This paper is NOT THE PUBLISHED VERSION; but the author's final, peer-reviewed manuscript. The published version may be accessed by following the link in the citation below.
Machine learning (ML) offers a variety of techniques to understand many complex problems in different fields. The field of heat transfer, and thermal systems in general, are governed by complicated sets of governing physics that can be made tractable by reduced-order modeling, and by extracting simple trends from measured data. Therefore, ML algorithms can yield computationally efficient models for more accurate predictions or to generate robust optimization frameworks. This study reviews past and present efforts that use ML techniques in heat transfer from the fundamental level to full-scale applications, including the use of ML to build reduced-order models, predict heat transfer coefficients and pressure drop, real-time analysis of complex experimental data, and optimize large-scale thermal systems in a variety of applications. The appropriateness of different data-driven ML models in heat transfer problems is discussed. Finally, some of the imminent opportunities and challenges that the heat transfer community faces in this exciting and rapidly growing field are identified.
The complex variable boundary element method or CVBEM is a numerical technique that can provide solutions to potential value problems in two or more dimensions by the use of an approximation function that is derived from the Cauchy integral equation in complex analysis. Given the potential values (i.e. a Dirichlet problem) along the boundary, the typical problem is to use the potential function to solve the governing Laplace equation. In this approach, it is not necessary to know the streamline values on the boundary. The modeling approach can be extended to problems where the streamline function is needed because there are known streamline values along the problem boundary (i.e. a mixed boundary value problem). Two common problems that have such conditions are insulation on a boundary and fluid flow around a solid obstacle. In this paper, five advances in the CVBEM are made with respect to the modeling of the mixed boundary value problem; namely (1) the use of Mathematica and Matlab in tandem to calculate and plot the flow net of a boundary value problem. (2) The magnitude of the size of the problem domain is extended. (3) The modeling results include direct computation and development of a flow net. (4) The graphical displays of the total flownet are developed simultaneously. And (5) the nodal point location as an additional degree of freedom in the CVBEM modeling approach is extended to mixed boundaries. A demonstration problem of fluid flow is included to illustrate the flownet development capability.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.