Machine learning algorithms to predict flow condensation heat transfer coefficient in mini/micro-channel utilizing universal data

Advances in micro and nanoengineered surfaces for enhancing boiling and condensation heat transfer: a review

“…Supervised ML models such as Articial Neural Network (ANN), Decision Tree, Random Forest, K-nearest neighbors regression (KNN-regression), Adaptive boosting (AdaBoost), and Support Vector Machine (SVM) have been developed as regression models to estimate the pressure drop and heat transfer coefficient in internal boiling and condensation processes. [284][285][286][287][288][289] Qiu et al built an ANN model using 16953 data points from 50 sources to predict internal ow boiling heat transfer in mini/micro-channels. 286 The best ANN model had 7 hidden layers with number of units at each layer varying from 10 to 75 and it was trained only on dimensionless parameters including bond number (Bd), boiling number (Bo), convection number (Co), Froude number (Fr), Peclet number (Pe), Prandtl number (Pr), Reynolds number (Re), Suratman number (Su), and Weber number (We).…”

Review of Machine Learning for Hydrodynamics, Transport, and Reactions in Multiphase Flows and Reactors

“…Fairly good accuracy of estimations for the data sets over a wide range of operating conditions was achieved. Zhou et al compared different ML algorithms including ANN, Adaptive Boosting (AdaBoost), RF, and XGBoost in order to evaluate their prediction performance for flow condensation heat transfer in mini/microchannels. In particular, the relative importance of extensive feature inputs was also explored in detail.…”

Fluid flow and heat transfer in microchannel heat sinks: Modelling review and recent progress

“…With the utilization of deep learning, the proposed model can successfully predict the heat transfer performance in helical coils, and especially achieved excellent performance in predicting outputs that have a very large range of value differences. Zhou et al [121] applied four machine learning-based models, i.e. ANNs, Random Forest, AdaBoost, and Extreme Gradient Boosting (XGBoost), to predict condensation heat transfer coefficients in mini/microchannel with a consolidated database of 4,882 data points and compared for predicting accuracy.…”