Performance analysis and machine learning algorithms of parabolic trough solar collectors using Al2O3-MWCNT as a hybrid nanofluid

Abstract

The critical importance of developing solar energy technologies for sustainable power generation places a high priority on the crucial efforts to optimize the performance of parabolic trough solar collector (PTSC) systems. This study focuses on optimizing the volume frictions of the hybrid nanofluids (1% Al2O3− 2% MWCNT/Syltherm-800, 1.5% Al2O3− 1.5% MWCNT/Syltherm-800, and 2% Al2O3− 1% MWCNT/Syltherm-800) in order to attain the most effective performance of the PTSC system. A mathematical model to investigate the PTSC performance under various volume concentrations is developed. Furthermore, this study trained three machine learning models (Decision Tree, Support Vector Machine, and Artificial Neural Network) using the generated data from the developed mathematical model to predict the PTSC outlet temperature more quickly, with the goal of achieving higher accuracy with fewer inputs. The findings of this study indicate that the PTSC system exhibits higher thermal efficiency when utilizing a combination of 2% Al2O3 and 1% MWCNT/Syltherm-800, with an average thermal efficiency of 70.54%. Moreover, the Artificial Neural Network (ANN) was the most accurate model out of the three. It performed remarkably well, showing an astounding R2 value of 99.99%, a mean absolute percentage error (MAPE) of 4.8x10−5, a root mean square error (RSME) of 0.012, and a mean absolute error (MAE) of 0.0057.