Predictive analysis of heat transfer characteristics of nanofluids in helically coiled tube heat exchanger using regression approach

Abstract

Nanofluids are the combination of base fluid and nanoparticles which offer higher thermal conductivity resulting higher heat transfer. In this research article, soft computing tool is used to find the accurate Nusselt number of coiled tube heat exchanger handling Al2O3/H2O nanofluids at three different volume concentrations and at different mass flow rate in terms of Dean number. The input predictor variables used in this model are convective heat transfer coefficient, thermal conductivity of nanofluids, and Dean number and the output response variable is Nusselt number. Linear regression, generalized linear regression, and Lasso and elastic-net regularized generalized linear models methodologies are taken to predict the Nusselt number. It is observed that the linear regression method shows an accurate agreement with experimental data with root mean square error value of 0.05614 and regression coefficient value is 0.99. It is studied that the experimental data holds good accordance with the predicted data given by the trained network. The average relative errors in the prediction of Nusselt number and heat transfer coefficients are found to be 0.3% and 0.2%, respectively.