| Autor: |
Ibrahim, M. I., Abdullah, M. Z., Kulla, D. M., Umaru, S., Dalhatu, A. |
| Předmět: |
|
| Zdroj: |
AIP Conference Proceedings; 2023, Vol. 2959 Issue 1, p1-16, 16p |
| Abstrakt: |
Renewable energies, specifically solar energy which is the richest renewable energy on earth, has been employed in numerous applications due to the fact that it is pollutant-free, widely available and inexhaustible. The main thrust of this study is to predict thermal and electrical efficiencies of photovoltaic-thermal (PV/T) setups in regard with load power and solar radiation by developing an adaptive neuro fuzzy inference system (ANFIS) and multiple linear regression (MLR) models. To achieve this goal, about 266 empirical measurements were performed on two different fabricated PV/T setups (conventional PV and a water-cooled PV/T system). Several statistical analyses were carried out to assess and validate the proposed models. Base on the results obtained, it was confirmed that there was an excellent agreement between the predicted model outputs and the actual experimental data. However, the ANFIS model outperformed MLR model due lowest MSE with a value of (19.500 for PV/T thermal efficiency, 0.0440 for PV/T electrical efficiency and 0.0021 for PV electrical efficiency). In addition, the lowest RMSE (4.4158 for PV/T thermal efficiency, 0.2097 for PV/T electrical efficiency and 0.0089 for PV electrical efficiency) and the highest correlation of determination (R2) (0.9003 for PV/T thermal efficiency, 0.9972 for PV/T electrical efficiency and 0.9998 for PV electrical efficiency) respectively. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
