Machine Learning in Heat Transfer: Taxonomy, Review and Evaluation

Abstract

In the field of heat transfer, machine learning (ML) is used to analyze the large amounts of data that are collected through experiments, field observations, and simulations. It’s important to write a review paper that looks at how ML techniques are used in different heat transfer applications. We made a standard database with 900 publications for systematic reviews. So, the main goal of this review is to show a systematic state-of-the-art by analyzing how well ML works in heat transfer applications using PRISMA guidelines. Based on the results, most studies used the correlation coefficient as the most reliable and overall way to judge the ML tools in different heat transfer applications. Also, the Decision Tree (DT), the Random Forest (RF), and the Artificial Neural Network (ANN) have the most uses. On the other hand, the best performance is when people work together and use hybrid ML techniques. We’ll also publish and keep updating the latest research results so we can keep up with how quickly technology changes.