Publication Date
2023
Document Type
Dissertation/Thesis
First Advisor
Cho, Kyu Taek
Degree Name
M.S. (Master of Science)
Legacy Department
Department of Mechanical Engineering
Abstract
Heat exchangers (HX) are crucial components of thermal control systems which facilitate proper transfer of heat in systems such as nuclear reactors. This research focuses on the heat transfer-enhancing effect associated with turbulence in fluid flowing conduits. Turbulence can be induced at low flow rates when surface roughness is present on channel walls. Rough surfaces are typically regarded as a flawed product of the Additive Manufacturing (AM) process during the construction of Heat Exchangers. Machining surface finishes onto AM parts is a typical post-processing method used to remove the roughness. Thorough investigations have not yet been published on the effects of leaving the resulting AM surface roughness on HX thermo-hydraulic performance.This new direction is explored through the novel modeling approach of synergistically combining a physics-based model with data-based prediction models. A Computational Fluid Dynamics (CFD) model is combined with numerous Machine Learning (ML) algorithms to correlate roughness on channel walls with thermo-hydraulic performance. The ML algorithms included both models capable of both Regression and Classification; Logistic Regression, Linear Regression, Support Vector Machines, K-Nearest Neighbors, Decision Trees, and Random Forest models were included in a combined final ensemble method. This final model was utilized to create the relationships between parameters and produce parameter processing maps.
Recommended Citation
Platt, Kaylen A., "Analysis of the Effect of Surface Roughness in additive Manufacturing on Thermo-Hydraulic Performance by ML Models supported by Physics-Based Modeling" (2023). Graduate Research Theses & Dissertations. 7345.
https://huskiecommons.lib.niu.edu/allgraduate-thesesdissertations/7345
Extent
56 pages
Language
eng
Publisher
Northern Illinois University
Rights Statement
In Copyright
Rights Statement 2
NIU theses are protected by copyright. They may be viewed from Huskie Commons for any purpose, but reproduction or distribution in any format is prohibited without the written permission of the authors.
Media Type
Text
