A numerical and experimental comparison of heat transfer in a quasi two-dimensional packed bed--颗粒学报PARTICUOLOGY

Tyslik, M., Ebert, M., Schiemann, M., Wirtz, S., & Lessig, C. (2024). A numerical and experimental comparison of heat transfer in a quasi two-dimensional packed bed. Particuology, 84, 136-144. https://doi.org/10.1016/j.partic.2023.03.014

A numerical and experimental comparison of heat transfer in a quasi two-dimensional packed bed

Matthias Tyslik ^a *, Mirko Ebert ^b, Martin Schiemann ^a, Siegmar Wirtz ^a, Christian Lessig ^b

a Department of Energy Plant Technology, Ruhr-University Bochum, Germany
b Institute for Simulation and Graphics, Otto-von-Guericke-Universität Magdeburg, Germany

10.1016/j.partic.2023.03.014

Volume 84, January 2024, Pages 136-144

Received 30 December 2022, Revised 9 March 2023, Accepted 17 March 2023, Available online 5 April 2023, Version of Record 12 April 2023.

E-mail: tyslik@leat.rub.de

Highlights

Abstract

Heat transfer plays a major role in many industrial processes taking place in packed beds. An accurate and reliable simulation of the heat exchange between particles is therefore crucial for a reliable operation and to optimize the processes in the bed. The discrete ordinates method (DOM) provides an established numerical technique to model radiative heat transfer in granular media that offers the possibility to consider the directional dependence of the radiation propagation. In this work, DOM is compared with Monte Carlo ray tracing, which provides an alternative method for heat transfer simulations. Geometrically simple configurations are used to investigate the influence of the angular discretization on the accuracy of the results and the computation time in both methods. The obtained insights are then transferred to a more complex configuration of a quasi two-dimensional test rig consisting of metal rods for which also experimental results are available. Our results show that both DOM and Monte Carlo ray tracing allow for an accurate simulation of heat transfer in packed beds. Monte Carlo ray tracing requires thereby computation times that are surprisingly competitive (although still somewhat slower) compared to DOM and allows for an easier computation of highly accurate reference solutions. In our preliminary comparison to the experimental test rig, Monte Carlo ray tracing also provides the advantage that it can more easily model highly specular materials such as stainless steel. Both methods are comparable for diffuse materials such as magnesium oxide.

Graphical abstract

Keywords

Heat transfer; Discrete ordinates method; Monte Carlo ray tracing

文章导读