A data driven artificial neural network model for predicting radiative properties of metallic packed beds. (March 2019)

• Record Type: Journal Article
• Authors: Kang, Hyun Hee; Kaya, Mine; Hajimirza, Shima
• Is Part Of: Journal of quantitative spectroscopy & radiative transfer. Volume 226(2019)
• Page Start: 66
• Page End: 72
• Publication Date: 2019-03
• Subjects: Porous media -- Metallic packed bed -- Radiative heat transfer -- Monte Carlo ray tracing -- Neural networks; Spectrum analysis -- Periodicals; Radiation -- Periodicals; Analyse spectrale -- Périodiques; Rayonnement -- Périodiques; Radiation; Spectrum analysis; Periodicals; 543.0858
• DOI: 10.1016/j.jqsrt.2019.01.013
• Languages: English
• ISSNs: 0022-4073

Abstract

Highlights: A neural network based surrogate model for parametric modeling of radiative heat transfer in metallic packed bed is developed. The Monte Carlo ray tracing simulation approximates transmissivity of metal packed bed with respect to bed geometry and material property. The training of neural network model with transmission output from Monte Carlo ray tracing simulation is proceeded. Neural network model coupled with Monte Carlo ray tracing model accurately and efficiently approximate radiative properties of metal packed beds. Abstract: In this work, we demonstrate that machine learning methods can be reliably used to predict radiative properties of dispersed media, i.e. packed beds, as a function of packed bed geometry and material properties. The computationally expensive Monte Carlo ray tracing (MCRT) method, which is widely used in this context, is replaced by Neural Networks (NN). We demonstrate that the data-driven surrogate prediction works accurately and generally. The results of both MCRT and NN models agree well with each other and with previously measured literature results. We also measure the uncertainty of the NN results using statistical methods. It is recommended that the developed model be used for efficient inverse problems and optimizations in relevant future work.