Search

Surface radiation heat transfer may become a dominant mode of heat transfer in many engineering applications, and its accurate and efficient modeling is the key to the efficient design of the involved systems. In the study, an improved discrete transfer method (DTM) is developed to meet practical needs for surface radiation modeling. A direct application of the DTM to modeling surface radiative heat transfer usually leads to a large error, due to the strong ray effects. In order to eliminate the ray effects, a total of nine discrete directions over a control angle instead of a single discrete direction as used in the traditional DTM is selected, and the incident radiative intensity over this control angle is then replaced by a weighted-averaged intensity from these nine discrete directions. This approach provides a more accurate value on the mean radiative intensity over a control angle, while it does not increase the total number of radiative intensity to be solved at a surface element. Compared to the modified DTM we developed previously, the present approach is much simpler in terms of numerical treatment. In order to examine the accuracy of the present method, two 3-D benchmark problems are selected and investigated. With a reasonably fine angular discretization, the present solutions are found to match the exact solutions and they are much more accurate than the results from the traditional DTM. [ABSTRACT FROM AUTHOR]

Radiative heat transfer is the most dominant mode of heat transfer in an optical fiber drawing process, and its accurate prediction is the key to a realistic simulation of flow and heat transfer in the system. In this study, the finite-volume method (FVM ) is developed to model radiative heat transfer in a gas enclosure as well as inside a glass preform with a Fresnel boundary at the gas–glass interface. Unlike diffuse boundaries, the reflections and transmissions at Fresnel boundaries are governed by Fresnel’s relation and Snell’s law, and they are strongly dependent on the angular direction. During the implementation of the FVM in semitransparent media, control-angle overlap may occur, and it is treated by assuming that the flux of radiant energy across a control angle is determined only by the discrete direction over this control angle and the corresponding radiative intensity. At the Fresnel boundary, a reflective or refractive intensity may not coincide with any discrete direction, and it is approximated by an intensity whose discrete direction makes the smallest angle with the corresponding reflective or refractive intensity. To validate the present FVM in semitransparent media with Fresnel boundaries, two benchmark problems are examined and the present solutions are found to be very accurate with a fine angular discretization. After that, the present model is applied to investigate an optical fiber drawing process with different boundary conditions and temperature distributions. For the diffuse boundary at the gas–preform interface, the results from the FVM are very close to those from the zonal method in each case. For the Fresnel boundary, the FVM provides a solution that tends to be higher in the upper neck-down region and lower in the low neck-down region than that from the same method with the diffuse boundary. The maximum difference betweentwo solutions varies and may reach 10–15% at some cases. [ABSTRACT FROM AUTHOR]

An ideal surface radiation model applied in many manufacturing and materials processing systems must be able to take into account specular, spectral, and shadowing effects with complex geometries, and must be computationally efficient to permit its inclusion in the fluid flow and heat transfer models. In this study, a novel surface radiative heat transfer method is developed to meet all these practical needs. The present model is based on the discrete transfer method (DTM). A direct application of the DTM to modeling surface radiative heat transfer may result in a large error due to strong ray effects. In order to eliminate these ray effects, the DTM is modified by considering radiation contribution from all surface cells intercepted by a control angle. Calculation of these surface cell areas represents one of the most important tasks in the modified DTM (MDTM), and it is described in detail in this study. To investigate the accuracy and efficiency of the MDTM, three benchmark problems covering different geometric and boundary conditions are considered, and the present solution is compared with the solutions from the exact approach, the DTM, and discrete ordinates method (DOM). For each problem, the accuracy of the MDTM, DTM, and DOM appears to be affected by the angular discretization. For a reasonable fine angular discretization, the solutions from the MDTM and DOM match the exact solution very well, while the solution from the DTM usually shows strong ray effects. The CPU times spent on the MDTM and DTM are very similar, but they are usually orders of magnitude less than that for the DOM. The present study indicates that the MDTM is not only accurate but also very efficient for modeling complicated surface radiation problems. Such a model will greatly benefit the simulation of many manufacturing and materials processing systems. [ABSTRACT FROM AUTHOR]

A spatial domain-based parallel algorithm is developed for simulating radiative heat transfer in a distributed computing environment. The radiative transfer equation is solved using an unstructured finite-volume method that is applicable for any 2D planar, axisymmetric, and 3D problems with structured, unstructured, or hybrid grids. The domain decomposition is carried out by equally partitioning the spatial domain into many subdomains along the longer geometric dimension. Communication among the subdomains on each processor is performed through a message-passing interface library. In order to examine the parallel performance of the unstructured radiation code, two benchmark problems are investigated for different absorption coefficients, scattering coefficients, and grid sizes in a parallel computer. To help us understand the change of parallel performance, a new parameter, the total inner iteration number, is introduced to analyze the results. For all the cases examined, as expected, the parallel performance is seen to degrade rapidly with an increase of the processor number. However, in contrast with other studies, the parallel performance is found to degrade with an increase of absorption coefficient for a temperature-prescribed problem. Also, the global iteration number is found to be not necessarily independent of the grid size. [ABSTRACT FROM AUTHOR]