Machine learning for heat transfer correlations

Beomjin Kwon, Faizan Ejaz, Leslie K. Hwang

Research output: Contribution to journalArticlepeer-review

49 Scopus citations


This paper explores machine learning approach as a heat transfer correlation. Machine learning significantly reduces the effort to develop multi-variable heat transfer correlations, and is capable of readily expanding the parameter domain. Random forests algorithm is used to predict the convection heat transfer coefficients for a high-order nonlinear heat transfer problem, i.e., convection in a cooling channel integrated with variable rib roughness. For 243 different rib array geometries, numerical simulations are performed to train and test ML model based on six input features. Machine learning model predicts closely to numerical simulation data with high determination of coefficient (R2), e.g., R2 > 0.966 for the testing dataset. The capability and limitation of random forests algorithm are discussed with validation dataset.

Original languageEnglish (US)
Article number104694
JournalInternational Communications in Heat and Mass Transfer
StatePublished - Jul 2020


  • Heat transfer correlation
  • Machine learning
  • Random forests
  • Variable rib roughness

ASJC Scopus subject areas

  • Atomic and Molecular Physics, and Optics
  • Chemical Engineering(all)
  • Condensed Matter Physics


Dive into the research topics of 'Machine learning for heat transfer correlations'. Together they form a unique fingerprint.

Cite this