Energy-exergy analysis and ANN-based performance prediction of a nano-coated PCM-integrated photovoltaic thermal system
APPLIED THERMAL ENGINEERING, cilt.302, 2026 (SCI-Expanded, Scopus)
- Yayın Türü: Makale / Tam Makale
- Cilt numarası: 302
- Basım Tarihi: 2026
- Doi Numarası: 10.1016/j.applthermaleng.2026.132008
- Dergi Adı: APPLIED THERMAL ENGINEERING
- Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Compendex, INSPEC, DIALNET, Business Source Ultimate (EBSCO)
- Anahtar Kelimeler: Artificial neural network (ANN), Energy–exergy analysis, Nano-enhancement, Photovoltaic–thermal (PVT), Thermal energy storage
- Gazi Üniversitesi Adresli: Evet
Özet
Photovoltaic-thermal (PVT) collectors simultaneously produce thermal and electrical energy, and their performance can be improved through thermal storage and absorber surface modifications. In this study, three doublepass V-grooved air-type PVT configurations were experimentally investigated: a conventional V-PVT system, a cetearyl alcohol-based PCM-integrated TS/V-PVT system, and a graphene nano-coated PCM-integrated NTS/VPVT system. The systems were tested under low and high airflow conditions, and their energy, exergy, electrical, and ANN-based prediction performances were evaluated. The results showed that the NTS/V-PVT configuration provided the best overall performance. Under high airflow conditions, the maximum average thermal efficiency reached 80.67% for NTS/V-PVT, compared with 54.10% for TS/V-PVT and 49.04% for V-PVT. The maximum average thermal exergy efficiency was obtained as 8.80% for NTS/V-PVT under low airflow conditions. In addition, the highest average electrical efficiency and electrical exergy efficiency were calculated as 5.993% and 2.159%, respectively. An MLP-based ANN model was developed using 121 experimental datasets with six input parameters and four output variables. The optimized model based on the Levenberg-Marquardt algorithm achieved MSE, RMSE, MAE, and R2 values of 0.00052, 0.0229, 0.0136, and 0.9934, respectively. The findings indicate that the combined use of graphene nano-coating and PCM-based thermal storage improves PVT performance, while the ANN model provides an accurate tool for predicting energy and exergy-based performance indicators.