Key points are not available for this paper at this time.
This study investigates the exergy efficiency of a hybrid solar collector using water and water-based titanium dioxide (TiO 2 ) nanofluids, employing advanced machine learning (ML) models to optimize performance evaluation. Support Vector Regression (SVR), Random Forest (RF), and a hybrid approach incorporating Wavelet Transform (WT) were utilized to assess the system's efficiency. Three statistical metrics, such as mean absolute error (MAE), coefficient of determination (R 2 ), and root mean square error (RMSE), denoted as E1, E2, and E3 respectively, were used to evaluate model performance. Two experimental setups were implemented: the first involved water flow rates of 0.5, 1.0, and 1.5 liters per minute, while the second employed a water-based TiO 2 nanofluid with a 0.1% volume concentration. Results indicate a direct correlation between increased mass flow rates and enhanced exergy efficiency, with energy efficiency ranging from 7.1% to 11.1% for water, and 12.8% to 20.4% for the TiO 2 nanofluid. The integration of WT with ML models significantly improved predictive accuracy, achieving final metrics of 0.874 (E1), 2.212 (E2), and 3.118 (E3). Wind speed, ambient temperature, and solar radiation were identified as critical factors influencing system performance, with hybrid models outperforming individual ML models in both accuracy and reliability.
Natesan et al. (Tue,) studied this question.