The current study begins with an experimental investigation focused on measuring the pressure drop of a water–air mixture under different flow conditions in a setup consisting of horizontal smooth tubes. Machine learning (ML)-based pipelines are then implemented to provide estimations of the pressure drop values employing obtained dimensionless features. Subsequently, a feature selection methodology is employed to identify the key features, facilitating the interpretation of the underlying physical phenomena and enhancing model accuracy. In the next step, utilizing a genetic algorithm-based optimization approach, the preeminent machine learning algorithm, along with its associated optimal tuning parameters, is determined. Ultimately, the results of the optimal pipeline provide a Mean Absolute Percentage Error (MAPE) of 5.99% on the validation set and 7.03% on the test. As the employed dataset and the obtained optimal models will be opened to public access, the present approach provides superior reproducibility and user-friendliness in contrast to existing physical models reported in the literature, while achieving significantly higher accuracy.

The present study is focused on identifying the most suitable sequence of machine learning-based models and the most promising set of input variables aiming at the estimation of heat transfer in evaporating R134a flows in microfin tubes. Utilizing the available experimental data, dimensionless features representing the evaporation phenomena are first generated and are provided to a machine learning-based model. Feature selection and algorithm optimization procedures are then performed. It is shown that the implemented feature selection method determines only six dimensionless parameters ($S{u}_l$: liquid Suratman number, $Bo$: boiling number, $F{r}_g$: gas Froude number, $R{e}_l$: liquid Reynolds number, $Bd$: Bond number, and $e/D$: fin height to tube’s inner diameter ratio) as the most effective input features, which reduces the model’s complexity and facilitates the interpretation of governing physical phenomena. Furthermore, the proposed optimized sequence of machine learning algorithms (providing a mean absolute relative difference (MARD) of 8.84% on the test set) outperforms the most accurate available empirical model (with an MARD of 19.7% on the test set) by a large margin, demonstrating the efficacy of the proposed methodology.