TY - JOUR
T1 - Enhancing Computational Efficiency in Porous Media Analysis
T2 - Integrating Machine Learning With Monte Carlo Ray Tracing
AU - Tabassum, Farhin
AU - Hajimirza, Shima
N1 - Publisher Copyright:
Copyright © 2024 by ASME.
PY - 2024/10/1
Y1 - 2024/10/1
N2 - Monte Carlo ray tracing (MCRT) is a prevalent and reliable computation method for simulating light-matter interactions in porous media. However, modeling these interactions becomes computationally expensive due to complex structures and enormous variables. Hence, machine learning (ML) models have been utilized to overcome computational burdens. In this study, we investigate two distinct frameworks for characterizing radiative properties in porous media for pack-free and packing-based methods. We employ two different regression tools for each case, namely Gaussian process (GP) regressions for pack-free MCRT and convolutional neural network (CNN) models for pack-based MCRT to predict the radiative properties. Our study highlights the importance of selecting the appropriate regression method based on the physical model, which can lead to significant computational efficiency improvement. Our results show that both models can predict the radiative properties with high accuracy (>90%). Furthermore, we demonstrate that combining MCRT with ML inference not only enhances predictive accuracy but also reduces the computational cost of simulation by more than 96% using the GP model and 99% for the CNN model.
AB - Monte Carlo ray tracing (MCRT) is a prevalent and reliable computation method for simulating light-matter interactions in porous media. However, modeling these interactions becomes computationally expensive due to complex structures and enormous variables. Hence, machine learning (ML) models have been utilized to overcome computational burdens. In this study, we investigate two distinct frameworks for characterizing radiative properties in porous media for pack-free and packing-based methods. We employ two different regression tools for each case, namely Gaussian process (GP) regressions for pack-free MCRT and convolutional neural network (CNN) models for pack-based MCRT to predict the radiative properties. Our study highlights the importance of selecting the appropriate regression method based on the physical model, which can lead to significant computational efficiency improvement. Our results show that both models can predict the radiative properties with high accuracy (>90%). Furthermore, we demonstrate that combining MCRT with ML inference not only enhances predictive accuracy but also reduces the computational cost of simulation by more than 96% using the GP model and 99% for the CNN model.
KW - convolutional neural network (CNN)
KW - Gaussian process
KW - machine learning (ML)
KW - Monte Carlo ray tracing
KW - porous media
KW - porous medium
KW - radiative heat transfer
KW - radiative properties
KW - thermophysical properties
UR - http://www.scopus.com/inward/record.url?scp=85200639711&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85200639711&partnerID=8YFLogxK
U2 - 10.1115/1.4065895
DO - 10.1115/1.4065895
M3 - Article
AN - SCOPUS:85200639711
SN - 1948-5085
VL - 16
JO - Journal of Thermal Science and Engineering Applications
JF - Journal of Thermal Science and Engineering Applications
IS - 10
M1 - 101003
ER -