transfer Part A-Applications

Pub Type

Journals

Download Paper

Local Download

Keywords

mean beam length, non-gray, multi-dimensional, radiative heat transfer, neural network

Citation

Yuen, W. and Tam, A. (2020), Point Mean Beam Length - a New Concept to Enhance the Computational Efficiency of Multi- Dimensional Non-Gray Radiative Heat Transfer, Numerical Heat Transfer Part A-Applications, [online], https://tsapps.nist.gov/publication/get_pdf.cfm?pub_id=930741 (Accessed October 10, 2025)