Deep neural network-based heat radiation modelling between particles and between walls and particles

Josef Franz Viktor Tausendschön*, Stefan Radl

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review


We present a Deep Neural Network (DNN)-based view factor model to calculate radiative heat transfer rates between particles, as well as between particles and walls in Discrete Element Method (DEM)-based simulations. A systematic analysis of the most promising markers available in DEM simulations to be correlated with the view factor is performed. Subsequently, a neural network is trained, and its predictive performance is analyzed. View factors are studied for a variety of systems ranging from dilute particle systems to dense (i.e., settled under gravity) particle beds. It is demonstrated that the trained DNN-model can model view factors at higher accuracy and with significantly less computational effort than other literature models. A validation with experimental data is provided, showing that the implemented model can predict the total heat flux, as well as particle temperatures in a packed bed accurately.

Original languageEnglish
Article number121557
JournalInternational Journal of Heat and Mass Transfer
Publication statusPublished - Oct 2021


  • Deep Neural Networks
  • Discrete Element Method (DEM)
  • Machine learning
  • Radiation modelling
  • Radiative heat transfer
  • View factors
  • Wall radiation

ASJC Scopus subject areas

  • Condensed Matter Physics
  • Mechanical Engineering
  • Fluid Flow and Transfer Processes

Cite this