Estimating the density of hybrid nanofluids for thermal energy application: Application of non-parametric and evolutionary polynomial regression data-intelligent techniques

There is no doubt that density is one of the most crucial thermophysical properties of hybrid nanofluids in thermal energy applications. Various research papers have been devoted to thermophysical properties of various hybrid nanofluids. However, a few of them focused on the simultaneous effects of nanoparticles, base fluids, and other factors on the density of hybrid nanofluids. In this research, a comparative study was conducted on non parametric and evolutionary machine learning paradigms, namely, Multivariate Adaptive Regression Spline (MARS) and Evolutionary Polynomial Regression (EPR) models to accurately predict the density of a wide variety type of nanofluids in thermal energy applications. Here, for providing the predictive models, 501 data points were collected from the reliable recent literature. Besides, the Gene Expression Programming (GEP) and Multivariate Linear Regression (MLR) models were examined for validating the outcomes of MARS and EPR models. The comprehensive assessment demonstrated that the MARS outperformed the other models.

Automated summary

In this study, different machine learning models were compared to predict the density of various nanofluids in thermal energy applications. The results demonstrated that the MARS model outperformed the others.