An imperative need for machine learning algorithms in heat transfer application: a review

Abstract

In recent years, modeling of heat exchanger is increased due to transient prediction, optimization, and performance calculations. Nanofluids play a vital role in increasing heat transfer performance of heat exchangers. This review gives an open knowledge on predicting heat transfer performance of various heat exchanger with nanofluid as coolant using various machine learning techniques. Machine learning is a promising data-driven approach for estimating heat exchanger parameters through regression classification, demonstrating promising prediction capabilities. This review article provides exemplary guidance on selecting suitable model to predict important criteria such as heat transfer coefficient, Nusselt number, overall heat transfer performance, and provides restrictions, and loopholes of machine learning techniques for heat transfer applications.