RAD-NNET, a neural network based correlation developed for a realistic simulation of the non-gray radiative heat transfer effect in three-dimensional gas-particle mixtures

Abstract

A neural network correlation, RAD-NNET, is developed to simulate the realistic effect of non-gray radiative absorption by a homogeneous mixture of combustion gases (CO 2 and H 2O) and soot using numerical data generated by RADCAL. RAD-NNET is then applied to assess the accuracy of some commonly accepted approximate approaches to evaluate radiative heat transfer in three-dimensional non-gray media. Results show that there are significant errors associated with the current approximate approaches. RAD-NNET can be readily implemented in commercial CFD codes to greatly enhance the accuracy of simulation of radiative heat transfer in practical engineering systems.