Session: K20-01: APPLICATIONS OF MACHINE LEARNING/AI FOR HEAT TRANSFER

Paper Number: 130465

130465 - Temperature Field Reconstruction of Surfaces Heated Through Radiative Heat Transfer Using Convolutional Neural Networks 

Abstract: 

Diverse applications require us to know the measure of a given physical quantity in real time. For many applications, we can rely on probes to fulfill this task. However, this approach can be challenging when we need to know these quantities in many locations over a given domain due to the spatial limitation involved. In this context, using Convolutional Neural Networks (CNN) can offer a good option to deal with such a problem. A well-trained CNN can reconstruct the entire distribution of a given physical quantity over a domain using only a few probes, allowing us to retrieve the desired distribution even in a limited space or complex geometries where adding a large number of probes could be impossible.

In this work, we will present the initial approach to developing a real-time tool for the thermal monitoring of nuclear reactor pressure vessels. Based on an experimental setup, we developed a computational model using the Multiphysics Object-Oriented Simulation Environment (MOOSE) framework, where both the Ray Tracing and Heat conduction modules were used to evaluate the temperature distribution over a non-planar metal surface heated through radiative heat transfer. We verified our computational model against the available experimental data. We used part of the data generated by the MOOSE model to train the Convolutional Neural Network responsible for the temperature reconstruction over the vessel. The CNN generalization was then verified against the experimental and computational data, and its performance metrics are presented in this work.

Presenting Author: Luiz Aldeia Machado The Pennsylvania State University

Presenting Author Biography: He has a bachelor's degree in Nuclear Engineering from the Federal University of Rio de Janeiro - Brazil (2021) and is currently pursuing his Ph.D. in Nuclear Engineering at the Pennsylvania State University. His areas of interest are multi-physics simulations, CFD, and computational thermal hydraulics.

Authors:

Luiz Aldeia Machado The Pennsylvania State University
Victor Coppo Leite The Pennsylvania State University
Elia Merzari The Pennsylvania State University
Lesley Wright Texas A&M University
Lander Ibarra Argonne National Laboratory
Roberto Ponciroli Argonne National Laboratory
Pramatha Bhat Texas A&M University
Yassin Hassan The Texas A&M University