Liquid-to-vapor phase change heat transfer evaluation and parameter sensitivity analysis of nanoporous surface coatings

Abstract

The goal of this research is to create a deep learning approach coupled with explainable artificial intelligence (DL-XAI) to evaluate the liquid-to-vapor phase change heat transfer in nanoporous surface coatings. Over the years, nanoporous coated surfaces have been employed to increase the efficacy of the liquid-to-vapor phase change heat transfer process. Despite a considerable body of experimental data on the pool boiling of coated nanoporous surfaces, the pool boiling phenomenon of these surfaces with various working fluids is poorly understood. In this study, random forest-based optimization (RF) has been used to tune the hyper-parameters for the DL model. The optimized model can predict the desired parameter with an R2 = 0.99. Besides, the parametric analysis has been performed to highlight the impact of each parameter on the output parameter. Through XAI, heat flux, conductivity of the substrate material and the pore diameter are unveiled to be the most influencing surface features for the nanoporous coatings. The proposed method can be used to optimize nanoporous coated surfaces for various working fluids.